SUPREME COURT OF THE UNITED STATES
_________________
No. 17–8995
_________________
JASON J. MONT, PETITIONER
v. UNITED
STATES
on writ of certiorari to the united states
court of appeals for the sixth circuit
[June 3, 2019]
Justice Sotomayor, with whom Justice Breyer,
Justice Kagan, and Justice Gorsuch join, dissenting.
A term of supervised release is tolled when an
offender “is imprisoned in connection with a conviction.” 18
U. S. C. §3624(e). The question before the Court is
whether pretrial detention later credited as time served for a new
offense has this tolling effect. The Court concludes that it does,
but it reaches that result by adopting a backward-looking approach
at odds with the statute’s language and by reading the terms
“imprisoned” and “in connection with” in unnatural isolation.
Because I cannot agree that a person “is imprisoned in connection
with a conviction” before any conviction has occurred, I
respectfully dissent.
I
A
The Sentencing Reform Act of 1984 empowers a
court to impose a term of supervised release following
imprisonment. See 18 U. S. C. §§3583(a), (b).
The clock starts running on a supervised release
term when the offender exits the jailhouse doors. §3624(e). During
the term, offenders are bound to follow court-imposed conditions.
Some apply to all supervised release terms, such as a requirement
to refrain from committing other crimes. §3583(d). Others apply
only at a sentencing court’s discretion, such as a condition that
the offender allow visits from a probation officer. See
§3563(b)(16); §3583(d). The probation officer, in turn, is tasked
with monitoring and seeking to improve the offender’s “conduct and
condition” and reporting to the sentencing court, among other
duties. §3603. During the supervised release term, the court has
the power to change its conditions and to extend the term if less
than the maximum term was previously imposed. §3583(e)(2). If an
offender violates any of the conditions of release, the court can
revoke supervised release and require the person to serve all or
part of the supervised release term in prison, without giving
credit for time previously served on postrelease supervision.
§3583(e)(3).
In the normal course, a supervised release term
ends after the term specified by the district court. But,
crucially, the term “does not run during any period in which the
person is imprisoned in connection with a conviction for a Federal,
State, or local crime unless the imprisonment is for a period of
less than 30 consecutive days.” §3624(e). In other words, certain
periods of “imprisonment” postpone the expiration of the supervised
release term.
A district court’s revocation power generally
lasts only as long as the supervised release term. If the court
issues a warrant or summons for an alleged violation before the
term expires, however, the court’s revocation power can extend for
a “reasonably necessary” period beyond the term’s expiration.
§3583(i).
B
Though the mechanics of supervised release
tolling may seem arcane, these calculations can have weighty
consequences. For petitioner Jason Mont, tolling enabled a court to
order an additional 31∕2 years of federal imprisonment after he
serves his current state sentence.
Mont was convicted in 2005 for federal drug and
gun crimes. The District Court sentenced him to prison time and to
five years of supervised release. In 2012, Mont was released from
prison and his supervised release term began. Left to run its
course, the term would have ended on March 6, 2017.[
1]
Mont’s time on supervised release did not go
well. In January 2016, his probation officer informed the District
Court that Mont had failed two drug tests and tried to pass two
further drug tests by using an “ ‘unknown’ ” liquid. 723
Fed. Appx. 325, 326 (CA6 2018). The officer noted that Mont also
had been charged with state marijuana-trafficking offenses. Upon
learning of these alleged violations of the supervised release
conditions, the District Court could have issued a warrant for
Mont’s arrest, but it did not do so at that time.
On June 1, 2016, Mont was arrested on a new
state indictment for trafficking cocaine, and the State took him
into custody. The probation officer reported the arrest to the
District Court, but the record does not reflect any action by the
court in response. After several months in custody, Mont pleaded
guilty to certain of the state charges. He also admitted to the
District Court that he had violated the terms of his supervised
release, and he requested a hearing. The District Court set a
November hearing to consider his alleged supervised release
violation, but continuances delayed that hearing. Months more
passed as Mont, still detained, awaited sentencing. In the
meantime, the original end date of his federal supervised release
term—March 6, 2017—came and went. On March 21, 2017, the state
court sentenced Mont to six years in prison and retroactively
credited the approximately 10 months he had spent in pretrial
detention toward his sentence.
At that point, Mont’s probation officer reported
Mont’s state convictions and sentences to the Federal District
Court, which—after its many earlier opportunities—finally issued a
warrant for Mont’s arrest on March 30, 2017. Mont objected,
claiming that the court had no power to issue the warrant because
his supervised release term had expired on March 6. The District
Court rejected that contention and sentenced Mont to 42 months in
prison, to run consecutively to his state sentence.[
2]
The United States Court of Appeals for the Sixth
Circuit affirmed. In its view, the District Court had jurisdiction
to revoke Mont’s supervised release because his pretrial detention
triggered the tolling provision in §3624(e) and thus shifted back
the end date of his supervised release term. The Sixth Circuit
construed the tolling provision to apply to Mont’s detention
because his state-court indictment ultimately led to a conviction
and Mont subsequently received credit for the period of detention
as time served for that conviction.
II
The majority errs by affirming the Sixth
Circuit’s construction of the tolling statute. Most naturally read,
a person “is imprisoned in connection with a conviction” only while
he or she serves a prison term after a conviction. The statute does
not allow for tolling when an offender is in pretrial detention and
a conviction is no more than a possibility.
The first clue to the meaning of §3624(e) is its
present-tense construction. In normal usage, no one would say that
a person “is imprisoned in connection with a conviction” before any
conviction has occurred, because the phrase would convey something
that is not yet—and, indeed, may never be—true: that the detention
has the requisite connection to a conviction. After all, many
detained individuals are never convicted because they ultimately
are acquitted or have their cases dismissed.[
3] Until a conviction happens, it is impossible to
tell whether any given pretrial detention is “connect[ed] with” a
conviction or not.
Reading the phrase “is imprisoned” to require a
real-time assessment of the character of a conviction does not just
match the colloquial sense of the phrase; it also gives meaning to
the tense of the words Congress chose. The Court generally “look[s]
to Congress’ choice of verb tense to ascertain a statute’s temporal
reach.”
Carr v.
United States,
560 U.S.
438, 448 (2010). Doing so abides by the Dictionary Act, which
provides that “words used in the present tense include the future
as well as the present” absent contextual clues to the contrary, 1
U. S. C. §1, and thus “the present tense generally does
not include the past,”
Carr, 560 U. S., at 448.
Applying this presumption here leads to the straightforward result
that the phrase “is imprisoned” does not mean “was imprisoned.”
Adhering to the present-tense framework of the statute, then,
pretrial detention does not meet the statutory definition, no
matter what later happens.
The other language in §3624(e)—“imprisoned in
connection with a conviction”—confirms this result. Had Congress
wanted to toll supervised release during pretrial confinement, it
could have chosen an alternative to the word “imprisoned” that more
readily conveys that intent, such as “confined” or “detained.” See
Black’s Law Dictionary 362 (10th ed. 2014) (defining “confinement”
as “the quality, state, or condition of being imprisoned or
restrained”);
id., at 543 (defining “detention” as “[t]he
act or an instance of holding a person in custody; confinement or
compulsory delay”). Instead, Congress selected a
word—“imprisoned”—that is most naturally understood in context to
mean postconviction incarceration.
Congress regularly uses the word “imprisoned”
(or “imprisonment”) to refer to a prison term following a
conviction. The United States Code is littered with statutes
providing that an individual shall be “imprisoned” following a
conviction for a specific offense. See,
e.g., 18
U. S. C. §§1832, 2199, 2344. Congress also classifies
crimes as felonies, misdemeanors, or infractions based on “the
maximum term of imprisonment authorized.” §3559(a). And even in the
Sentencing Reform Act itself, which added the tolling provision at
issue, Congress used the word “imprisonment” to refer to
incarceration after a conviction. See §3582(a) (describing the
factors courts consider when imposing “a term of imprisonment”);
§3582(b) (referring to “a sentence to imprisonment”);
§3582(c)(1)(B) (discussing when courts may “modify an imposed term
of imprisonment”).
This Court also has previously equated the word
“imprisonment” with a “prison term” or a “sentence”—phrases that
imply post-trial detention. See
Tapia v.
United
States,
564 U.S.
319, 327 (2011) (referring in passing to “imprisonment” as a
“prison term”);
Barber v.
Thomas,
560 U.S.
474, 484 (2010) (“[T]erm of imprisonment” can refer “to the
sentence that the judge imposes” or “the time that the prisoner
actually serves” of such a sentence); see also
Argersinger
v.
Hamlin,
407 U.S.
25, 37 (1972) (“[N]o person may be imprisoned for any offense
. . . unless he was represented by counsel at his
trial”).
To be sure, dictionary definitions of the word
“imprison” sweep more broadly than just post-trial incarceration.
See
ante, at 6. But the word “imprisoned” does not appear in
this statute in isolation; Congress referred to imprisonment “in
connection with a conviction.” As part of that phrase and given its
usual meaning, the word “imprisoned” is best read as referring to
the state of an individual serving time following a conviction.
The present tense of the statute and the phrase
“imprisoned in connection with a conviction” thus lead to the same
conclusion: Pretrial detention does not toll supervised
release.[
4]
III
The majority justifies a contrary
interpretation of the tolling provision only by jettisoning the
present-tense view of the statute and affording snippets of text
broader meaning than they merit in context.
The majority’s first error is its conclusion
that courts can take a wait-and-see approach to tolling. If a
conviction ultimately materializes and a court credits the
offender’s pretrial custody toward the resulting sentence, the
majority reasons, then the pretrial detention retroactively will
toll supervised release. If not, then there will be no tolling. See
ante, at 6–8. The offender’s supervised release status thus
will be uncertain until the court calculates tolling either “upon
the defendant’s release from custody or upon entry of judgment.”
Ante, at 8.
The majority’s retrospective approach cannot be
squared with the language of §3624(e). Because Congress phrased the
provision in the present tense, the statute calls for a
contemporaneous assessment of whether a person “is imprisoned” with
the requisite connection to a conviction. The majority erroneously
shifts the statute’s frame of reference from that present-tense
assessment (what is) to a backward-looking review (what was or what
has been).[
5]
The majority’s textual argument hinges on what
the majority perceives to be an advantage of the retrospective
approach: It accounts for the fact that the statute provides for
tolling only if a period of imprisonment lasts longer than 30 days.
§3624(e). According to the majority, the 30-day provision shows
Congress’ expectation that courts look backwards when evaluating
whether tolling is appropriate. If Congress anticipated such an
analysis as to the length of the detention, the majority implies,
surely it provided more generally for backward-looking review of
the relationship between the detention and any ensuing conviction.
See
ante, at 8.
This argument, however, assumes a problem of the
majority’s own making. The 30-day minimum creates no anomalies if
the statute is read to toll supervised release only during
detention following a conviction. Under that more natural reading,
courts in most cases will not be left in the dark about the length
of a period of detention or its relationship to a conviction; the
conviction and sentence of imprisonment at the time imposed will
answer both questions.[
6]
Under the majority’s approach, however, this
language creates a dilemma. Unlike a term of imprisonment following
a conviction, the duration of pretrial confinement is uncertain at
its outset. Thus if (as the majority contends) Congress meant to
toll such periods of detention, the 30-day limitation means that
every single time a person on supervised release enters detention,
it will be unclear for up to a month whether the supervised release
term is being tolled or not. See
ante, at 8 (conceding that
there will be “no way for a court to know on day 5 of a defendant’s
pretrial detention whether the period of custody will extend beyond
30 days”). If pretrial detention lasts longer than 30 days, the
uncertainty will continue until a judgment of conviction is entered
and credit for pretrial detention is computed.
But the difficulties inherent in predicting how
long pretrial detention will last (and whether that detention
eventually will turn out to have any connection to a conviction,
see
supra, at 4–5, and n. 3) most naturally compel the
conclusion that Congress never intended to force district courts to
grapple with them in the first place. These uncertainties generally
would not arise—and courts thus would not need to rely on
hindsight—if the Court were to adopt Mont’s reading. Yet the
majority instead takes as a given that the statute tolls supervised
release during pretrial detention, and then uses the uncertainties
inherent in that process to justify a backward-looking
analysis.
The majority’s error is compounded by the
centerpiece of its textual analysis, which relies on artificially
isolating the terms “imprisoned” and “in connection with.” The
majority says that imprisonment is a term so capacious as to
encompass pretrial detention,
ante, at 6–7, and that the
phrase “in connection with” sweeps broadly enough to include
pretrial detention that is ultimately credited to a new sentence,
ante, at 7.
Whether or not these phrases independently have
the far-reaching meaning that the majority ascribes to them—a
conclusion that is by no means inevitable—the terms are still
limited by their relationship to each other and by the
present-tense framework of the statute. Individual phrases must not
be taken “ ‘in a vacuum,’ ” because doing so overrides
the “ ‘fundamental canon of statutory construction that the
words of a statute must be read in their context and with a view to
their place in the overall statutory scheme.’ ”
Home Depot
U. S. A., Inc. v.
Jackson,
ante, at ___
(slip op., at 5) (quoting
Davis v.
Michigan Dept. of
Treasury,
489 U.S.
803, 809 (1989)). As discussed, in the context of a phrase
referring to conviction, the term “imprisoned” most naturally means
imprisonment following a conviction.
Supra, at 5–7. And seen
from the point at which a person is detained and awaiting a
verdict, his confinement is not “in connection with” a conviction
that has not happened and may never occur.
IV
The majority’s approach has the further flaw
of treating tolling as the only meaningful avenue to preserve a
district court’s revocation power when an offender is detained
pretrial. But the statute already provides a way for a court to
extend its revocation power: If a court issues a warrant or summons
while the supervised release term is running, that action triggers
an extension of the court’s revocation authority “beyond” the
supervised release term “for any period reasonably necessary for
the adjudication” of the matters that led to the warrant or
summons. See §3583(i).
In this very case, the District Court had at
least three opportunities to issue a warrant prior to the
expiration of Mont’s original supervised release term. Mont’s
probation officer notified the District Court of Mont’s potential
supervised release violations in January 2016, more than a year
before Mont’s supervised release was set to expire. 723 Fed. Appx.
325, 326 (CA6 2018). In June 2016, the probation officer alerted
the District Court to Mont’s arrest.
Ibid. And in October
2016, Mont filed a written admission with the District Court that
he had violated supervised release.
Id., at 326–327. The
District Court was empowered at each step of this process to issue
a warrant. Indeed, the court apparently intended to do just that
after Mont’s written admission, though the Sixth Circuit later
found that there was no evidence of such a warrant in the record.
See
id., at 329, n. 5.
In sum, the delayed revocation process provides
a straightforward, and statutorily prescribed, path for district
courts to decide which charges are significant enough to justify a
warrant and thus to extend the court’s revocation power. The
majority’s overly broad reading of the tolling provision is thus
unnecessary as well as a distortion of the clear statutory
text.
V
Lacking a strong textual basis for its
backward-looking analysis, the majority is left to rely on
intuitions about how best to fulfill the statute’s purpose.
To begin with, the majority emphasizes that
supervised release and incarceration have different aims. See
ante, at 8–10. True enough. The Court has explained that
supervised release is intended “to assist individuals in their
transition to community life,” and as a result is not
“interchangeable” with periods of incarceration.
United
States v.
Johnson,
529 U.S.
53, 58–60 (2000). But the goals of supervised release can be
fulfilled to some degree even when an offender is detained. Cf.
Burns v.
United States,
287 U.S.
216, 223 (1932) (noting that a probationer is still “subject to
the conditions of” probation “even in jail”). Offenders on
supervised release may well be able to comply with several
mandatory conditions of supervised release while detained, such as
submitting to a DNA sample or taking drug tests. See §3583(d). And
probation officers have experience coordinating with correctional
facilities in the prerelease context. See §3624(c)(3) (providing
that the probation system “shall, to the extent practicable, offer
assistance to a prisoner during prerelease custody”).
Even if an offender’s detention does make it
meaningfully harder to fulfill the goals of supervised release,
moreover, the majority’s reading permits the same incongruities.
Under the majority’s interpretation, supervised release continues
to run for offenders who are confined pretrial for less than 30
days and for those who are detained pretrial but are later
acquitted or released after charges are dropped. See Tr. of Oral
Arg. 34.[
7] At best, the
majority offers a half-a-loaf policy rationale that cannot justify
departing from the best reading of the statute’s text.
The majority also invokes the general principle
against double-counting sentences, see,
e.g., §3585(b), and
objects that Mont’s reading of the statute would give defendants a
“windfall.”
Ante, at 9–10. This argument, however, fails to
recognize the distinct character of pretrial detention. Its purpose
is to ensure that an alleged offender attends trial and is
incapacitated if he or she is a danger to the community, not to
punish the offender for a conviction. See
United States v.
Morales-Alejo,
193 F.3d 1102, 1105 (CA9 1999) (citing §3142(c);
United
States v.
Salerno,
481 U.S.
739, 748 (1987)). A State or the Federal Government may later
choose to credit an equivalent period of time toward a new
sentence, but that credit does not retroactively transform the
character of the detention itself into “imprison[ment] in
connection with a conviction,” §3624(e)—particularly in the context
of this present-tense statute.
In any event, the majority’s approach creates a
serious risk of unfairness. Offenders in pretrial detention will
have no notice of whether they are bound by the terms of supervised
release. This effectively compels all offenders to comply with the
terms of their release, even though only some will ultimately get
credit for that compliance, because otherwise they risk being
charged with a violation if their supervised release term is not
tolled.[
8] Although the
majority indicates that offenders generally will comply with the
terms of their release simply by following prison rules, the range
of supervised release conditions is too broad to guarantee complete
overlap with prison directives. See,
e.g., Doherty,
Indeterminate Sentencing Returns: The Invention of Supervised
Release, 88 N. Y. U. L. Rev. 958, 1012–1013
(2013) (describing mandatory condition of cooperating in DNA
collection and special conditions of taking prescribed medications
and undergoing periodic polygraph testing). Altogether, I am not
nearly as sanguine as the majority that the uncertainty created by
the majority’s expansive tolling rule “matters little from either
the court’s or the defendant’s perspective.”
Ante, at
11.
* * *
The Court errs by treating Mont’s pretrial
detention as tolling his supervised release term. Because its
approach misconstrues the operative text and fosters needless
uncertainty and unfairness, I respectfully dissent.