[J-34-2025] IN THE SUPREME COURT OF PENNSYLVANIA MIDDLE DISTRICT
TODD, C.J., DONOHUE, DOUGHERTY, WECHT, MUNDY, BROBSON, McCAFFERY, JJ.
COMMONWEALTH OF PENNSYLVANIA, : No. 19 MAP 2023 : Appellant : Appeal from the Order of the York : County Court of Common Pleas, : Criminal Division, at No. CP-67-CR- v. : 1632-2004, dated April 27, 2022. : : ARGUED: May 13, 2025 ROBERT W. ARNETT, : : Appellee :
OPINION
JUSTICE WECHT DECIDED: March 26, 2026 In 1995, Pennsylvania enacted its inaugural version of Megan’s Law, 1 a
comprehensive regulatory scheme designed, among other things, to “[p]rotect the safety
and general welfare of the people of this Commonwealth by providing for registration,
community notification and access to information regarding sexually violent predators and
offenders who are about to be released from custody and will live in or near their
1 See 42 Pa.C.S. §§ 9793-9795 (repealed). The first sexual offender regulatory scheme was enacted by New Jersey in 1994. See N.J.S.A. §§ 2C:7-1 to 2C:7-11 (repealed). That law, and many like it that would follow, was named “Megan’s Law,” because it was passed as a response to the rape and murder of a seven-year-old girl, Megan Kanka. Shortly thereafter, Congress conditioned the allocation of certain federal funding to states upon their passage of similar sexual offender registration and notification statutes. See 42 U.S.C. § 14071 (repealed). Pennsylvania’s enactment responded to this federal legislation. neighborhood.”2 Since then, that law has been amended, revised, replaced, and
renamed. 3 All told, there have been four versions of Megan’s Law and two of SORNA,
most of which have faced a variety of constitutional challenges. 4 These laws have been
challenged upon due process grounds, 5 upon the manner in which the law was passed, 6
and as violations of the ex post facto clauses in the United States and Pennsylvania
Constitutions. 7 Until today, however, this Court has not had occasion to examine the
propriety of the mechanisms utilized to challenge a sexual offender registration statute.
2 42 Pa.C.S. § 9799.51(b)(1). 3 Megan’s Law was renamed as the “Sexual Offender Registration and Notification Act,” commonly referred to by its acronym, “SORNA.” See 42 Pa.C.S. §§ 9799.11- 9799.75. 4 See, e.g., Commonwealth v. Gaffney, 733 A.2d 616 (Pa. 1999) (Megan’s Law I); Commonwealth v. Williams, 733 A.2d 593 (Pa. 1999) (“Williams I”) (Megan’s Law I); Commonwealth v. Williams, 832 A.2d 962 (Pa. 2003) (“Williams II”) (Megan’s Law II); Commonwealth v. Killinger, 888 A.2d 592 (Pa. 2005) (Megan’s Law II); Commonwealth v. Wilson; 910 A.2d 10 (Pa. 2006) (Megan’s Law II); Commonwealth v. Neiman, 84 A.3d 603 (Pa. 2013) (Megan’s Law III); Commonwealth v. Muniz, 164 A.3d 1189 (Pa. 2017) (SORNA I); Commonwealth v. Lacombe, 234 A.3d 602 (Pa. 2020) (SORNA II). 5 See Williams I, supra n.4. 6 See Neiman, 84 A.3d at 616 (holding that Megan’s Law III violated the “single subject” provision of the Pennsylvania Constitution); see PA. CONST. art. III, § 3 (“No bill shall be passed containing more than one subject, which shall be clearly expressed in its title, except a general appropriation bill or a bill codifying or compiling the law or a part thereof.”). 7 See Muniz & Lacombe, supra n.4. The prohibition on ex post facto laws appears twice in the United States Constitution. The first proscription is found in Article 1, Section 9, and serves as a limitation on Congress’ authority to pass laws: “No Bill of Attainder or ex post facto Law shall be passed.” U.S. CONST. art. 1, § 9. The limitation appears for the second time in Article 1, Section 10, and, in this usage, constitutes a restriction on the power of the states: “No State shall . . . pass any Bill of Attainder, ex post facto Law, or Law impairing the Obligation of Contracts, or grant any Title of Nobility.” U.S. CONST. art. 1, § 10. Pennsylvania’s ex post facto provision is found in Article 1, Section 17 of our Constitution, and states that: “No ex post facto law, nor any law impairing the obligation of contracts, or making irrevocable any grant of special privileges or immunities, shall be passed.” PA. CONST. art 1, § 17.
[J-34-2025] - 2 We begin that process today by considering whether a petition for relief under the Post
Conviction Relief Act 8 is a viable mechanism to challenge the constitutionality of such a
law. Because the PCRA only authorizes petitions related to a person’s “conviction or
sentence,”9 and because SORNA II is non-punitive, i.e., not a criminal sentence, 10 we
hold that the PCRA is not available in this context.
In 2003, Robert Arnett engaged in sexual intercourse with a fourteen-year-old girl.
Arnett was arrested and charged with aggravated indecent assault, 11 statutory sexual
assault, 12 three counts of indecent assault, 13 and two counts of corruption of the morals
of a minor. 14 On June 14, 2004, Arnett pleaded guilty to those charges. The trial court
sentenced him to serve an aggregate term of five to ten years’ incarceration.
Under the sexual offender registration statute that governed at the time, Arnett’s
aggravated indecent assault conviction subjected him to lifetime registration as a sexual
offender. 15 After Arnett was released from prison, Megan’s Law was replaced by SORNA,
and then by SORNA II, which also subjects Arnett to lifetime compliance. 16
8 Hereinafter “PCRA.” See 42 Pa.C.S. §§ 9541-46. 9 Id. § 9543(a)(2). 10 See Lacombe, supra n.4; see also Commonwealth v. Torsilieri, 316 A.3d 77 (Pa. 2024) (“Torsilieri II”). 11 18 Pa.C.S. § 3125. 12 Id. § 3122.1. 13 Id. § 3126. 14 Id. § 6301. 15 See 42 Pa.C.S. § 9795.1(b)(2)(i) (expired). Following an assessment, Arnett was determined not to be a sexually violent predator. See id. § 9795.4 (expired). 16 See id. § 9799.55(b)(2)(i)(A). Because Arnett’s offenses occurred before December 20, 2012, he is subject to Subchapter I of SORNA II. See id.; see also id. § (continued…)
[J-34-2025] - 3 On August 13, 2020, over fifteen years after his judgment of sentence became
final, 17 Arnett filed a pro se PCRA petition. The PCRA court appointed counsel, who, on
March 15, 2021, filed an amended PCRA petition 18 on Arnett’s behalf. Among other
things, Arnett argued that SORNA II unconstitutionally infringed upon his fundamental
right to reputation, 19 because it rests upon the “irrebuttable presumption that all sex
offenders, regardless of their individual characteristics or even the particular offense
involved, ‘pose a high risk of engaging in further offenses even after being released from
incarceration or commitments.’” 20 Arnett contended that, statistically, the presumption
9799.52 (subjecting to Subchapter I those who were “convicted of a sexually violent offense committed on or after April 22, 1996, but before December 20, 2012 whose period of registration with the Pennsylvania State Police . . . has not expired” or who were “required to register with the Pennsylvania State Police under a former sexual offender registration law of this Commonwealth on or after April 22, 1996, but before December 20, 2012, whose period of registration has not expired”). 17 Under the PCRA, a judgment of sentence becomes final “at the conclusion of direct review, including discretionary review in the Supreme Court of the United States and the Supreme Court of Pennsylvania, or at the expiration of time for seeking the review.” Id. § 9545(b)(3). Arnett was sentenced in 2004, over fifteen years before he filed his PCRA petition. 18 Arnett styled his amended petition as an “Amended PCRA Petition/Motion for Writ of Habeas Corpus.” Despite the dual title, Arnett asked only that the PCRA court grant his petition and enjoin “the enforcement of sex offender registration requirements on him.” Amended Petition, 3/5/2021, at 5 (unpaginated). As discussed below, we directed supplemental briefing in this case only to review whether the PCRA is a suitable method for challenging a sexual offender’s obligation to comply with SORNA II. We leave the viability of habeas corpus as such a mechanism for a later day. 19 See PA. CONS. Art. 1, §1 (“All men are born equally free and independent, and have certain inherent and indefeasible rights, among which are those of enjoying and defending life and liberty, of acquiring, possessing and protecting property and reputation, and of pursuing their own happiness.”). 20 Amended Petition, 3/5/2021, at 4 (unpaginated) (quoting 42 Pa.C.S. § 9799.51(a)(2)). We rejected this precise legal claim in Torsilieri II. Arnett also argued that he was not required to register as a sexual offender at all, because Megan’s Law III was in effect at the time that he committed his offenses, which this Court held (continued…)
[J-34-2025] - 4 was not universally true and that, instead of a blanket presumption, there existed
reasonable alternative means to ascertain whether an offender poses a risk of recidivism.
The Commonwealth filed an answer. It did not argue specifically that the PCRA
was the incorrect mechanism to challenge the constitutionality of SORNA II. Although
the Commonwealth stressed that the petition was “grossly untimely” under the PCRA, it
acknowledged that, in Lacombe, this Court declined, for the time being, to find that “the
PCRA, or any other procedural mechanism, is the exclusive method for challenging
sexual offender registration statutes[.]”21 As to the merits, the Commonwealth criticized
Arnett for failing to analyze the specific terms of SORNA II or offer cases interpreting
SORNA II in support of his claim that the statutory scheme violated his reputational right.
The Commonwealth emphasized that Arnett did not mention, let alone account for, the
provision in SORNA II that provided a mechanism through which a sexual offender can
be relieved of his reporting obligations. 22
The PCRA court held a hearing on November 15, 2021, at which neither party
presented any lay or expert testimony. In an April 27, 2022 opinion and order, the PCRA
court agreed with Arnett that SORNA II was unconstitutional, and it granted his petition.
unconstitutional. See Neiman, 84 A.3d at 616. The PCRA court ultimately rejected this argument. When Megan’s Law III was stricken as unconstitutional, Megan’s Law II reverted to operation. At all times, Arnett was under the continuous obligation to comply with one version of Megan’s Law or SORNA. That claim no longer is a part of this appeal. 21 Answer, 5/12/2021, at 3 (quoting Lacombe, 234 A.3d at 618). 22 See 42 Pa.C.S. § 9799.59. This section permits a sexual offender to request exemption from SORNA II’s obligation to register with the PSP, verify his or her address, employment, or education status, and to appear on the publicly accessible website. The offender becomes eligible to petition for this exemption once twenty-five years have passed after the offender commenced registering as a sexual offender, “during which time the petitioner has not been convicted in this Commonwealth or any other jurisdiction or foreign country of an offense punishable by imprisonment of more than one year, or the petitioner's release from custody following the petitioner's most recent conviction for an offense, whichever is later.” Id. § 9799.59(a)(1).
[J-34-2025] - 5 The Commonwealth appealed the order to the Superior Court. 23 On February 27,
2023, a panel of that court explained that, by granting Arnett’s PCRA petition, the PCRA
court “declared SORNA’s Subchapter I unconstitutional.”24 Appeals of such orders, the
panel noted, fall within this Court’s exclusive jurisdiction. 25 Thus, the panel transferred
the Commonwealth’s appeal to this Court.
As Arnett’s PCRA petition proceeded through the lower courts, this Court was
considering the legal issue at the heart of Arnett’s challenge to SORNA II. 26 In 2018, the
Chester County Court of Common Pleas declared SORNA II unconstitutional because,
inter alia, the statutory scheme “violated [a sexual offender’s] right to due process by
impairing his right to reputation, as protected by the Pennsylvania Constitution, through
the utilization of an irrebuttable presumption.” 27 On appeal, we were unable to resolve
the issue on the existing record. In large part, the trial court’s ruling was predicated upon
three expert affidavits, to which the Commonwealth stipulated, challenging the legislative
presumption that all sexual offenders pose a high risk of recidivism. 28 Because the trial
23 On May 27, 2022, the PCRA court directed the Commonwealth to file a concise statement of errors complained of on appeal pursuant to Pa.R.A.P. 1925(b). On June 15, 2022, the Commonwealth timely complied. On July 22, 2022, the PCRA court issued a Pa.R.A.P. 1925(a) opinion. 24 Commonwealth v. Arnett, 293 A.3d 642, 2023 WL 2232705, at *2 (Pa. Super. Feb. 27, 2023) (memorandum). 25 Id. at *1-2 (quoting 42 Pa.C.S. § 722(7)). 26 See Commonwealth v. Torsilieri, 232 A.3d 567 (Pa. 2020) (“Torsilieri I”). 27 Id. at 574 (summarizing the trial court opinion). The court also held that SORNA II was unconstitutional because it failed to afford an offender adequate notice and an opportunity to be heard, violated the doctrine of separation of powers, and deprived an offender of the right to a trial by jury as provided by the Sixth Amendment to the United States Constitution. Id. at 574-75. 28 Id. at 594.
[J-34-2025] - 6 court did not have the “benefit of the opposing science,” we remanded the case to the
trial court “to provide both parties an opportunity to develop arguments and present
additional evidence and to allow the trial court to weigh that evidence in determining
whether [the sexual offender] has refuted the relevant legislative findings supporting
[SORNA II].” 29 Notably, in Torsilieri I, only Subchapter H of SORNA II was challenged.
While Torsilieri I percolated in the lower court on remand, Arnett’s appeal reached
this Court. On July 19, 2023, we placed Arnett’s appeal on hold pending a final resolution
in Torsilieri.
On remand, the Torsilieri trial court held three days of evidentiary hearings, at
which three experts testified for the defense that recent empirical data demonstrated that
sexual offenders do not pose a high risk of recidivism. 30 The Commonwealth presented
its own expert, who opined that, because many sexual offenses go unreported, the data
relied upon by the defense experts was unreliable. 31 The trial court credited the defense
experts and rejected the Commonwealth’s expert’s criticism. Thus, the court concluded,
Subchapter H of SORNA II was unconstitutional because it was predicated upon an
irrebuttable presumption that infringed upon an individual’s reputational rights under
Article I, Section 1 of the Pennsylvania Constitution. 32
This Court reversed. To overturn a legislative policy under the irrebuttable
presumption doctrine, a challenger must demonstrate: (1) an interest protected by the
due process clause; (2) utilization of a presumption that is not universally true; and (3) the
29 Id. at 596. 30 Torsilieri II, 316 A.3d at 84. 31 Id. 32 Id. at 85.
[J-34-2025] - 7 existence of a reasonable alternative means to ascertain the presumed fact. 33 Because
the first prong was not disputed, we proceeded directly to the second prong. We noted
that, while this prong suggests that a presumption must be true in every instance, without
exception, such a standard is a practical impossibility. 34 Thus, the question to be asked
is whether there is a verifiable consensus regarding the veracity of the presumption.35
For a sexual offender subject to SORNA II, this is a “heavy burden” to meet. He or she
must “establish that there exists a scientific consensus that sexual offenders pose no
greater risk of committing additional sexual crimes than other groups not subject to similar
registration laws.” 36 A consensus is more than “a mere disagreement among experts” 37
with the status quo. Before a court may take the “extraordinary step” of overturning a
legislative presumption, there must be “clear and indisputable evidence,” “as the General
Assembly made a considered policy choice that sex crimes were uniquely abhorrent to
the victims and society, and relying on the presumption that, as a group, those who
commit such crimes are more likely to commit another crime of a sexual nature.”38
The trial court found that the irrebuttable presumption at the heart of SORNA II
was not universally true. That court determined that there existed a consensus among
the relevant scientists that sexual offenders do not pose a high risk of reoffending. The
court credited the defense experts’ testimony that 80-95% of sexual offenders will not
33 In re J.B., 107 A.3d 1, 15-16 (Pa. 2014). 34 Torsilieri II, 316 A.3d at 98. 35 Id. 36 Id. at 98-99. 37 Id. at 99. 38 Id.
[J-34-2025] - 8 reoffend. 39 Upon review, we held that, by considering individual recidivism rates, the trial
court focused upon the wrong statistical measure. The correct “meaningful statistical
measure,” we explained, is “whether the percentage of those who have committed a
sexual offense and go on to commit a second sexual offense—the group SORNA
targets—is higher than the percentage of those who first commit a non-sexual offense
followed by a second, sexual offense.”40 Applying the correct rubric, we stressed that the
defense experts had “concede[d] that adult sexual offenders reoffend at a rate . . . at least
three times higher than other individuals convicted of non-sexual offenses.” 41 “[R]ather
than refuting it, the evidence supports the legislative presumption; the evidence
validat[ed] the statutory underpinnings” of SORNA II. 42 We thus reversed the trial court,
because the defense had failed to meet its “heavy burden.” 43
We next considered whether SORNA II was punitive in effect. If it were, other
constitutional concerns would arise, such as the doctrine of separation of powers, the
prohibition on cruel and unusual punishments, and the right to trial by jury. When a civil
regulatory scheme is challenged as penal in nature, we apply the analytical framework of
Kennedy v. Mendoza-Martinez. 44 The first part of that rubric requires courts to inquire
whether, by enacting the regulatory scheme, the legislature intended to punish those
39 Id. at 85. 40 Id. at 99. 41 Id. 42 Id. (emphasis in original). 43 Id. at 99-100. 44 372 U.S. 144 (1963).
[J-34-2025] - 9 subjected to its terms or conditions. 45 With regard to Subchapter H of SORNA II, we held,
“the clearly expressed legislative purpose, findings, and declaration of policy all establish
that, rather than intending to punish, the General Assembly desired to enact a civil,
regulatory scheme.” 46
Discerning the General Assembly’s non-punitive intent does not end the inquiry.
The second part of the Mendoza-Martinez framework necessitates an examination of
whether the regulatory scheme is nonetheless punitive in effect, notwithstanding the
legislative intent. For this examination, a court must consider the following seven factors:
[w]hether the sanction involves an affirmative disability or restraint, whether it has historically been regarded as a punishment, whether it comes into play only on a finding of scienter, whether its operation will promote the traditional aims of punishment—retribution and deterrence, whether the behavior to which it applies is already a crime, whether an alternative purpose to which it may rationally be connected is assignable for it, and whether it appears excessive in relation to the alternative purpose assigned. 47
We reviewed each of the applicable factors 48 and determined that two (factors two
and four) weighed in favor of finding Subchapter H to be punitive, while three (factors one,
six, and seven) favored non-punitivity. We balanced those factors against each other,
and we held as follows:
45 Torsilieri II, 316 A.3d at 102 (citing Torsilieri I, 232 A.3d at 588; Lacombe, 234 A.3d at 618). 46 Id. In fact, the General Assembly expressly directed to courts that Subchapter H “shall not be construed as punitive.” 42 Pa.C.S. § 9799.1(b)(2). 47 Mendoza-Martinez, 372 U.S. at 168-69. 48 As we did in Muniz, 164 A.3d at 1214, Lacombe, 234 A.3d at 603-04, 606, and Torsilieri I, 232 A.3d at 589, we found that factors three—whether the law comes into play only on a finding of scienter—and five—whether the behavior to which the sanction applies is already a crime—were not particularly significant in an evaluation of sexual offender registration statutes and assigned them no weight in the overall balancing of factors. Torsilieri II, 316 A.3d at 102.
[J-34-2025] - 10 In our view, weighing the Mendoza-Martinez factors does not compel the conclusion that Subchapter H is punitive. Here, the General Assembly created a tier-based classification system organized by seriousness of the offense, which, in turn, is tied to the degree of harm caused by the crime. This is a policy-based decision vested in the legislature. Like Subchapter I, we find that Subchapter H significantly changed the original version of SORNA with the apparent goal of ensuring that the legislation was not punitive in nature. Indeed, Subchapter H has a significantly less burdensome impact on the life of the offender than its predecessor. Moreover, we find compelling the Commonwealth's argument that not only does Subchapter H offer a valid non-punitive purpose of informing and protecting the public, but that [the record contained no] compelling evidence establishing that its registration and notification requirements were excessive, i.e., not rationally or reasonably related to this legislative purpose. 49
In sum, while Arnett’s present appeal was on hold, we held in Torsilieri II that
Subchapter H of SORNA II did not violate a sexual offender’s due process rights under
the irrebuttable presumption doctrine and that it was not punitive in its effect on the
offender. Once the opinion issued, we turned our attention back to Arnett’s appeal.
Rather than summarily dispose of that appeal in light of Torsilieri II, we instead decided
to examine, among other things, the propriety of Arnett’s utilization of the PCRA as the
mechanism for challenging the constitutionality of the governing sexual offender
registration law. Thus, on August 20, 2024, we agreed to hear oral arguments in the
case. We directed the parties to provide supplemental briefing that specifically addressed
the following issues:
1. Whether, in light of this Court’s decisions in Commonwealth v. Torsilieri, 316 A.3d 77 (Pa. 2024), and Commonwealth v. Lacombe, 234 A.3d 602 (Pa. 2020), holding that Pennsylvania’s Sex Offender Registration and Notification Act’s (SORNA) registration requirements are non-punitive, [Arnett’s] constitutional challenge to Subchapter I of SORNA is properly brought under the Post Conviction Relief Act (PCRA), 42 Pa.C.S. §§ 9541-[]46?
49 Id. at 109.
[J-34-2025] - 11 2. If so, did the PCRA court have jurisdiction to consider the merits of [Arnett’s] facially untimely PCRA petition?
3. What effect, if any, does this Court’s decision in [Torsilieri II] have on a PCRA court’s conclusion that Subchapter I of SORNA imposed an unconstitutional irrebuttable presumption upon [Arnett]? 50
In the three decades since our General Assembly enacted Megan’s Law I, this
Court never has “required that sexual offender registration statutes be challenged” using
any particular “procedural mechanism.”51 Our hesitation has been due, in large part, to
the “frequent changes to sexual offender registration statutes, along with more onerous
requirements and retroactive application, [which] complicate [a] registrant[’]s ability to
challenge new requirements imposed years after their sentence[] become[s] final.” 52 This
laissez-faire approach has resulted in challenges arising from various procedural starting
points. Williams II began with a challenge to Megan’s Law II lodged in a pre-sentence
“Motion for Extraordinary Relief.” 53 The registrant in Muniz claimed that SORNA I was
unconstitutional in a post-sentence motion (and on direct appeal), 54 as did the challenger
in Torsilieri. 55 The arguments in Lacombe directed at Subchapter I of SORNA II arose in
a “petition to terminate [] sexual offender registration requirements” and in a pro se PCRA
50 Order, 8/20/2024, at 1. 51 Lacombe, 234 A.3d at 617. 52 Id. 53 Williams II, 832 A.2d at 965. There were two challengers to Megan’s Law II in Williams II: Gomer Williams and Bruce Peters. Like Williams, Peters filed a pre- sentencing “Motion for Relief.” Id. In both instances, the trial court found Megan’s Law II to be unconstitutional and granted the respective petitions. Id. at 970. 54 Muniz, 164 A.3d at 1193. 55 Torsilieri II, 316 A.3d at 81-82.
[J-34-2025] - 12 petition. 56 We also have seen constitutional arguments emanate from a “Petition to
Enforce Plea Agreement or for a Writ of Habeas Corpus” 57 and from a petition for a writ
of mandamus. 58
This permissive approach has been understandable, given the fact that, between
successful court challenges and legislative amendment, sexual offender laws frequently
undergo vast and substantial revision. It is not uncommon for offenders subject to one
version of a law to find themselves obliged years later to conform to a new statutory
scheme, one that imposes more stringent requirements on the offenders and that requires
compliance for a much longer period of time. These revisions sometimes occur decades
after an offender has been convicted and sentenced. Consequently, there must be some
flexibility with regard to the manner in which challenges to these laws are brought.
While some latitude is warranted in these circumstances, it does not follow that the
universe of motions or petitions available to a challenger knows no bounds. Each
mechanism, be it a petition for a writ of mandamus, a petition for review in the
Commonwealth Court, a motion for extraordinary relief, etc., can serve only the purposes
for which it exists, and can be invoked only according to that mechanism’s terms and
conditions. That some flexibility is necessary in this dynamic, oft-changing area of law
does not license courts or litigants to disregard jurisdictional or constitutional limitations
or, as is the case here, the intent and plain terms of a statute. We must enforce the law
as written by our legislators. 59 When a statute’s terms preclude that statute from serving
56 Lacombe, 234 A.3d at 606. 57 Commonwealth v. Martinez, 147 A.3d 517, 523 (Pa. 2016). 58 A.S. v. Pa. State Police, 143 A.3d 896, 903 n.7 (Pa. 2016). 59 “The domain of the judiciary is in the field of the administration of justice under the law; it interprets, construes[,] and applies the law.” Commonwealth v. Sutley, 378 A.2d 780, 783 (Pa. 1977) (quoting Commonwealth v. Widovich, 145 A. 295, 299 (Pa. 1929)) (continued…)
[J-34-2025] - 13 as a vehicle to challenge the constitutionality of a sexual offender regulatory scheme,
those terms must be enforced, especially when those terms relate to cognizability or
jurisdiction. The PCRA’s terms do just that.
Before proceeding to the particulars of the PCRA, we pause to note aspects of
SORNA’s history that substantiate the conclusion that the PCRA cannot serve as a
mechanism for challenging a sexual offender registration law. SORNA I replaced
Megan’s Law as Pennsylvania’s sexual offender regulatory scheme in 2012. In Muniz,
this Court found that SORNA I was punitive in effect and amounted to criminal
punishment. 60 We held that SORNA I’s registration requirements, when applied
retroactively, constituted an unconstitutional ex post facto law. 61 The General Assembly
responded by enacting a new regulatory system. In SORNA II, the General Assembly
bifurcated SORNA within the Sentencing Code, creating two subchapters: Subchapter H
and Subchapter I. The former applies to those sexual offenders whose triggering
offenses were committed on or after December 20, 2012. The latter applies to those
whose offenses occurred before that date.
In Lacombe, this Court considered an ex post facto challenge to Subchapter I of
SORNA II. After applying and balancing the Mendoza-Martinez factors, this Court held
that the General Assembly had remedied the constitutional defects that plagued SORNA
I 62 and that Subchapter I of SORNA II was not punitive in effect and, thus, did not
(footnote omitted); see also Commonwealth v. Small, 238 A.3d 1267, 1284 (Pa. 2020) (“Courts must apply statutes as they are written.”). 60 Muniz, 164 A.3d at 1218. 61 Id. 62 Lacombe, 234 A.3d at 626 (“As the above Mendoza-Martinez analysis clearly reflects, Subchapter I effected significant changes from the original version of SORNA, retroactive application of which we found unconstitutional in Muniz.”).
[J-34-2025] - 14 constitute a criminal punishment. 63 Later, in Torsilieri II, we held the same with regard to
Subchapter H of SORNA II in Torsilieri II. 64 The legal character of both subchapters of
SORNA II is now settled. Neither constitutes criminal punishment. Because the PCRA
applies only to those matters related to a conviction or sentence, the legal designation is
largely dispositive of the question at bar.
We begin with the PCRA’s stated purposes. The statute states, in no uncertain
terms, that the PCRA “provides for an action by which persons convicted of crimes they
did not commit and persons serving illegal sentences may obtain collateral relief.” 65 A
challenge to the constitutionality of SORNA II’s registration and reporting requirements
does not relate to either of these purposes. A successful constitutional challenge to this
regulatory scheme would have no impact upon the criminal conviction(s) that subjected
the offender to SORNA II in the first place, nor would it alter the offender’s sentence or
require resentencing. Because such a challenge does not fall within the ambit of the
PCRA’s dual purposes, it follows that the General Assembly did not intend for the PCRA
to “provide for an action” challenging a non-punitive regulatory scheme. In fact, the PCRA
states that it “is not intended . . . to provide relief from collateral consequences of a
criminal conviction.” 66 The PCRA exists to provide a means to challenge collaterally a
criminal defendant’s conviction or sentence, not the non-punitive consequences
attendant to that conviction or sentence.
63 Id. at 626-27 (“We hold [that] Subchapter I does not constitute criminal punishment[.]”). 64 Torsilieri II, 316 A.3d at 109-110. 65 42 Pa.C.S. § 9542. 66 Id.
[J-34-2025] - 15 The availability of the PCRA as a mechanism to challenge SORNA II is precluded
not only by the PCRA’s intent, but by the statute’s terms as well. Subsection 9543 of the
PCRA—entitled “Eligibility for relief”—establishes the criteria that a PCRA petitioner must
meet in order to be eligible for relief. 67 A petitioner must “plead and prove by a
preponderance of the evidence” that he is “currently serving a sentence of imprisonment,
probation or parole for the crime,” or that he is “awaiting execution of a sentence of death
for the crime.” 68 This criterion demonstrates that the General Assembly did not intend the
PCRA to be used to test the constitutionality of non-punitive civil statutory schemes. An
offender’s obligation to register under SORNA II does not begin until he or she is released
from prison, and that obligation can last for at least a decade and, for many, a lifetime.
The vast majority of registration obligations will persist long after an offender completes
his or her sentence, at which point the offender is no longer eligible to seek PCRA relief.
This is not the population upon which the General Assembly intended to confer an avenue
for relief when it created the PCRA.
For those that are serving a sentence (or awaiting execution), the PCRA lists the
types of claims that may be raised in a petition for relief. A petitioner may assert a violation
of his or her constitutional rights, may claim that counsel was ineffective, may argue the
existence of newly-discovered evidence, may complain that his or sentence is illegal,
etc. 69 These claims are cognizable only if the asserted error or violation contributed in
some way to the “sentence or conviction.”70 By its unambiguous terms, the PCRA is
confined to collateral attacks on a conviction or sentence. Lacombe and Torsilieri II
67 Id. § 9543. 68 Id. § 9543(a)(1)(i)-(ii). 69 Id. § 9543(a)(2)(i)-(viii). 70 Id. § 9543(a)(2).
[J-34-2025] - 16 establish that SORNA II is not a criminal punishment. Hence, a constitutional challenge
to SORNA II”s registration terms (and other obligations) is not related to the offender’s
“sentence or conviction.” Such a claim is not cognizable under the PCRA. 71
For these reasons, the PCRA is precluded by its own terms from being invoked as
a means to challenge the constitutionality of a non-punitive sexual offender regulatory
scheme. 72 Because the PCRA was not available to Arnett, we need not consider the
second question upon which we directed supplemental briefing in this case—whether
Arnett established the PCRA court’s jurisdiction over his facially untimely petition. 73 That
question is now moot.
We do not dismiss Arnett’s petition with prejudice. As noted earlier, Arnett styled
his petition as both a PCRA petition and a petition for a writ of habeas corpus. The PCRA
court appointed counsel to assist Arnett. Appointed counsel filed an amended petition,
which counsel also styled as both a PCRA petition and a petition for a writ of habeas
71 Challenging SORNA II under the PCRA is complicated further by the PCRA’s jurisdictional time limit. To be timely, and thus invoke the PCRA’s court’s jurisdiction, a PCRA petitioner must file a petition within one year of his or her judgment of sentence becoming final. 42 Pa.C.S. § 9545(b)(1). The vast majority of sexual offenders, whose SORNA II obligations commence upon release from prison, will be jurisdictionally barred from seeking relief under the PCRA because their judgments of sentence finalized long before the first time they appear before the PSP to register as a sexual offender. The only option to circumvent the time bar is to plead and prove one of these statutory exceptions: (1) governmental interference; (2) the discovery of a newly-discovered fact; or (3) the novel recognition of a constitutional right that applies retroactively. Id. § 9545(b)(1)(i)-(iii). Facially, neither the governmental interference exception nor the retroactive constitutional right exception facially is applicable. And, even if a change in the law could be considered a newly-discovered fact, a sexual offender still would not be able to demonstrate that his or her claim is cognizable. 72 Our holding depends in substantial part upon our determinations that SORNA II, in its current form, is non-punitive in effect. If a future iteration of SORNA is found to be punitive, today’s ruling will require reconsideration. 73 Order, 8/20/2024, at 1.
[J-34-2025] - 17 corpus. While we reject Arnett’s use of the PCRA, his pursuit of a writ of habeas corpus
remains. Consistent with this Court’s decision in Lacombe not to limit the mechanisms
available to a challenger without full development of the issue, 74 and although both the
Commonwealth and Arnett agree that a petition for a writ of habeas corpus (and a petition
for a writ of coram nobis) is a feasible method of challenging SORNA II, 75 we leave
consideration of the viability of other non-PCRA mechanisms for another day. We
directed the parties to brief only the availability of the PCRA. Consideration of any other
mechanisms would exceed the scope of that directive.
Justice Brobson would decide the issue now. He would hold that a petition for a
writ of habeas corpus is not available to challenge the legality or constitutionality of
SORNA. He would rule that the lower court lacked subject matter jurisdiction in this
case. 76 However, doing so would vitiate the restraint that we deemed necessary in this
context in Lacombe. Although a court may consider matters pertaining to subject matter
jurisdiction sua sponte, 77 we have declined to do so where no party has raised a fully
developed argument against the mechanism used by a sexual offender to challenge the
constitutionality of SORNA. 78 In Lacombe, where no such challenge was developed, we
found that subject matter jurisdiction existed and addressed the merits of only the issues
74 See Lacombe, 234 A.3d at 618. 75 See Supplemental Br. for the Com. at 6-7; Supplemental Br. for Arnett at 10. 76 See Conc. and Diss. Op. at 3-5. 77 Commonwealth v. Little, 314 A.2d 270, 272 (Pa. 1974) (“Whether a court has subject matter jurisdiction over an action is a fundamental issue of law which may be raised at any time in the course of the proceedings, including by a reviewing court sua sponte.”) (emphasis added). 78 See Lacombe, 234 A.3d at 617-18.
[J-34-2025] - 18 raised in the appeal. 79 We did not venture into any unraised matters. Our reluctance to
dismiss an action in this context on jurisdictional grounds sua sponte stems from the fact
that SORNA frequently undergoes substantial legislative revisions, imposing upon
offenders new and different requirements, often years after a sentence becomes final. 80
Here, the principal parties do not dispute meaningfully the propriety of the use of habeas
corpus petitions to challenge SORNA. Only the OAG takes the opposite position, and
only briefly, which leaves us with, at best, a one-sided argument. Thus, we exercise the
same restraint that we found to be imperative in Lacombe and we decline to review the
matter further. 81
Justice Brobson would forge ahead. Although Justice Brobson agrees that the
PCRA cannot be used to contest SORNA’s requirements, 82 he then asks “what vehicle
may Arnett use to challenge the constitutionality of those requirements and what court
79 Id. at 618 (“[W]e decline to find the PCRA, or any other procedural mechanism, is the exclusive method for challenging sexual offender registration statutes and we thus conclude the trial court had jurisdiction to consider Lacombe's ‘Petition to Terminate His Sexual Offender Registration Requirements.’”). 80 See id. at 617 (“Our approach in this regard takes into account the fact that frequent changes to sexual offender registration statutes, along with more onerous requirements and retroactive application, complicate registrants’ ability to challenge new requirements imposed years after their sentences become final.”). 81 We do not cavalierly “waive off jurisdictional concerns,” as Justice Brobson alleges. Conc. and Diss. Op. at 2. We are instead adhering to the protocol agreed upon by a majority of Justices in Lacombe. Justice Brobson questions our “commitment to restraint” because we address the merits of the lower court’s decision without addressing whether a habeas corpus petition suffices to invoke that court’s jurisdiction. See Conc. and Diss. Op. at 7 n.4. We hold the contrary perspective: in declining to address questions not prompted by the questions upon which we directed supplemental briefing, and in confining ourselves to the issues fully developed by the parties, our “commitment to restraint” is clear. 82 Conc. and Diss. Op. at 2.
[J-34-2025] - 19 has jurisdiction over the matter?”83 Ignoring the approach that we adopted in Lacombe,
Justice Brobson suggests that a writ of mandamus filed in the Commonwealth Court’s
original jurisdiction is the proper mechanism when a party files an action against the PSP
seeking removal from SORNA’s registry. This conclusion may have some facial appeal
and statutory support. 84 However, neither SORNA’s text or history, nor this Court’s
precedents, require all claims for SORNA relief be funneled exclusively into one particular
court. In fact, SORNA does not divest courts of common pleas of jurisdiction entirely.
SORNA directs sexual offenders seeking to be removed from the registry via SORNA’s
removal mechanism to file such petitions with the sentencing court, not with the
Commonwealth Court. 85 Having envisioned a role for courts of common pleas, it is not
at all clear that the General Assembly intended that those courts—who are tasked with
determining whether an offender is a sexually violent predator, ordering an offender to
comply with SORNA at sentencing, presiding over trials for offenders charged with failing
to comply with SORNA’s obligations, and deciding whether an offender should be
released from those obligations—be stripped of jurisdiction to hear any other SORNA-
related claims. It is a question we should only decide in the normal course, with briefing
and argument.
The need to proceed with caution becomes evident upon a cursory review of the
issue. The question of whether habeas corpus is available to challenge SORNA is not
one that can be summarily decided, as Justice Brobson would. Historically, the
availability of habeas corpus has been limited to those in custody, 86 or to those facing the
83 Id. 84 See id. at 5 n.1 (citing 42 Pa.C.S. § 761(a)(1)(b)). 85 See 42 Pa.C.S. § 9799.59 (a)(2). 86 Preiser v. Rodriguez, 411 U.S. 475, 484 (1973).
[J-34-2025] - 20 possibility of being placed in custody. 87 Although we have held that SORNA is not
punitive, and, therefore, not a criminal sentence, we twice have held that SORNA’s
obligations create a regulatory scheme that is akin to a traditional sentence of probation. 88
A violation of these probation-like conditions results in a new criminal charge and a
potential new sentence. 89 Thus, we have at least acknowledged that SORNA operates
in some ways like a traditional form of criminal punishment and, once subject to its terms
and conditions, future incarceration is a possibility. 90 Whether those factors suffice to
allow SORNA to be challenged via a petition for a writ of habeas corpus is far from clear,
and is not a decision that should be made in a cursory fashion. Such arguments (and
any others) should be raised and developed by trained advocates, not by this Court. For
this reason, we decline to opine on any issues beyond those necessary to dispose of the
present appeal.
Thus, to the extent that Arnett’s petition constitutes a habeas corpus petition, we
do not dismiss it in its entirety. Instead, we proceed to the question of “what effect, if any,
does this Court’s decision in Torsilieri II have on a PCRA court’s conclusion that
Subchapter I of SORNA II imposed an unconstitutional irrebuttable presumption upon
87 U.S. ex rel. Dessus v. Com. of Pa., 452 F.2d 557, 559–60 (3d Cir. 1971) (noting that “the sine qua non of federal habeas corpus jurisdiction is that petitioner be in custody. . . . Thus, custody is the passport to federal habeas corpus jurisdiction. Without custody, there is no detention. Without detention, or the possibility thereof, there is no federal habeas jurisdiction.” (cleaned up) (emphasis added)). 88 See Lacombe, 234 A.3d at 623; Torsilieri II, 316 A.3d at 105. 89 See 18 Pa.C.S. § 4915.1 (grading an offender’s failure to comply with Subchapter H’s registration, counseling, and reporting requirements a felony); § 4915.2 (grading an offender’s failure to comply with Subchapter I’s registration, counseling, and reporting requirements a felony). 90 See U.S. ex rel. Dessus, 452 F.2d at 460 (indicating that even the possibility of incarceration is sufficient to warrant habeas corpus review).
[J-34-2025] - 21 Arnett?” 91 The answer is straightforward. Torsilieri II requires that we vacate the PCRA
court’s order.
In Torsilieri II, we considered whether “the General Assembly’s determination . . .
that individuals who commit sexual offenses pose a high risk of committing additional
sexual offenses constitutes an unconstitutional irrebuttable presumption violative of due
process, because it impairs the right to reputation under the Pennsylvania Constitution.”92
Arnett raised a nearly identical question in his “Amended PCRA Petition/Motion for Writ
of Habeas Corpus.” 93 However, Arnett argues that Torsilieri II is not controlling here,
because that case involved a challenge to Subchapter H of SORNA II whereas his
challenge is directed at Subchapter I. He contends, inter alia, that, because Subchapter
I applies retroactively, while Subchapter H applies prospectively, the two subchapters
must be treated differently. The evidence that this Court found insufficient to invalidate
the irrebuttable presumption in Torsilieri II has no applicability here, Arnett maintains,
because that evidence was directed at offenders subject to Subchapter H, not those
subject to Subchapter I, many of whom necessarily are older and farther removed from
the time of their crimes. According to Arnett, due to their age, Subchapter I offenders are
less likely to recidivate, so the standards applicable to Subchapter H offenders should
have no bearing here. 94 We disagree.
91 Order, 8/20/2024, at 1 (cleaned up). 92 Torsilieri II, 316 A.3d at 79. 93 See “Amended PCRA Petition/Motion for Writ of Habeas Corpus,” 3/5/2021, at 4- 5 (unpaginated). 94 Arnett also argues that Torsilieri II should not govern his constitutional challenge because he raised an as-applied challenge, not a facial challenge. However, it is not at all clear that he raised an as-applied challenge in the court below. In his amended petition, Arnett supported his constitutional argument with broad, group-based empirical data. None of the arguments focus upon his individual circumstances, age, etc. Arnett (continued…)
[J-34-2025] - 22 Subchapter H and Subchapter I are in many ways the same. Subchapter I applies
only to those whose offenses were committed before December 20, 2012. Subchapter
H governs those whose offenses occurred thereafter. Both subchapters require a sexual
offender to comply with SORNA II’s numerous obligations for lengthy periods of time. And
both require the offender to report in person to the Pennsylvania State Police, and to
update that agency when the offender changes addresses, employment, and educational
status. Both require the offender’s photograph and personal information to be displayed
on a publicly accessible website. Failure to comply with the terms and conditions of either
subchapter constitutes a separate criminal offense. 95
In some ways, the subchapters are different. Because Subchapter I applies
retroactively, and thus is subject to ex post facto concerns, some of its requirements are
less stringent than those imposed by Subchapter H. For instance, a Tier III sexual
offender subject to Subchapter I must report in person to the PSP once per year,96
whereas under Subchapter H, a Tier III offender must report quarterly. 97 And, Arnett’s
premise is correct: Subchapter I offenders, generally, are older and further removed
temporally from their crimes than Subchapter H offenders.
makes no argument that the presumption that sexual offenders are more likely to recidivate is inapplicable to him. Nor did he adduce any evidence at the PCRA hearing specifically related to his circumstances and how they differ from the underlying irrebuttable presumption. Regardless, because both an as-applied challenge and a facial challenge stem from the same presumption, which is evaluated using the same legal test, the outcome in this instance would be the same. 95 For a detailed description of the terms and conditions of Subchapter I, see Lacombe, 234 A.3d at 615-17; for Subchapter H, see Torsilieri II, 316 A.3d at 123-28 (Wecht, J., concurring and dissenting). 96 See 42 Pa.C.S. § 9799.60(b). 97 Id. § 9799.15(e)(3).
[J-34-2025] - 23 However, the differences between the subchapters are irrelevant for present
purposes. None of these differences distinguish Subchapter I from Subchapter H for
purposes of applying our ruling in Torsilieri II. This is because both subchapters are
predicated upon the same presumption. In Subsection 9799.11(a) of Subchapter H, the
General Assembly stated as one of its “findings” the legislative judgment that “[s]exual
offenders pose a high risk of committing additional sexual offenses and protection of the
public from this type of offender is a paramount governmental interest.” 98 In Subsection
9799.51(a) of Subchapter I, the General Assembly declared that sexual offenders “pose
a high risk of engaging in further offenses even after being released from incarceration or
commitments, and protection of the public from this type of offender is a paramount
government interest.”99 The presumptions for all relevant purposes are the same. We
must treat them the same.
In Torsilieri II, we held that the presumption did not violate sexual offenders’ due
process or reputational rights. Because the expert testimony in that case confirmed that
“adult sexual offenders reoffend at a rate of at least three times higher than other
individuals convicted of non-sexual offenses,” we opined that the relevant scientific data
supported the legislative presumption. 100 Arnett has offered no evidence to counter this
conclusion. Arnett presented no evidence at his PCRA hearing, let alone expert
testimony, to support his argument that he is not one of those sexual offenders likely to
reoffend or that those subject to Subchapter I differ substantively from those “adult sexual
offenders” that we considered in Torsilieri II. Because Subchapter I and Subchapter H
98 Id. § 9799.11(a)(4). 99 Id. § 9799.51(a)(2). 100 Torsilieri II, 316 A.3d at 99.
[J-34-2025] - 24 arise from the same presumption, and because we upheld that presumption in Torsilieri
II, we discern no basis on this record to deviate here.
Torsilieri II controls the disposition of this case. The lower court erred in ruling
Subchapter I of SORNA II unconstitutional. We vacate the lower court’s order, and we
dismiss Arnett’s petition.
Chief Justice Todd and Justices Donohue, Dougherty, Mundy and McCaffery join
the opinion.
Justice Brobson files a concurring and dissenting opinion.
[J-34-2025] - 25