UNITED STATES of America, Appellee, v. Christian J. MORALES, Defendant, Appellant, 801 F.3d 1

HOWARD, Chief Judge. This appeal presents the question of whether a state Rhode Island conviction for first degree child molestation, R.I. Gen. Laws § 11-37-8.1, is “comparable to or more severe than” one of the offenses listed in Tier III of the federal Sex Offender Registration and Notification Act (“SOR-NA”), 42 U.S.C. § 16911(4). Appellant-Defendant Christian Morales, who had previously been convicted under that Rhode Island law, was sentenced in federal court to 65 months in prison and a lifetime of supervised release for failing to register as a sex offender under SORNA, 18 U.S.C. § 2250(a).

At sentencing, the district court utilized the prior Rhode Island conviction to deem Morales a Tier III offender, resulting in a base-offense level two levels higher than if he had been deemed a Tier II offender. Finding the Tier III designation to be plain error, we vacate Morales’s prison term and remand for re-sentencing. I. In December 2006, Morales entered a plea of nolo contendere to two counts of first degree child molestation in the state of Rhode Island. R.I. Gen. Laws § 11-37-8.1. At the time that he committed the sexual assault, he was 18 and the victim was 13.

The state of Rhode Island sentenced Morales to a 30-year incarcerative term, with all but seven years suspended. As a result of that sentence, Morales was required to register as a sex offender under SORNA. That law classifies offenders into three tiers based on the severity of the sex offense. Those categories, in turn, detail the frequency and longevity of an individual’s registration requirements. For instance, a Tier III offender must register for the remainder of his or her life, while a Tier II offender must register for 25 years. 42 U.S.C. § 16915.

In 2010, a federal grand jury indicted Morales in the District of Rhode Island for failing to register, and Morales subsequently pled guilty. Prior to sentencing, the probation officer prepared a pre-sen-tence report classifying Morales as a Tier III sex offender. U.S.S.G. § 2A3.5(a) (setting the base-offense level at 16 for a Tier III offender, as defined in SORNA, rather than 14 for a Tier II offender). The district court accepted this designation, which increased Morales’s Guidelines sentence range from 46-57 months in prison to 57-71 months.

At sentencing, the court imposed a mid-guidelines incarcerative sentence of 65 months. With respect to supervised release, however, the district court emphasized Morales’s behavior since the time of the predicate conviction (including an alleged sexual assault on a minor during his unregistered period) and concluded that a lifetime of supervised release was necessary. At a subsequent hearing, the court reaffirmed its view on supervised release and made clear that public safety demanded an upward variance to the statutory maximum. Morales timely appealed, asserting a litany of challenges.

Finding the Tier III contention to be the only arguably meritorious claim, we requested further briefing and oral argument solely on that issue. II. We begin by setting forth the statutes at issue before delving into the merits. SOR-NA classifies sex offenders into three tiers with each category corresponding to specific, enumerated crimes or to offenses incorporated from other federal sexual abuse laws. The most egregious offenders are grouped into Tier III. 42 U.S.C. § 16911(4)(A).

Meanwhile, Tier II of the statute captures, inter alia, sexual offenses against victims aged 13 through 16 if the perpetrator is four or more years older than the victim. § 16911(3)(A)(iv). The final category, Tier I, serves as a catch-all provision for convicted sex offenders not otherwise grouped into Tier II or Tier III. § 16911(2). Most relevant for our purposes is Tier III. This tier covers individuals who have committed crimes “comparable to or more severe than” a number of enumerated offenses. § 16911(4)(A). Those offenses essentially break down into two categories.

First, Tier III includes sexual offenses against a child aged 12 or under. Part of

section (

i) in Tier III adopts the definition of “aggravated sexual abuse” from 18 U.S.C. § 2241, which penalizes crossing state lines “with intent to engage in a sexual act with a person who has not attained the age of 12 years,” or actually engaging in such conduct. § 16911(4)(A)(i) (incorporating § 2241). Also included in this category, from

section (ii) of Tier III, is “abusive sexual contact ... against a minor who has not attained the age of 13 years.” § 16911(4)(A)(ii) (incorporating 18 U.S.C. § 2243(a)). “Abusive sexual contact” is defined as any sexual offense against a 12 year old if the perpetrator is 16 years or older. Id. Second, Tier III encompasses sex offenses that are committed with force, result in additional harm, or are perpetrated against particularly vulnerable victims. These offenses, also found in

section (i), incorporate other aspects of “abusive sexual contact” from 18 U.S.C. § 2241, along with “sexual abuse” as defined in 18 U.S.C. § 2242. The final statute of interest is the source of Morales’s predicate conviction. That Rhode Island law merely states, “A person is guilty of first degree child molestation sexual assault if he or she engages in sexual penetration with a person fourteen (14) years of age or under.” R.I. Gen. Laws. § 11-37-8.1. III. Given the lack of an objection below, the parties agree that our review is for plain error only.

This requires Morales to show that “(1) an error occurred which was (2) clear or obvious and which not only (3) affected his substantial rights, but also (4) seriously impaired the fairness, integrity, or public reputation of the judicial proceedings.” United States v. Tavares, 705 F.3d 4,16 (1st Cir.2013) (citation omitted). We begin by asking whether any error occurred. i. Morales contends that it was error for the district court to enhance his offense level based on a Tier III designation, since his Rhode Island conviction was not “comparable to or more severe than” any offense listed in that

section of SORNA. Our analysis of this argument proceeds in two steps. First, we must ask what analytical approach applies to this comparative inquiry. Second, under that framework, we must then determine whether the specific Rhode Island law that Morales was convicted under is, in fact, “comparable to or more severe than” any offense in Tier III. a. The threshold question in this case is whether our comparison of the statutes is limited to the elements of each crime or whether we can account for Morales’s specific conduct when determining whether he is a Tier III offender.

At its core, this requires us to give some meaning to the term “offense” as it is utilized in Tier III of the SORNA statute. The Supreme Court has provided significant guidance on how to answer this question. In Descamps v. United States, — U.S. -, 133 S.Ct. 2276, 186 L.Ed.2d 438 (2013), the Court described the analytical framework for comparing a state predicate offense with the generic crimes listed in the Armed Career Criminal Act (“ACCA”).

Key to its decision was a distinction between indivisible statutes (those not containing alternative elements) and divisible statutes (those providing alternative elements). For the latter set of statutes, limited factual consideration is appropriate to determine under which portion of the statute the offense lies. However, for indivisible predicate statutes, like the Rhode Island law at issue here, the comparison must be limited solely to the elements of the crimes. The Court emphasized three main justifications for this. Descamps, 133 S.Ct. at 2287-89, citing Taylor v.

United States, 495 U.S. 575, 110 S.Ct. 2143, 109 L.Ed.2d 607 (1990). First, the text of the ACCA, using the term “convictions” rather than a phrase such as “has committed,” implied that Congress was focused on a defendant’s convictions irrespective of the underlying facts. Descamps, 133 S.Ct. at 2287-88; contra Nijhawan v. Holder, 557 U.S. 29, 129 S.Ct. 2294, 174 L.Ed.2d 22 (2009) (where the statute’s requiremeiit of a loss exceeding $10,000 called for an inquiry into the specific circumstances leading to the offense).

Second, the Court'was concerned that fact-finding on the predicate offense could run afoul of the Sixth Amendment right to a jury trial. Descamps, 133 S.Ct. at 2288-89. Finally, such a categorical approach eschewed the possibility of any mini- or collateral-trial at sentencing to probe the predicate offense. Id. at 2289. At least two of these considerations strongly militate towards adopting the same method in this context. First, the text yields the same result.

Certainly, the word “offense” itself does not provide us the same clarity as the use of the word “conviction” in the ACCA, see Nijhawan, 557 U.S. at 33-34, 129 S.Ct. 2294; see also Black’s Law Dictionary (9th ed. 2009) (defining offense as “violation of the law; a crime”), but we have recently ascribed meaning to that word in the context of the sexual, abuse statutory scheme. In United States v. Jones, 748 F.3d 64 (1st Cir.2014), a defendant challenged his life sentence, imposed under 18 U.S.C. § 2241© (providing the definition of aggravated sexual assault incorporated into the Tier III statute).

To determine whether an enhanced sentence under that statute was appropriate as a result of a predicate sexual offense, we asked whether a state law penalizing sexual penetration with a victim under 13 (Jones’s predicate crime) “would have been an offense” under federal law (§ 2241©). Id. at 73-74. After citing Descamps, we determined that “offense” in this context was limited to the “state court judgment and the statute of conviction — not at what [the defendant] did to trigger the statute’s application.” Id. at 73. In other words, we adopted a categorical approach.

We see no reason to depart from that understanding of “offense.” It is axiomatic that when considering two statutes on the same subject “courts construe words or phrases from a prior act on the same subject in the same sense,” Sutherland Statutory Construction § 51:2 (7th ed.), and that “identical words used in different parts of the same act are intended to have the same meaning,” Dep’t of Revenue of Or. v. ACF Indus., Inc., 510 U.S. 332, 342, 114 S.Ct. 843, 127 L.Ed.2d 165 (1994) (citation pmitted). Here, the SORNA statute explicitly incorporates § 2241(

c) and both sections use the term “offense” in precisely the same way; they each mandate a comparison of a predicate state offense with the federal law. The government, perhaps recognizing this, does not provide any reason to avoid this consistent reading. Instead, it merely adds a footnote in its brief saying that “some courts have expressed doubt about whether the so-called ‘categorical approach’ even applies in this setting.” Absent any justification to find otherwise, Jones answers this textual question.

Since it limits “offense” to the elements of the crime, this Descamps consideration implores us to follow the categorical approach. The third Descamps consideration, the need to avoid collateral trials about the factual grounding of the predicate offense, is also directly relevant. There, relying on its previous explanation in Taylor, the Court maintained that requiring a district court to delve into the facts of a predicate offense could turn into a daunting task. Indeed, In some cases, the indictment or other charging paper might reveal the theory or theories of the case presented to the jury.

In other cases, however, only the Government’s actual proof at trial would indicate whether the defendant’s conduct constituted [the offense]. Would the Government be permitted to introduce the trial transcript before the sentencing court, or if no transcript is available, present the testimony of witnesses? Could the defense present witnesses of its own and argue that the jury might have returned a guilty verdict on some theory that did not require a finding that the defendant committed the [generic offense]? Taylor, 495 U.S. at 601, 110 S.Ct. 2143.

Relatedly, the Court observed that where a defendant pleads to a lesser included offense of one listed in the ACCA, this fact-intensive approach could subject a defendant to a mini-trial on, and an enhanced punishment from, that broader crime. Id. In effect, this could deprive him or her of the benefit of the plea bargain. Not much more need be said here, as a similar concern also points to the answer in this case. Under a fact-centric analysis, an equally intensive investigation could be required to determine whether a defendant’s predicate actions fall within the Tier Ill category.

Although it is plausible that the issue could be resolved from the plea agreement, plea colloquy, or judgment alone, it is also conceivable that the court would need to review the entire record. As the Court in Taylor noted, it could even require further factual development outside of the initial trial record. This is particularly concerning in this context where requiring the victim to testify again could increase the likelihood of secondary trauma. See, e.g., L. Christine Brannon, The Trauma of Testifying in Court for Child Victims of Sexual Assault v. the Accused’s Right to Confrontation, 18 Law & Psychol.

Rev. 439 (1994). Finally, depending on the nature of the defendant’s plea to the underlying offense, the approach could also render his or her bargain meaningless. Ultimately, limiting our analysis to the elements of indivisible state predicate offenses, along with the generic crimes referenced in SORNA, best comports with the Supreme Court’s considerations in Descamps. Indeed, we are not alone in reaching this conclusion. See United States v. Backus, 550 Fed.Appx. 260 (6th Cir.2014) (applying the categorical approach to the SORNA context); United States v.

Cabrera-Gutierrez, 756 F.3d 1125 (9th Cir.2013) (same); see also United States v. Forster, 549 Fed.Appx. 757 (10th Cir.2013) (casting some doubt on the approach but applying it nonetheless); contra United States v. Gonzalez-Medina, 757 F.3d 425 (5th Cir.2014) (applying a fact-based approach when comparing an “age differential” distinction between a state and federal statute). Therefore, our analysis will be limited solely to the elements of the relevant statutes. b. The heart of this case turns on whether the Rhode Island law is “comparable to or more severe than” any of the SORNA Tier III offenses.

At its core, Tier III breaks down into two categories: (1) offenses that apply irrespective of the victim’s age—i.e., if the offense either includes some additional conduct or harm, or is committed' against a particularly vulnerable individual, § 16911(4)(A)(i); and, (2) any sexual crime against a victim 12 or under, § 16911 (4)(A)(i)-(ii). We address each category in turn. Initially, Tier III applies in a number of situations irrespective of the victim’s age.

Specifically, it applies to offenders who engage in certain, additional conduct (other than the sexual act alone) in the course of committing the crime; who inflict some additional harm during the offense; or who abuse a specified victim. § 16911(4)(A)(i) (enumerating these offenses and including: offenders who pom-mit the crime utilizing force or the threats of force, offenders who commit the crime against an unconscious victim, or offenders who eommit the crime against a victim who is incapable of appraising the nature of the conduct). The Rhode Island statute lacks analogous characteristics.

Instead, it contains only two elements: the sexual act and the age of the victim. It does not include additional elements such as threats or force, nor does it narrowly protect specific classes of victims in a comparable fashion. Quite simply, the Rhode Island law in this case penalizes significantly broader behavior than this category of Tier III offenses. Our examination continues, however, since Tier III also encompasses any sexual act against a victim aged 12 or under. Specifically,

section (ii) of Tier III includes offenses where the victim is 12 years old or under and the perpetrator is at least four years older, § 16911(4)(A)(ii), and

section (i) (in addition to the crimes previously discussed) includes offenses where the perpetrator crosses state lines with the intent to engage in sexual conduct, or actually does engage in such conduct, if the individual is under the age of 12, § 16911 (4)(A)(i). Such crimes are considered so severe that, without anything more, they warrant Tier III designation. The government anchors its argument in the latter,

section (

i) crime. It emphasizes that

section (

i) of the statute penalizes the mere intent to sexually abuse an individual under 12 (together with the actus reus of crossing state lines), while Rhode Island penalizes more severe conduct: an actual sexual act. The government simply views the difference in the age cut-offs as inconsequential. The government’s position runs head first into a congressional judgment that lies at the core of the tiered framework.

The structure of the law makes clear that while “comparable to” may, as the government argues, provide us some flexibility in examining the offenses, the question of age is so essential to the framework that the congressional cut-off must be strictly construed. See Saysana v. Gillen, 590 F.3d 7, 13 (1st Cir.2009) (stating that the meaning “of a statutory provision is often made clear not only by the words of the statute but by its structure as well”). This is manifested by the distinctions within

section (

i) of Tier III, by the interplay of

section (

i) and (ii), and by the differences between Tier II and Tier III.' First, the structure of

section (

i) of Tier III implies that this congressional line-drawing was critical. At the risk of repetition, Congress determined that the intent to abuse a victim under 12 (or actually doing so) warrants Tier III classification. But, where a victim is any other age, Congress made a different statement; although the conduct may be severe, without additional evidence of other conduct or harm (such as threats or force), or some other particularly vulnerable attribute about the victim (such as being unable to comprehend what was happening), only Tier II applies.

Thus, for crimes against victims over 12, only additional conduct, or a more vulnerable victim, renders the crime commensurate with an offense against a victim under that age. Comparing the different sub-sections in Tier III yields the same conclusion.

Section (

i) of Tier III already covers all conduct where the victim is under 12. § 16911(4)(A)(i) (incorporating 18 U.S.C. § 2241(c) (“intent to engage in ... [or] knowingly engages in a sexual act with another person who has not attained the age of 12.”)).

Section (ii) adds only one additional class of offenses: specified sexual abuse against 12 year old children. § 16911(4)(A)(ii). If the age limit in

section (

i) could be disregarded, as the government suggests, then a state crime penalizing a sexual act against a 12 year old would already be captured by

section (

i) of Tier III. This would leave the second subsection of Tier III without any purpose. See Duncan v. Walker, 533 U.S. 167, 174, 121 S.Ct. 2120, 150 L.Ed.2d 251 (2001) (noting the importance of giving every word or

section in a statute meaning when possible). The differences across the tier hierarchy further underscore the age distinction’s importance. Compare § 16911(3)(A)(iv) with § 16911(4)(A). If a person engages in sexual conduct against a victim 16 or under and' is convicted under a state’s general statutory rape law, Tier II status is imposed. § 16911(3)(A)(iv). If the victim is 12 or under, however, then the same exact conduct warrants Tier III classification. In crafting this law, Congress plainly did not envision placing every offender who violated any statutory rape law under the Tier III umbrella.

But, here, too, the government’s position would likely do just that. Indeed, it would raise the question of whether Tier III applied to a violation of any state’s statutory rape law. By moving a significant number of Tier II offenders into Tier III in this way, an entire

section of Tier II could thus be left without any purpose. Nor is the tier system the only place that Congress emphasized this age distinction in the sexual abuse statutory regime. See Util. Air Regulatory Grp. v. EPA — U.S. -, 134 S.Ct. 2427, 2442, 189 L.Ed.2d 372 (2014) (noting that a provision in a statute must be read in the broader context of the law); Robinson v. Shell Oil Co., 519 U.S. 337, 342-45, 117 S.Ct. 843, 136 L.Ed.2d 808 (1997) (trying to decipher the meaning of a term by examining its usage in other provisions in the statute).

For example, Congress specifically enhanced the penalty for “offenses involving young children.” 18 U.S.C. § 2244(c). In that law, Congress said that if a victim had not attained the age of 12 “the maximum term of imprisonment that may be imposed for the offense shall be twice that otherwise provided in this section.” Id. The purpose of that enhancement, Congress specifically noted, was its concern over an increase in abuse (roughly one-third of all sexual offenses) against children 11 and under. H.R.Rep. No. 105-557 (1998).

This suggests that Congress was not setting an arbitrary age limit, but was intentionally drawing a line at that specific age. Ultimately, the age of the victim is a critical component of the tier system. Although it now sings a different tune, even the government has acknowledged this fact.

As the Department of Justice (“DOJ”) noted in its own guidelines implementing the law, the tier designation increases requirements corresponding to a number of factors, most prominently, “the nature and seriousness of the offense, the age of the victim, and the extent of the offender’s recidivism.” National Guidelines for Sex Offender Registration and Notification, 73 Fed. Reg. 38,030 (July 2, 2008) (emphasis added). The DOJ seems to accept the idea, which we also adopt, that the victim’s age is one of the core elements distinguishing the tiers in SOR-NA.

Given that, a state law simply does not target comparatively grave conduct when it fails to include the same age cutoff. The Rhode Island law at issue here is significantly broader than a Tier III offense, since the state law penalizes sexual conduct alone—without anything more—against victims over the eongressionally-designated age of 12. Although Rhode Island can certainly draw the line at 14 when crafting its own laws and setting its own registration requirements, it does not necessarily follow that specific federal requirements are automatically triggered.

Where a distinction exists on such a foundational issue, we cannot consider the two laws comparable. Admittedly, this would be a different case if the Rhode Island statute required proof of some additional harm (even if it defined the harm differently than the federal statute) to victims over 12, since it would then mirror Congress’s judgment on this crucial point. See, e.g., N.H.Rev.Stat. Ann. § 632-A:2(R; M.G.L. ch. 265 § 23A. It would also be a different ease if Rhode Island changed its laws to penalize sexual conduct against any individual 12 or under, as there would then be overlap with the federal scheme.

See, e.g., Me. Rev. Stat. Ann. tit. 17-A § 253(l)(e); N.H.Rev.Stat. Ann § 632-A:2(I)(Z) & (II). It might even be a different case if the Rhode Island law under which Morales was convicted specifically sanctioned abuse against victims with a mental capacity equivalent to a child 12 or under, as that might indicate an attempt to regulate equivalent harm. But, in penalizing sexual conduct alone against older victims, the law sweeps far too broadly in terms of the severity of the offenses it covers to be a Tier III congener.

Accordingly, the Rhode Island law cannot be read to be “comparable to or more severe than” any SORNA Tier III offense. Thus, the district court erred when it characterized Morales as a Tier III sex offender. ii. Although we find error, our analysis does not end there. Instead, Morales must still establish the other prongs of plain error. First, he must show that the error was “plain.” In other words, it must be “clear” or “obvious.” See United States v. Olano, 507 U.S. 725, 734, 113 S.Ct. 1770, 123 L.Ed.2d 508 (1993).

The government contends that we have never interpreted the SORNA tier regime in this context and that the conclusion we reach today is far from self-evident. The plainness of an error is considered at the time of an appeal, and we thus account for developments in the law even if the district court did not have the benefit of those changes. See Henderson v. United States, — U.S. -, 133 S.Ct. 1121, 1124-25, 185 L.Ed.2d 85 (2013); United States v. Farrell, 672 F.3d 27, 36 (1st Cir.2012).

Despite the implication of the government’s argument, the absence of a decision directly on point does not remove the potential for a finding of plain error. Instead, the inquiry is always whether the error is open to doubt or question. See Henderson, 133 S.Ct. at 1130 (“ ‘[P]lain’ means that lower court decisions that are questionable but not plainly wrong ... fall outside the Rule’s scope.”); Puckett v. United States, 556 U.S. 129, 143, 129 S.Ct. 1423, 173 L.Ed.2d 266 (2009).

Two interwoven aspects of the error here make it “plain.” First, although the district court in this case should have conducted some analysis given Descamps, it was our recent decision in Jones that settled the matter definitively. With that case, one the district court did not have the benefit of reviewing, it became sufficiently clear that the Descamps analytical framework applied to the tier structure at issue.

Second, with the Jones decision in place (and assuming, arguendo, that the government’s broader conception of “comparable to” applies) all this case then required was a comparison of the federal and state law. Although we have engaged in a thorough examination, even a cursory review of the statutes illustrates the importance of the age issue in the federal law and the breadth of the Rhode Island statute in comparison. See United States v.

Dávila-Félix, 667 F.3d 47, 54-57 (1st Cir.2011) (finding plain error despite undertaking a significant legal analysis asking whether a prior conviction qualified as a predicate offense for sentencing enhancement purposes). Simply put, no faithful reading of the elements of the statutes, as our analysis shows, could lead one to conclude that Morales was a Tier III offender. Thus, given the framework mandated by Jones, and the clarity with which Congress spoke in SORNA, it was sufficiently obvious that Morales was incorrectly classified as a Tier III offender.

Morales must next show that the error affected his substantial rights. In other words, it “must have been prejudicial.” Olano, 507 U.S. at 734, 113 S.Ct. 1770. Here, it is important to distinguish Morales’s incarcerative sentence from his supervised release term. As to the former, this case is somewhat analogous to others where much is at stake (for example, when an armed career criminal designation plays a role) and plain error is found because the district court erroneously considered a pri- or conviction to be a predicate offense. See United States v. Torres-Rosario, 658 F.3d 110, 116 (1st Cir.2011).

As we noted in Torres-Rosario, “district courts have regained considerable discretion in sentencing but the guidelines are still highly influential.” Id. Thus, where the guidelines are augmented because the court improperly considers a prior conviction to be a predicate offense, we must, at a minimum, be on alert to the presence of prejudice. In this case, the district court seemingly followed the guidelines, and imposed an incarcerative sentence directly in the middle of the range.

Given the district court’s focus on the supervised release term rather than on the incarcerative sentence, we are unable to say with sufficient confidence that the erroneous guidelines range did not single-handedly drive the district court’s incarcerative sentencing decision. (“Now, the enhancement puts you in a guideline range of 57 to 71 months. If you are truly a child abuser, then the difference between 57, 60, 65, 71 months just keeps you off the street for a few more months, and eventually you’re going to return to society.

So just as important as the prison term, it seems to me, are the terms of your release.”) Given the apparent role that the incorrect guideline range played, and the reasonable probability that a different sentence will be imposed, remand is appropriate. See, e.g., United States v. Antonakopoulos, 399 F.3d 68, 81 (1st Cir.2005) (noting with respect to a Booker error that “[e]ven in cases where the judge was silent, there may be cases in which the appellate panel is convinced by the defendant based on the facts of the case that the sentence would, with reasonable probability, have been different”).

Although Morales has shown that his substantial rights were affected as to his prison term, he cannot show prejudice with respect to his supervised release sentence. At the initial sentencing, the district court noted that a lifetime of supervised release was required because Morales was a recidivist and the public needed to be protected. After the initial sentencing, the government informed the court that it had changed it position. It believed that the guidelines called for a maximum of five years of supervised release even though the statute permitted the court to vary upwards to a lifetime sentence.

At a subsequent hearing, the court again emphasized that public safety concerns mandated the lifetime order and it thus reaffirmed that part of its decision. Nothing from either proceeding indicates that the tier designation played any role in this part of the sentence, and we see no reason to disrupt it. Morales must finally show that the error seriously impaired the fairness, integrity, or reputation of the judicial proceedings. In other words, he must show that there is “a threat of injustice if we affirm.” United States v. Rodriguez, 630 F.3d 39, 42-43 (1st Cir.2010).

Given the significant possibility that Morales will receive a reduced sentence upon remand with a different base-offense level, and since SORNA imposes enhanced, lifelong requirements for Tier III offenders, such a threat does exist without correcting this error. Nor, we note, is there any injustice to the government from a remand. See United States v.

Ramos-Gonzalez, 775 F.3d 483, 507-08 (1st Cir.2015) (“To the extent relevant to the plain error inquiry, the government asserts no offsetting circumstances.”) Since the error was one purely of law, the government faced no undue prejudice in responding to the argument for the first time on appeal. Moreover, the government faces no significant harm moving forward, as it can still argue for an upward variance upon remand. Ultimately, this is one of the rare cases in which the plain error review standards have been satisfied and a remand to correct the mistake of law is required. ÍV.

As the district court committed plain error when it characterized Morales as a Tier III sex offender and when it then utilized that designation in its guidelines calculation, we vacate the sentence with respect to Morales’s prison term and remand for proceedings consistent with this opinion. . We have considered Morales's additional contentions, and find them to be wholly unpersuasive. .

These crimes include: causing another to engage in a sexual act "by using force ... or by threatening that other person” with force, § 2241(a); knowingly rendering a person unconscious or drugging that individual and then engaging in a sexual act with him or her, § 2241(b); or, engaging or attempting to engage in abuse defined in § 2241(a) & (

b) with a person between the ages of 12 and 15 if the perpetrator is four years older than the victim, § 2241©. § 1691 l(4)(A)(i). This category also includes forcing "another person to engage in a sexual act by threatening or placing that other person in fear” or engaging in a sexual act with a victim if such an individual is "incapable of appraising the nature of the conduct; or physically incapable of declining participation in, or communicating unwillingness to engage in, that sexual act.” § 16911 (4)(A)(i) (incorporating 18 U.S.C. § 2242). .

Despite its tepid argument against the categorical approach, the government vigorously insists that the phrase "comparable to" should have broad meaning. Morales, meanwhile, argues for a narrower construction of the term, similar to that given by two other circuits. See Backus, 550 Fed.Appx. at 263 (defining "comparable” as "prohibit[ing] the same activity”); Cabrera-Guiterrez, 756 F.3d at 1133-34 (stating that a statute that covers more activity than the federal statute is not comparable to the federal law). We need not decide this issue.

Instead, we conclude that, no matter what the precise meaning of the term, the Rhode Island offense in question is not "comparable to” the federal offenses listed in Tier III. . We also note that other sections of Rhode Island law do proscribe force or additional harm in this context. See, e.g., R.I. Gen. Laws § 11-37-2. This evidences the legislature's capacity to target such activity when it so intends. . Although the government focuses on the

section (

i) intent crime, the importance of the age distinction would apply equally when comparing the Rhode Island statute to the offenses barred in

section (ii) of Tier III (sexual conduct against a 12 year old child). However, we could potentially resolve that

section (ii) comparison in another way. Unlike the offenses listed in sub-section (i), sub-section (ii) could either be read as a single unit (i.e. finding that "is comparable to or more severe than” applies to the entire section), or it could be read as two separate clauses ("is comparable to or more severe than ... abusive sexual contact” and is "against a minor who has not attained the age of 13 years.”) Under the latter approach, the “comparable to” language might not apply to the age limit and thus that clause could be read as a standalone requirement.

Courts could then engage in a factual inquiry into the victim’s actual age when considering whether an offense matched sub-section (ii) of Tier III. Because the victim of the underlying crimes for which Morales was convicted was 13, we need not make that determination, since we would reach the same result regardless of which analytical framework applied to that one specific clause of Tier III. .

Although we do not find that the plain error standard has been met with respect to supervised release, determining the appropriate balance of an incarcerative term and supervised release is an exercise committed to the sound discretion of the district court. The district judge should therefore be permitted, but not required, to revisit the supervised release aspect of the sentence on remand.