Rico v. United States - Annotated Opinion

 Whatever the wisdom of a policy like that, we see nothing in the law authorizing it. The Sentencing Reform Act provides courts with many tools to address defendants who fail to report or otherwise violate their supervised release conditions. But automatically extending a term of supervised release is not among them. An array of textual clues proves the point.

 Start with what the Act says about when supervised release begins and ends. The Act instructs that a term of supervised release starts “the day the person is released from imprisonment.” §3624(e). The Act then tells us when a term of supervised release must end, generally setting its maximum length at one, three, or five years, depending on the severity of the defendant’s underlying offense. §3583(b). Neither provision hints at anything like the Ninth Circuit’s automatic extension rule. To the contrary, that rule risks flouting the Act by permitting courts to extend supervised release beyond even the maximum terms set by Congress.

 Next, consider how the Act treats absconders. As a condition of supervised release, a court may order a defendant to report to a probation officer and inform authorities if he changes his residence. See §§3563(b)(15), (17). Should a defendant violate those (or any other) conditions, a court may (and sometimes must) revoke his supervised release and send him back to prison with a new term of supervised release to follow. §§3583(e)(3), (g). That the Act authorizes those potent tools for addressing absconders—but makes no mention of anything like the Ninth Circuit’s automatic extension rule—is telling too.

 Relevant as well, the Act authorizes other extension and (true) tolling rules. Take a few examples. Section 3583(e)(2) permits a court to “extend a term of supervised release,” but only subject to important limitations. A court must generally hold a hearing and consider various sentencing factors. See §3583(e)(2); Fed. Rule Crim. Proc. 32.1. A court cannot extend supervised release beyond the maximum term set by statute. See §3583(e)(2). Nor may a court extend a term of supervised release if that term has already expired. See ibid.; Brief for United States 8, 38, 45. The Ninth Circuit’s automatic extension rule disregards not just one but each of these limits. Under its approach, no hearing is needed, a term of supervised release may extend beyond the statutory maximum, and a court-ordered expiration date becomes meaningless.

 Section 3583(i) also addresses extensions. It provides that a court’s power to revoke supervised release “extends beyond the expiration of the term of supervised release for any period reasonably necessary for the adjudication of matters arising before its expiration if, before its expiration, a warrant or summons has been issued on the basis of an allegation of such violation.” Here, then, the Act speaks directly to what a court may do after the expiration of a defendant’s judicially ordered term of supervised release. But (again) the Act bounds a court’s authority. A court may adjudicate only “matters arising before” the expiration of the defendant’s term of supervised release. It may do so only if a warrant or summons issued during that term. And nothing in this provision allows a court to exceed statutory maximums. The Ninth Circuit’s rule, meanwhile, blows past all those constraints, allowing courts to revoke supervised release for violations arising after the judicially decreed supervised release term expires—and sometimes even after the statutory maximum period—all without the necessity of a warrant or summons.

 If more evidence were needed to illustrate how unlikely the Ninth Circuit’s rule is, §3624(e) would supply it. That provision indicates that a “term of supervised release does not run during any period in which the person is imprisoned in connection with a conviction” for 30 consecutive days or more. This is a true tolling rule, one that stops the clock running on a defendant’s term of supervised release. Yet, like the Act’s extension rules, it is narrow in scope and nothing at all like the rule the Ninth Circuit has adopted.

 The quantity of all these instructions has a quality of its own, too. Cumulatively, the Act details how long supervised release may run, the tools courts can use to address defendants who fail to report or commit other violations, when courts may extend and toll terms of supervised release, and how they can go about addressing violations even after a term of supervised release has expired. To our eyes, the absence of anything like the Ninth Circuit’s rule in all these exacting instructions is striking and strongly suggestive that the Ninth Circuit’s rule more nearly represents an adornment to Congress’s work than a permissible interpretation of it.

 Seeking to persuade us otherwise, the government starts this way. Under §3601, it observes, a defendant on supervised release “shall, during the term imposed, be supervised by a probation officer.” Many other provisions speak similarly. See, e.g., §§3603(2), (3), 3624(e). As a matter of ordinary meaning, the government submits, supervision “requires both observation and direction.” Brief for United States 12. Because Ms. Rico was subject to neither of those things from early 2018 until 2023, the government reasons, she should receive no “credit” for serving on supervised release during that period. Ibid. And, as a result, the district court was within its rights to hold that she violated the terms of her supervised release when she committed her January 2022 drug offense. Id., at 18–23.

 If attractive at first blush, this line of reasoning suffers from a few underlying flaws. For one thing, the provisions the government cites do not outline anything like the Ninth Circuit’s automatic extension rule. Instead, they merely describe the individual responsible for supervising the defendant—the probation officer—along with some of his duties. For another, to the extent these provisions speak about the length of supervised release at all, they indicate that a probation officer shall supervise a defendant only “during the term imposed” by the sentencing court. See §§3601, 3624(e). And that limitation would seem to do more to hurt than help the government’s cause.

 For another thing still, an anomaly lies at the heart of the government’s theory. The government contends that Ms. Rico was not supervised from early 2018 until 2023 and, accordingly, should not have that period counted toward her term of supervised release. But in the same breath, the government argues that her January 2021 state offenses and her January 2022 state drug offense count as federal violations because her term of supervised release continued to run during the entire length of her abscondment. In a very real sense, then, the government asks us to imagine that Ms. Rico was both off and on supervised release at the same time. Really, it is quite the puzzle.

 If its arguments from statutory text fall short, the government replies, its arguments from precedent fill the gap. In particular, the government suggests, our decisions in Mont and United States v. Johnson, 529 U. S. 53 (2000), require us to adopt the Ninth Circuit’s rule. But here, too, we just don’t see it. Mont simply recognized that §3624(e)’s express terms suspend (or truly toll) a defendant’s term of supervised release while he remains imprisoned for a separate state offense before trial. 587 U. S., at 521. And if that express tolling rule highlights anything, it is only, as we have seen, the absence of anything like the Ninth Circuit’s rule in the statutory text. See Part II–A, supra.

 Johnson is even further afield. There, the defendant remained in prison longer than he should have. 529 U. S., at 54. After realizing the mistake, the district court ordered the defendant’s release and instructed that he should begin serving his term of supervised release. Id., at 55. The defendant appealed a portion of that order, arguing that his term of supervised release should have started “when his lawful term of imprisonment expired.” Id., at 55–56. Ultimately, we rejected that argument, reasoning that the Act’s terms could not sustain it. Id., at 56–58. How any of that might require us to adopt the Ninth Circuit’s rule escapes us. Perhaps Johnson’s only relevance to this case lies in this parallel to our case: We proceed today much as we did there, rejecting yet one more effort to adorn the Act with a rule Congress could have adopted but did not enact.

 Moving past our precedents, the government next appeals to the common law. For centuries, the government observes, common-law courts have held that the time between a defendant’s escape from prison and his recapture generally does not count toward the discharge of his sentence. It is a rule, the government notes, premised on the commonsense view that an escapee should take no “ ‘manner of advantage from it.’ ” Brief for United States 25 (quoting 2 W. Hawkins, Pleas of the Crown 200 (6th ed. 1788)). And it is a rule, the government says, we should assume Congress meant to carry forward under the Sentencing Reform Act—and one that should ultimately lead us to adopt the Ninth Circuit’s rule. Brief for United States 29.

 The government’s premises here may be sound, but its conclusion is not. Congress may sometimes legislate against the backdrop of the common law. We can also assume for argument’s sake that the government is right about the common-law rule and courts generally paused the clock on a defendant’s sentence and its attendant conditions when he escaped from prison and remained at large. But, as we have seen, the government seeks a very different rule here. Its proposed rule does not stop any clock or merely ensure a defendant does not take advantage of his abscondment. It imposes a new punishment on him by automatically extending his term of supervised release.

 Look at it this way. When a prisoner escapes, he is in no sense serving his prison sentence. In contrast, when a defendant on supervised release fails to report, everyone agrees he remains bound by the terms of his release. That is why, for failing to report and any other violation he may commit during his judicially ordered term of supervised release, a court may send a defendant to prison and authorize more supervised release yet. §3583(e)(3). The court can do all that, too, even after the defendant’s prescribed term of supervised release expires, so long as a warrant or summons issues beforehand. §3583(i). The Act thus already provides courts with many ways to ensure a defendant does not profit from a supervised release violation. But what the Act does not do is automatically extend the defendant’s period of supervised release beyond what a judge has ordered. And that additional rule is hardly necessary to ensure that a defendant should take “no manner of advantage” from his abscondment.

 The government urges us to consider Anderson v. Corall, 263 U. S. 193 (1923). That case, the government insists, illustrates how background common-law principles should inform our understanding of the Act. In Corall, a federal court sentenced the defendant to prison, followed by a term of parole, a form of conditional release different from (but not wholly unlike) supervised release. Id., at 193–194. While the defendant was on parole, he committed a new state crime and was sentenced to state prison. Id., at 194. After his release, federal authorities revoked the defendant’s parole and returned him to federal prison. Id., at 194–195. The defendant protested that his federal parole clock had expired while he was in state prison, but we disagreed. Instead, applying background common-law principles, we held that the defendant’s state confinement stopped the clock on his federal parole much as a prisoner’s escape stops the clock on his prison sentence. Id., at 196–197.

 We fail to see how any of this moves the needle. Corall might teach that, thanks to background common-law principles, a defendant’s federal parole clock stops running during his time in state prison. But what does that prove? When it comes to supervised release, as we have seen, Congress has already codified a similar rule, providing that a defendant’s incarceration for 30 days or more pauses his term of supervised release. §3624(e). And none of that means the government is entitled to an additional and very different rule that would automatically extend a defendant’s term of supervised release beyond what a court has ordered and possibly even beyond the statutory maximum. Simply put, Corall did not even contemplate, let alone endorse, anything like the Ninth Circuit’s rule.

 In the end, the government is left to resort to a policy plea. The Act may supply courts with many tools to address absconding defendants. A court may even revoke a defendant’s supervised release after his judicially ordered term of supervised release has expired so long as a warrant or summons issues before the term ends. See §3583(i). But in cases where a defendant absconds late in his supervised release term, the government protests, a probation officer may not realize his absence in time to secure a warrant or summons, leaving a court powerless to address violations after the term expires. It’s a problem, the government insists, we should remedy by endorsing the Ninth Circuit’s rule.

 This plea is misdirected. If the government thinks §3583(i)’s warrant-or-summons requirement too demanding, the proper place to register that complaint is with those who drafted it. This Court is not free to rewrite the directions Congress has provided. Nor is it clear to us, for that matter, how serious the government’s supposed problem is. Defendants may commit many violations late in their terms of supervised release. Some of those violations, like secretive or nascent criminal activities, may be difficult to detect. By comparison, a defendant’s failure to report to a probation officer as instructed is presumably among the easiest violations for federal authorities to notice. So even if we had the authority to rewrite the law’s terms to address “late-in-term-and-difficult-to-detect-violations,” it is not clear why we would start with failure to report violations.*²

 Because the Sentencing Reform Act does not authorize the rule the Court of Appeals adopted and the government advances, the judgment of the Ninth Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion.