PRECEDENTIAL

      UNITED STATES COURT OF APPEALS
           FOR THE THIRD CIRCUIT
               _______________

                    No. 22-1786
                  _______________

          UNITED STATES OF AMERICA

                          v.

                JAMES CHANDLER,
                                 Appellant
                  _______________

    On Appeal from the United States District Court
       for the Eastern District of Pennsylvania
             (D.C. No. 2-21-cr-00049-001)
     District Judge: Honorable R. Barclay Surrick
                  _______________

                       Argued

                    April 26, 2023

Before: JORDAN, KRAUSE, and BIBAS, Circuit Judges

                (Filed: June 11, 2024)
                   _______________
Abigail E. Horn                   [ARGUED]
FEDERAL COMMUNITY DEFENDER OFFICE
EASTERN DISTRICT OF PENNSYLVANIA
601 Walnut Street
The Curtis Center, Suite 540 West
Philadelphia, PA 19106
Counsel for Appellant

Eileen C. Geiger                [ARGUED]
Justin Oshana
U.S. ATTORNEY’S OFFICE
615 Chestnut Street, Suite 1250
Philadelphia, PA 19106
Counsel for Appellee
                      _______________

                 OPINION OF THE COURT
                     _______________

JORDAN, Circuit Judge.

        James Chandler used a fake gun to twice rob on-duty
United States Postal Service employees. In one of the
robberies, he also kidnapped his victim. His use of what
appeared to be a revolver was meant to scare his victims into
compliance, and it did. The District Court therefore did not err
when it enhanced Chandler’s sentence for using the replica gun
in the robberies and the kidnapping, nor in accepting his guilty
plea to armed robbery charges. As for the kidnapping,
Chandler’s motivation was, at least in part, that the mail carrier
was a government employee, so a further sentence
enhancement was justified. We will therefore affirm his
conviction and sentence.




                                2
I.     BACKGROUND

        Early one evening, a uniformed mailman parked his
delivery truck in West Philadelphia and began his rounds.
Chandler walked up behind him, pulled out what appeared to
be a handgun, and pressed the muzzle to his back. Chandler
was wearing a blue zip-up sweatshirt emblazoned with the
letters “USPS.” He threatened to kill the mailman and
demanded packages, cards, and cash. He then forced the
mailman to get into the back of the truck, before fleeing with
four packages and several dollars.

       A few weeks later, Chandler struck again. This time,
the victim was a mail carrier who had parked her mail truck in
West Philadelphia, half a mile from the location of the first
robbery. Chandler, wearing the same USPS sweatshirt, again
approached from behind and pointed the same fake revolver at
his victim. He made her open the truck and put about three
dozen packages into sacks. Under duress, she drove him to an
address a mile away. Chandler then fled with the packages.

        The next day, police arrested Chandler. In his
waistband, they found the revolver, which turned out to be a
gun replica that could not be fired. The postal carrier he robbed
identified him and he confessed to stealing packages. He
consented to a search of his apartment, where postal inspectors
found the blue zip-up USPS sweatshirt stuffed inside a
pillowcase.

      Chandler was charged with two counts of armed
robbery of a postal worker, under 

(a), and one
count of kidnapping a government employee, under




                               3
§ 1201(a)(5). He pled guilty to all three charges. As relevant
here, the parties disputed two enhancements at sentencing: one,
for using a dangerous weapon during a kidnapping and a
robbery, U.S.S.G. §§ 2A4.1(b)(3), 2B3.1(b)(2)(D), and the
other, for kidnapping a government employee, id.
§ 3A1.2(a)(1)(A), (a)(2). Agreeing with the government, the
sentencing judge imposed both enhancements. Those rulings
resulted in a final offense level of 37 and a guidelines range of
262 to 327 months. The judge sentenced Chandler to the
bottom of that range: 262 months’ imprisonment, with 5 years
of supervised release.

       Chandler now appeals the application of those two
enhancements, arguing that the judge erred in holding that a
replica of a gun constitutes a dangerous weapon, and further
erred in holding that his kidnapping of the second mail carrier
was motivated by her status as a government employee. He
also appeals his conviction for armed robbery, rather than
unarmed robbery, again arguing that a replica firearm is not a
dangerous weapon. 1




       1
         Chandler appeals his sentence, despite pleading guilty
to the offenses without a plea agreement, as he preserved his
appellate rights regarding “rais[ing] an objection that [the
District Court] varied upward from the sentencing guidelines
improperly or departed upwards from the sentencing
guidelines improperly[.]” (J.A. at 29.) He also appeals his
armed robbery conviction, arguing that the District Court erred
by accepting his guilty plea to armed robbery because he only
used a replica firearm. See infra Section II.B.




                               4
II.    DISCUSSION 2

       A.     Under the Guidelines, a Replica Firearm Is a
              “Dangerous Weapon” 3

       The sentencing guidelines provide for enhanced
sentences for both kidnapping and robbery if “a dangerous
weapon” was used.          Id. §§ 2A4.1(b)(3) (kidnapping);
2B3.1(b)(2)(D) (robbery). Chandler contests the dangerous-
weapon enhancements to both his robbery and kidnapping
charges, given that he did not use a real gun. “The government
bears the burden of proving by a preponderance of the evidence
that” the enhancements are applicable. United States v.
Napolitan, 

 (3d Cir. 2014).

       The relevant guidelines sections do not define
“dangerous weapon,” but the guidelines’ commentary defines
the term to include, besides “an instrument capable of inflicting
death or serious bodily injury,”


       an object that is not an instrument capable of
       inflicting death or serious bodily injury but []
       closely resembles such an instrument; or
       [something] the defendant used … in a manner

       2
        The District Court had jurisdiction under 

. We have jurisdiction pursuant to 

 and

(a).
       3
         We review the District Court’s interpretation of the
guidelines de novo. United States v. Grier, 

(3d Cir. 2007) (en banc).




                               5
       that created the impression that the object was
       such an instrument (e.g., a defendant wrapped a
       hand in a towel during a bank robbery to create
       the appearance of a gun).

U.S.S.G. § 1B1.1 cmt. n.1(E).

        That comment informed our jurisprudence for more
than 30 years, but reliance on it has recently become
questionable. 4 The Supreme Court held in Kisor v. Wilkie, 

 (2019), that courts must perform a “now-essential”
textual analysis before turning to the commentary. See United
States v. Adair, 

 (3d Cir. 2022) (citing
Kisor, 

). Unless the guideline’s text is
ambiguous and the comment provides clarity, the text alone
controls. United States v. Nasir, 

 (3d Cir.
2021) (en banc).

        Ambiguity is thus the key to permissible reliance on the
guideline’s commentary. The Supreme Court has instructed
that a regulation is “genuinely ambiguous … after a court has
resorted to all the standard tools of interpretation” and come up


       4
         Prior to 2022, we uniformly held that the use of fake
or simulated firearms in furtherance of crimes like robbery
qualified as use of a “dangerous weapon” subject to sentencing
enhancement under the guidelines. See, e.g., United States v.
Bell, 

 (3d Cir. 2020) (toy gun); United States v.
Hoffa, 

 (3d Cir. 2009) (hand in a pocket);
United States v. Orr, 

 (3d Cir. 2002)
(dismantled pellet gun); United States v. Dixon, 

 (3d Cir. 1992) (hand in a towel).




                                6
short. Kisor, 

. “[O]nly when that legal
toolkit is empty and the interpretive question still has no single
right answer can a judge … wave the ambiguity flag[.] … To
make that effort, a court must carefully consider the text,
structure, history, and purpose of a regulation, in all the ways
it would if it had no agency to fall back on.” 

(cleaned up).

       Even if genuine ambiguity does exist, we still will not
defer to the agency’s reading unless it is reasonable. 

 The
commentary must clarify the ambiguity rather than change the
meaning of the text. Kisor, 

. If the
commentary is reasonable, we afford it controlling deference;
but only if the agency’s interpretation: (1) was “one actually
made by the agency”; (2) “implicate[s the agency’s]
substantive expertise”; and (3) “reflect[s] fair and considered
judgment.” 

 (internal quotation marks omitted).

       Chandler says that the term “dangerous weapon” is not
genuinely ambiguous so we cannot rely on the commentary.
Alternatively, he argues that, even if the term is ambiguous, the
commentary is an unreasonable interpretation of the guideline.

       We are unpersuaded by his arguments for two reasons.
First, the term “dangerous weapon” is indeed genuinely
ambiguous, being subject to a range of views, so we can rely
on the Sentencing Commission’s commentary to reasonably
interpret the phrase to include replica firearms. Second, the
meaning of the words “dangerous weapon,” as understood
from Supreme Court precedent at the time the pertinent
guideline, § 1B1.1, was promulgated, encompassed replica
firearms, thus making the commentary reasonable and entitled




                                7
to controlling weight. We address each of those points more
fully in turn.

              1.     The Phrase “Dangerous Weapon” Is
                     Genuinely Ambiguous

       Federal courts have grappled for many years with the
scope of the term “dangerous weapon” in sentencing
decisions. 5 This debate continued with the recent split decision
of the Sixth Circuit in United States v. Tate, a case involving a
bank robber who placed his hand in a shoulder bag in a manner
that led the teller to believe that the robber was about to pull
out a gun. 

 (6th Cir. 2021). A majority of

       5
         Compare United States v. Mahler, 

 (4th
Cir. 1989) (holding that a displayed replica handgun used in a
bank robbery was not covered by the dangerous-weapon
enhancement), with United States v. Shores, 

 (11th Cir. 1992) (holding the opposite for merely
possessing – never brandishing – a toy gun in an attempted
bank robbery), and Hoffa, 

 (applying the
enhancement to a bank robber who placed his hand in his
pocket to create the appearance of a firearm); United States v.
Sturgis, 

 (4th Cir. 1995) (recognizing that
a bite from an HIV-positive prisoner could be an assault with
a “dangerous weapon”), with United States v. Taylor, 

 (2d Cir. 2020) (refusing to apply the enhancement to a
robber who placed his hand on his waistline to suggest a
firearm), and United States v. Rocha, 

 (9th
Cir. 2010) (holding that defendant did not use a dangerous
weapon when he used his bare hands to pull someone’s feet out
from under them, slamming them onto a concrete floor).




                               8
the panel affirmed the application of the dangerous-weapon
sentencing enhancement but one judge saw the matter
differently and concluded that “[n]o ordinary English speaker
… would say that a robber possesses a dangerous weapon
when the robber merely pretends to possess one.” Id. at 386
(Murphy, J., concurring) (emphasis in original). The majority
held, however, that “literal or dictionary definitions of words
will often fail to account for settled nuances or background
conventions that qualify the literal meaning … of legal
language.” Id. at 378 (majority opinion) (internal quotation
marks omitted). In the majority’s view, “creating the
impression of possessing a gun can trigger the sentencing
enhancement applicable to robberies.” Id. at 381.

        Our dissenting colleague in this case, like the
concurring judge in Tate, defines “dangerous weapon” as “one
by the use of which a fatal wound may probably or possibly be
given.” Dissent at 2 (quoting Dangerous Weapon, Black’s Law
Dictionary (5th ed. 1979); and citing Dangerous Weapon (def.
1), Ballentine’s Law Dictionary (3d ed. 1969), and Weapon
(def. 1), Webster’s Third New International Dictionary
(1966)). This ends the debate for him. And that definition of
course fits the words, but it is not the only meaning. Both the
Black’s and Ballentine’s definitions encompass a broader
definition of the words “dangerous weapon” than the dissent
suggests.

        The Black’s definition notes that “the manner of use
enters into the consideration as well as other circumstances[.]”
Dangerous Weapon, Black’s Law Dictionary (5th ed. 1979).
Similarly, later, the sixth edition of the same dictionary says
that “[w]hat constitutes a ‘dangerous weapon’ depends not on
nature of the object itself but on its capacity, given manner of




                               9
its use, to endanger life or inflict great bodily harm.”
Dangerous Weapon, Black’s Law Dictionary (6th ed. 1990);
see also Weapon, Black’s Law Dictionary (6th ed. 1990) (“An
instrument of offensive or defensive combat, or anything used,
or designed to be used, in destroying, defeating, threatening,
or injuring a person.” (emphases added)). As words change
meaning over time, not just when a new edition of Black’s Law
Dictionary is issued, a more contemporaneous edition, issued
three years after the guideline 6 rather than nine years earlier,
provides more persuasive evidence of how the guidelines’
drafters used the word at the time the dangerous-weapon
enhancement was adopted. Thus, in 1987, manner of use,
including the potential to create harm through threats, was
most likely part of the understanding of how to define a
“dangerous weapon.”

        More to the point, however, while providing guidance,
dictionaries do not create precedent. The Supreme Court does.
As the majority in Tate observed, 

, the
phrase “dangerous weapon” is a legal term of art that for
decades has embraced a functional meaning as reflected in the
Supreme Court’s opinion in McLaughlin v. United States, 

 (1986). This must be taken into account: “[W]e do
not woodenly interpret a legal text ‘in a vacuum,’ but instead
discern ‘the meaning of a statement’ in a law from the ‘context
in which it is made[.]’” Tate, 

 (internal citation
omitted) (first quoting Abramski v. United States, 

, 179 (2014); and then quoting United States v. Briggs, 

 (2020)).

       6
          The dangerous-weapon enhancement was present in
the first guidelines, promulgated on November 1, 1987.




                               10
       One year before the dangerous-weapon enhancement
appeared in the guidelines, the McLaughlin Court held that
something that looks like an operable gun – there, an unloaded
gun – qualifies as a “dangerous weapon” because it was
characteristically dangerous, can instill fear in the average
citizen, can be used as a bludgeon, and can invite a dangerous
and violent response. 7 

 (holding that each
reason is “independently sufficient [to] support [that]
conclusion”). Thus, the presence of even a replica of a gun
raises the temperature during a crime. It increases the
likelihood that a reasonable police response will include the
use of deadly force, which may also create “greater risk to the
physical security of victims, bystanders, and even the

       7
          Our dissenting colleague focuses on the Supreme
Court’s footnote in McLaughlin citing congressional floor
debate which “indicate[d] that Congress regarded incitement
of fear as sufficient to characterize an apparently dangerous
article … as ‘dangerous[.]’” 

 n.3. He argues that
legislative history is “irrelevant[,]” “cannot universally justify
stretching the meaning of dangerous weapon[,]” and “no
longer carries the weight that it once did.” Dissent at 5. But
we are not relying on legislative history. We are following
precedent set by the Supreme Court, which held that the fear
created by an unloaded gun is sufficient to qualify the weapon
as dangerous. Cf. Gilles v. Davis, 

 (3d. Cir.
2005) (following the Supreme Court’s counsel “to follow its
directly applicable precedent, even if that precedent appears
weakened by pronouncements in its subsequent decisions, and
to leave to the Court the prerogative of overruling its own
decisions.” (internal quotation marks and citations omitted)).




                               11
perpetrators.” United States v. Dixon, 

 (3d
Cir. 1992) (internal quotation marks omitted). It also increases
the risk of violent response by victims themselves, as well as
bystanders. Cf. Bruce v. Powell, No. 19-cv-13028, 

, at *4 (D.N.J. Nov. 29, 2021) (“The victim’s friend
… reached for his gun first, which provoked [the defendant] to
reach for his gun, which in turn caused the victim to reach for
his gun.”). When the guidelines enhancement was written, the
Sentencing Commission was, we can be sure, fully aware of
the Supreme Court’s analysis in McLaughlin and its legal
definition of “dangerous weapon.” 8 We are not free to ignore

       8
          It is also worth noting that both the 2014 and 2019
editions of Black’s Law Dictionary do more than “hint” that
“dangerous weapon” has a broader legal meaning: it directly
excerpts McLaughlin in the definition of dangerous weapon
itself. It defines “dangerous weapon” as:
       [a]n object or device that, because of the way it
       is used, is capable of causing serious bodily
       injury.
       “Three reasons, each independently sufficient,
       support the conclusion that an unloaded gun is a
       ‘dangerous weapon.’ First, a gun is an article
       that is typically and characteristically dangerous;
       the use for which it is manufactured and sold is a
       dangerous one, and the law reasonably may
       presume that such an article is always dangerous
       even though it may not be armed at a particular
       time or place. In addition, the display of a gun
       instills fear in the average citizen; as a
       consequence it creates an immediate danger that




                               12
the import of McLaughlin now. We are asked to interpret the
same term (in the same robbery context) that the Supreme
Court defined in McLaughlin.           Despite our dissenting
colleague’s view, we decline to bypass instruction from the
highest court of the land in favor of a fifty-year-old dictionary
definition. McLaughlin shapes our understanding of the
meaning of “dangerous weapon,” as it surely shaped the
understanding of the term had by the guidelines’ drafters.

        The Sentencing Commission first defined “dangerous
weapon” in the commentary as “an instrument capable of
inflicting death or serious bodily injury.” U.S.S.G. § 1B1.1
cmt. n.1(d) (1987). 9 Within a couple of years, for the

       a violent response will ensue. Finally, a gun can
       cause      harm     when       used       as    a
       bludgeon.” McLaughlin v. U.S., 

 [] (1986).
Weapon, Black’s Law Dictionary (11th ed. 2019); Weapon,
Black’s Law Dictionary (10th ed. 2014). Thus, it seems plain
to us that the term “dangerous weapon” encompasses two
different definitions of dangerous weapon: one of an
instrument inherently dangerous and one of an instrument or
article perceived as dangerous or that could cause danger,
while not dangerous per se.
       9
           Our dissenting colleague suggests that the
Commission rejected McLaughlin the year after its issuance.
However, for support he cites to informal agency guidance that
discusses “toy guns” (not replicas) and expressly disclaims that
it “does not necessarily represent the official position of the
Commission, should not be considered definitive, and is not




                               13
“purpose[] of … clarify[ing] the definition of a dangerous
weapon[,]” the Commission amended the commentary to say,
“[w]here an object that appeared to be a dangerous weapon was
brandished, displayed, or possessed, treat the object as a
dangerous weapon[.]” 

 app. C, amend. 71 (1989) (emphasis
added). That 1989 amendment, and later amendments, were
added to be “in accord[,]” 

 app. C, amend. 601 (2003), with
decisions like ours in United States v. Dixon, where we applied
the dangerous-weapon enhancement to a bank robber who
wrapped a towel over her hand and pointed menacingly at bank
employees. 

; see also United States v. Davis,

 (D.C. Cir. 2011) (“[T]he current version
of the Commentary [regarding the dangerous-weapon
enhancement] all but codifies the holding in United States v.
Dixon[.]”).

       Our decision in Dixon relied heavily on McLaughlin to
hold that the Commission intended to “equate[] the image of a
‘dangerous weapon’ with its reality for purposes of sentence
enhancement.” Dixon, 

. And rightly so; this
equation allows the guideline to conform to the reality of the
dangers created by things that appear to be guns. The
Commission’s 1989 Amendment can thus be seen as


binding upon the Commission, [or] the court[.]” U.S.
Sentencing Commission Answers Questions Most Frequently
Asked About the Sentencing Guidelines, 1 Fed. Sent’g Rep.
423, 423, 425-26 (1989). Additionally, the Commission
amended the commentary just six months after that guidance
to encompass items such as replica or toy guns, in accord with
McLaughlin. U.S.S.G. app. C, amend. 71 (1989) (amended
November 1, 1989).




                              14
purposefully clarifying the definition of “dangerous weapon”
rather than expanding it. 10 Cf. Kisor, 

 (the
presumption that the promulgating agency “is in the better
position [to] reconstruct its original meaning” “holds good for
a significant category of contemporaneous readings. Want to
know what a rule means? Ask its author.”) (internal quotation
marks and citation omitted); U.S.S.G. app. C, amend. 71
(1989) (promulgated two years after the guideline).

      Notably, every time we have analyzed this issue after
the 1989 Amendment we have held that instruments used
during the commission of a crime to give the appearance or
prompt the perception of a bona fide dangerous weapon were




       10
            Our dissenting colleague seems wary of the
commentary to the Amendment stating that it merely
“clarifies” the Guideline. Dissent at 4; U.S.S.G. § 1B1.1 app.
C, amend. 71 (1989). We think, however, that it is appropriate
to take the commentary’s assertion of the Amendment’s
purpose at face value because the Amendment indeed seems to
be just a clarification. While we have rejected application of a
later guideline’s commentary to earlier guidelines based on ex
post facto challenges, see United States v. Menon, 

 (3d Cir. 1994), that is not at issue here. Rather, outside of
that context, we have said, “[W]hen an amendment to a
guideline is intended to clarify the meaning of the existing
guideline, the court [should] give it substantial weight in
interpreting that guideline.” 

 (citations omitted and
emphasis added).




                               15
properly a basis for a “dangerous-weapon” enhancement. 11
See, e.g., supra n.4. The only cases Chandler cites for support
of his position were decided before the 1989 Amendment, and
still support application of the enhancement. United States v.
Dzielinski, 

 (7th Cir. 1990) (affirming the
district court’s application of the enhancement pre-1989
Amendment because it “relied upon the failure of the
[g]uidelines to adequately consider the threat posed by the
possession of what appeared to be a dangerous weapon”);
United States v. Mahler, 

 (4th Cir. 1989)
(similarly affirming the district court’s “plainly reasonable”
action of “treating the replica of a gun as an empty gun for the
purpose of increasing the sentencing level” pre-Amendment).
Chandler does not point to any post-1989 cases that have
maintained his position, and we could not find any. 12

       11
         Three cases decided this past year also came out the
same way. See generally United States v. Jefferson, No. 22-
1306, 

 (6th Cir. Mar. 7, 2023) (passing a
note while robbing a bank indicating that he had a gun and,
during one of the robberies, miming having a gun by putting
his hand in his pocket and making a gun-like hand motion);
United States v. Wen, No. 22-50207, 

 (9th
Cir. May 17, 2023) (possession of car, hammer, box cutter, and
rope in connection with offense constituted dangerous
weapons); United States v. Padilla, 

(D.N.M. 2023) (lifting his shirt to show cashier an object that
looked like a gun).
       

 F.3d at 75, decided a few years ago, is the
closest thing to support for Chandler’s position. The Second
Circuit held that a robber merely placing his hand “near his




                               16
        So, while nothing in the structure of § 1B1.1 of the
guidelines sheds light on the meaning of “dangerous weapon,”
the text, history, and purpose of the guideline suggest that the
term does not have just one, narrow meaning. That does not,
however, remove all question of the scope of the term, though
it helps. Because the term “dangerous weapon” retains
arguable ambiguity even after a consideration of “the text,
structure, history, and purpose of [the] regulation,” Kisor, 

, we now turn to the Commission’s commentary
and ask whether it is “reasonable” and “entitle[d] … to
controlling weight[,]” 

.

              2.     The    Commentary       Definition   is
                     Reasonable and Entitled to Controlling
                     Weight and, Thus, Deference

       The Commission’s commentary is a reasonable reading
of the guideline’s text, falling within “the outer bounds of
permissible interpretation.” 

. As examined above,
the text of § 1B1.1 is ambiguous because the term “dangerous
weapon” can be read literally to focus on inherently dangerous
weapons, such as firearms, and it can also be read to encompass
objects or actions that simulate an inherently dangerous


waistband, not inside it[,]” to suggest he had a firearm did not
constitute a “dangerous weapon” for enhancement purposes.
Id. But the court’s analysis strongly emphasized the fact that
the robber’s hand was “unconcealed” and could therefore “not
appear to be itself a weapon.” Id. And the court drew a
contrast with caselaw that uniformly held that a “concealed”
hand could “itself appear[] to be a dangerous weapon.” Id. at
76.




                              17
weapon and thereby create dangerous conditions or reactions.
The commentary’s gloss defines “dangerous weapon” to
include both of those plausible meanings, thus following
McLaughlin and being well within the bounds of reasonable
interpretation of “dangerous weapon” as a legal term of art, not
a mere “fiction[,]” as the dissent suggests. U.S.S.G. § 1B1.1
cmt. n.1(E) (2021); id. app. C, amend. 71 (1989); Dissent at 5.

        Moreover, under the longstanding prior-construction
canon, when “judicial interpretations have settled the
       13

meaning of an existing statutory provision, repetition of the
same language in a new statute indicates, as a general matter,
the intent to incorporate its … judicial interpretations as well.”
Merrill Lynch, Pierce, Fenner & Smith Inc. v. Dabit, 

 (2006) (rejecting a narrow interpretation of a statutory
term when the Supreme Court had previously adopted a broad
construction of the same term used in a different statute). 14


       13
          See Shapiro v. United States, 

 (1948)
(Vinson, C.J.) (“In adopting the language used in the earlier
act, Congress ‘must be considered to have adopted also the
construction given by this Court to such language, and made it
a part of the enactment.’” (quoting Hecht v. Malley, 

 (1924)).
       14
          See also Antonin Scalia & Bryan A. Garner, Reading
Law: The Interpretation of Legal Texts 324-25 (2012) (when
the Supreme Court authoritatively interprets a term as it is used
in a particular field of law, that term acquires a “technical legal
sense … that should be given effect in the construction of later-
enacted” laws and rules governing that field, even when based
on a single decision).




                                18
That canon applies when interpreting the guidelines as well.
The Sentencing Commission is “a body that straddles both the
legislative and judicial branches of the government,” Nasir, 

, and is steeped in sentencing policy and criminal
law jurisprudence, cf. United States v. Mercado, 

 (3d Cir. 2023) (finding the Sentencing Commission
“optimally positioned to opine” on criminal sentencing due to
its “expertise”). 15 It could “hardly have been unaware[,]”
Merrill Lynch, 

, of the landmark McLaughlin
decision when it employed the phrase “dangerous weapon” in
the guidelines. With that understanding, and because the
application note’s interpretation is consistent with the text,
history, and purpose of § 1B1.1, we conclude that the
Commission’s definition is reasonable and reflects its “fair and
considered judgment.” Kisor, 

. The
definition of “dangerous weapon” in § 1B1.1’s commentary
should therefore be afforded deference. See id. at 2415-18
(citing generally Auer v. Robbins, 

 (1997)).

              3.     Lenity Cannot Save Chandler’s Position

       Chandler and our dissenting colleague also contend that
the rule of lenity should resolve any ambiguity in Chandler’s
favor. Chandler’s argument relies on the assumption that his
replica firearm was not dangerous because it could not be used
as a bludgeon, a possibility noted in McLaughlin. This, of
course, ignores the force of the arguments just outlined
concerning the other threats created by a plausible facsimile of

       15
           Due to the commentary falling within the
Commission’s “substantive expertise[,]” we can give it
controlling weight. Kisor, 

.




                              19
a gun, including the increased risk that a violent response will
be triggered and the fear instilled in victims, both of which
were present here and which McLaughlin emphasized are
“each independently sufficient.” 

. But, even
taking Chandler’s argument on its own terms, it comes up
short. The record does not support the assumption that his
replica revolver could not be a bludgeon. 16 The record shows
only that he admitted that he “placed … [a] replica handgun”
on the back of a mail carrier and that, when he was arrested,
the police found the revolver in his waistband and later
confirmed it to be a replica. (J.A. at 30.) A replica is “a copy
exact in all details,” Replica, Merriam-Webster’s Collegiate
Dictionary (10th ed. 2002), or a “reasonably exact duplicate[,]”
Replica, Black’s Law Dictionary (11th ed. 2019). That
certainly implies a close similarity in heft. Thus, the fake gun
could have been used in a pistol whipping, which, all by itself,
makes his replica a dangerous weapon.

       As noted earlier, the government must prove
enhancements by a preponderance of the evidence. Napolitan,

. While there is, as we have discussed, legal
ambiguity in the phrase “dangerous weapon,” there is no such
ambiguity in Chandler’s conduct. It clearly falls within the
scope of the relevant application note, U.S.S.G. § 1B1.1 cmt.
n.1(E), which we have concluded can properly be considered.
See generally Nasir, 

.

       16
           The District Court adopted the facts from the
presentence report at sentencing, including Chandler’s use of a
replica firearm during his crimes. Neither the District Court,
nor any of the parties, discussed or placed any evidence in the
record as to the weight of the replica firearm.




                              20
       Our dissenting colleague disagrees, and argues that
when a guideline is ambiguous, before deferring to the
Sentencing Commission’s commentary, we must first apply
the rule of lenity. We disagree.

        Again, Kisor instructs courts to “exhaust all the
‘traditional tools’ of construction[,]” including “the text,
structure, history, and purpose of a regulation[.]” 

. Those interpretative tools do not typically imply resort
to judicial doctrines. See, e.g., Shular v. United States, 

 (2020) (Kavanaugh, J., concurring) (“[W]hen
‘a reviewing court employs all of the traditional tools of
construction, the court will almost always reach a conclusion
about the best interpretation,’ thereby resolving any perceived
ambiguity. That explains why the rule of lenity rarely comes
into play.” (quoting Kisor, 

 (Kavanaugh, J.,
concurring in judgment))). If they did, the rule of lenity – itself
a judicial doctrine – would seem to trump the traditional tools
of interpretation any time ambiguity arises, which would
render the tools irrelevant.

       Moreover, the next analytical step called for by Kisor
when a regulation is found to be genuinely ambiguous is an
inquiry into the agency’s interpretation of the regulation, not
an application of separate judicial doctrines. See 

. Similarly, nowhere in the Nasir majority opinion –
which constitutes our application of Kisor to the guidelines –
do we mention the rule of lenity. 17 

(discussing and applying Kisor).

       17
         Although we noted that the rule of lenity might have
a role to play in the United States v. Nasir concurrence, 17




                                21
       Nor does application of the rule of lenity make sense
here. First, “the rule of lenity only applies if, after seizing
everything from which it can be derived, we can make no more
than a guess as to what Congress intended.” United States v.
$734,578.82 in U.S. Currency, 

 (3d Cir.
2002) (emphasis added) (quoting Muscarello v. United States,

 (1998)). Here, we can look to the
commentary, which clarifies what is otherwise ambiguous,
obviating the need to consider the rule of lenity. If lenity
principles applied whenever the sentencing guidelines are held
to be ambiguous, then like the other Kisor interpretative tools,
commentary issued by the sentencing commission would be
made superfluous. 18 That surely could not have been


F.4th at 472-74 (Bibas, J., concurring), that view was not
adopted by the majority, and is not a binding command to let
lenity displace the Sentencing Commission’s guidance.
       18
          Our dissenting colleague says that the rule of lenity
only displaces commentary that is not favorable to the
defendant. That assumes two things that we do not.
       First, it assumes guidelines commentary can properly be
broader or harsher than the guidelines themselves. It cannot.
As discussed above and in Nasir, cf. 

, we do not
defer to commentary if it expands the guideline at issue. Here,
the commentary “purposefully clarif[ies] the definition of
‘dangerous weapon’ rather than expanding it.” Supra at page
12. Thus, deference is appropriate, and the rule of lenity has
no role.
       Second, accepting arguendo that the commentary
differs from the guidelines, our colleague’s position also
assumes that all judges see alike on whether commentary is




                              22
Congress’ intent. 19 Nor ours in Nasir. The Commission’s
writing of the commentary, and our examination of it as
needed, is not designed as an exercise in futility.

       The District Court correctly applied the dangerous-
weapon enhancement to Chandler because of his use of a
replica firearm.




“neutral or defendant friendly[.]” Dissent at 7 (cleaned up).
That is not so, as is evident here, since we see the commentary
as explaining, not expanding, the scope of the guideline.
        Our colleague also states that “lenity … must come
before deference” as a “traditional canon[] of statutory
construction[.]” Dissent at 7 (internal quotation marks
omitted). That undermines his statement that the rule of lenity
would only displace commentary which is unfriendly to the
defendant. Were we to apply the rule of lenity first, we would
never have occasion to defer to commentary, as we would have
already decided to apply the most defendant-friendly reading
of the guideline, regardless of what the commentary had to say.
      19
          Cf. United States v. Cooper, 

 (3d
Cir. 2005), as amended (Feb. 15, 2005) (“It is a well known
canon of statutory construction that courts should construe
statutory language to avoid interpretations that would render
any phrase superfluous.” (citing TRW Inc. v. Andrews, 

 (2001)).




                              23
       B.     The District Court Did Not Err in Accepting
              Chandler’s Guilty Plea

       For the first time on appeal, Chandler argues that the
District Court erred by accepting his guilty plea to armed
robbery. 20 Because he only used a replica firearm, he claims
he was effectively unarmed. Armed robbery, versus unarmed
robbery, increases the maximum sentence from ten years to
twenty-five years. 

(a). For the same reasons
that the District Court did not err in applying the dangerous-
weapon enhancement, we conclude that the Court did not
commit plain error in accepting the armed robbery guilty plea.
Chandler was indeed armed with a dangerous weapon when he
robbed the mail carriers.




       20
          We review unpreserved claims for plain error, which
is more deferential than de novo or clear error review. Fed. R.
Crim. P. 52(b); United States v. Olano, 

(1993). “There must be an ‘error’ that is ‘plain’ and that
‘affect[s] substantial rights.’ … Rule 52(b) leaves the decision
to correct the forfeited error within [our] sound discretion …
and [we] should not exercise that discretion unless the error
‘seriously affect[s] the fairness, integrity or public reputation
of judicial proceedings.’” Olano, 

 (first and last
alterations in original).




                               24
       C.     The District Court Properly Enhanced
              Chandler’s Sentence for Kidnapping a
              Government Employee in the Course of Her
              Duties 21

       The District Court enhanced Chandler’s kidnapping
sentence under U.S.S.G. § 3A1.2(a) and (b).               That
enhancement applies if the victim was a “government officer
or employee,” and “the offense of conviction was motivated by
such status[.]” Id. The Court found that Chandler’s motivation
to kidnap the mail carrier was based in part on her status as a
government employee. The parties agree that she was a
government employee; therefore, the only question is whether
the kidnapping was “motivated by such status.”              Id.
§ 3A1.2(a)(2).

       “Motive is a question of fact[,]” Monteiro v. City of
Elizabeth, 

 (3d Cir. 2006), and we review the
District Court’s factual findings for clear error. United States
v. Richards, 

 (3d Cir. 2012). The record with
respect to the District Court’s finding on Chandler’s motive is


       21
          We review the District Court’s interpretation of the
sentencing guidelines de novo, United States v. Adair, 

 (3d Cir. 2022), but we accept its findings of fact
“unless they are clearly erroneous and … give due deference
to the [D]istrict [C]ourt’s application of the guidelines to the
facts[,]” 

(e); United States v. Richards, 

, 219 n.2 (3d Cir. 2012); see also United States
v. Napier, 

 (3d Cir. 2001) (for sentencing
guidelines decisions, “[f]actual findings … are simply
reviewed for clear error”).




                              25
admittedly scant. The Court stated only that “[the] postal
workers … had their uniforms on and Mr. Chandler knew
exactly what he was doing.” (J.A. at 43-44.) But while the
judge did not give comprehensive remarks, he was not silent
either, and we do not require perfect explanations from
sentencing judges. Cf. DiFederico v. Rolm Co., 

 (3d Cir. 2000) (“[I]t is [our] responsibility … to accept the
ultimate factual determination of the [District Court] unless
that determination either (1) is completely devoid of minimum
evidentiary support displaying some hue of credibility, or (2)
bears no rational relationship to the supportive evidentiary
data.” (internal quotation marks omitted)).

        A defendant may be said to be motivated by a victim’s
status as a government employee when that status has
“influenc[ed] [the defendant’s] choice” to commit the crime.
Motivate (def. 1) & Motive (def. 1b), Webster’s Third New
International Dictionary (1986); see also Motive, Black’s Law
Dictionary (11th ed. 2019) (“Something, esp[ecially] willful
desire, that leads one to act.”). Chandler argues that knowledge
alone cannot trigger this enhancement. We agree. 22 See
United States v. Sulik, 

 (6th Cir. 2019).
Instead, a victim’s status as a government employee must have
been at least part of the reason why the defendant committed
the crime. Cf. Hassan v. City of New York, 

(3d Cir. 2015) (“Motive asks … why did he do it?”) (cleaned
up). That status need not, however, be the only reason for the

       22
         Some degree of knowledge must be present, however.
Section 3A1.2(a) implicitly includes a knowledge requirement
because, as common sense dictates, to be motivated by a
victim’s status, the defendant must know it.




                               26
crime; multiple motives can coexist. See Motivate (def. 1),
Webster’s Third New International Dictionary (1986) (giving
examples of “factors that [motivate] people”); United States v.
Feeback, 

 (8th Cir. 2022) (applying the
“official victim” enhancement when a defendant made illegal
threats to Veterans Affairs employees regarding his benefits
because “[i]t ma[de] no difference that money may have [also]
played a role”).

      The record here amply supports that the robberies were
motivated by the victims’ status as postal workers. Chandler
donned a sweatshirt emblazoned with “USPS,” short for
United States Postal Service, and robbed the two USPS
employees, in the same area, separated by just a few weeks. 23
The question is whether that motivation can be imputed to the
kidnapping of his second victim. 24

       23
           Chandler asserts that his sole motivation for the
robberies was “to obtain packages from a delivery person,
regardless of employment status” (Opening Br. at 14, 20), not
“to steal packages carried by the Postal Service.” (Answering
Br. at 12). He insists that his motivation was purely economic
and that had he “taken packages from a UPS, FedEx, or
Amazon driver, his motive would have been the same.” (Reply
Br. at 6.) But Chandler did not rob a private package delivery
contractor. He robbed two USPS workers while wearing a
USPS sweatshirt. There is no evidence in the record that he
robbed or attempted to rob any other package delivery workers.

        For purposes of analysis, we look only to the second
       24

robbery, which occurred contemporaneously with the
kidnapping of the mail carrier. The government urges us to




                              27
       The Guideline’s enhancement states that the “offense of
conviction” must have been motivated by the government
employee status. U.S.S.G. § 3A1.2(a). The government asks
us to read “offense of conviction” broadly and to consider all
relevant conduct regarding both the robberies and the
kidnapping. 25 Chandler says we should focus on the language
of the indictment pertaining to the kidnapping count and its
elements, and not “mix and match” counts to apply the
enhancement. (Opening Br. at 19.) He relies on United States
v. Boney, but that case addresses the choosing of an applicable
guideline. 

 (3d Cir. 2014). It does not stand
for the proposition that a court cannot consider all related
conduct when deciding whether to apply an enhancement. 

       While the “‘offense of conviction’ includes only the
substantive crime for which a particular defendant was
convicted” and “the facts undergirding” it, United States v.
Pressler, 

, 157-58 n.7 (3d Cir. 2001), our decision
in United States v. Murillo makes clear that the “conduct of the


consider Chandler’s first armed mail theft, which occurred a
month prior, as “relevant conduct” to determine his motive for
kidnapping the mailwoman. (Answering Br. at 16 (“[T]he
[D]istrict [C]ourt’s factual determinations may be based on the
record in its entirety[.]”).) We need not consider that earlier
incident because the circumstances of the second robbery are
sufficient to demonstrate his motivation for the kidnapping.
       25
          The enhancement was not added to the robbery
convictions since the robbery-of-a-postal-employee statute
already takes the victim’s status into account. 

(a) (robbing a postal service worker).




                              28
offense of conviction” includes “all conduct in furtherance of
the offense of conviction[,]” 

 (3d Cir. 1991).
Though Murillo addressed sentencing enhancements for a
defendant’s role in the offense, its reasoning applies equally to
the question of motive. Often, “a court simply cannot
determine [motive] by looking at only the specific elements of
the offense of conviction. [Motive] is a concept that requires a
court to consult the events leading up to the offense of
conviction, not just a snapshot of events at the instant the
offense occurred.” 

. So, while we focus on the
motive for the kidnapping, we also consider the robbery that
immediately preceded and prompted it. The two “were
inextricably intertwined[.]” (Answering Br. at 18.) But for the
robbery, there is no reason to believe that the kidnapping would
have occurred at all.

        Chandler also invokes United States v. Cherry, where
we held that a court cannot “look through” the enhanced
conviction to an underlying offense to find an official victim.

 (3d Cir. 1993) (rejecting applying the
enhancement based on a flight from prosecution offense, which
lacks an official victim, or any victim at all). That, however,
says nothing about our case, in which the District Court looked
at the facts attendant to the kidnapping to determine Chandler’s
motive for the crime. That was appropriate, since motive can
almost never be discerned without examining circumstantial
evidence. Watson v. Rozum, 

 (3d Cir. 2016)
(“Because motivation is almost never subject to proof by direct
evidence, [one] must rely on circumstantial evidence to prove
… motive.”).

      The District Court thus properly applied the official
victim enhancement to the kidnapping offense. The Court




                               29
stressed that Chandler had chosen to target a mail carrier in
uniform. And it adopted the Presentence Report’s finding that
the mailwoman’s government-employee status had motivated
his kidnapping. Cf. United States v. Collado, 

 (3d Cir. 1992) (when “the district court makes no
independent findings of fact in relation to sentencing issues,
but instead adopts the reasons set forth by the probation officer
in the presentence investigation report, we view the report as
containing the only findings of fact that support the court’s
sentencing decision”). That is a sufficient basis for us to
discern the Court’s finding of fact, although it would of course
be helpful for district courts to be more explicit.

       Chandler robbed the mail carrier to steal USPS
packages from her, and he kidnapped her to get away with
those packages. The two crimes were tightly linked. By aiding
Chandler’s escape, the kidnapping served to facilitate the
robbery itself, and the motivations for the two crimes cannot
be unwound.

       Most of our sister courts have read this enhancement the
same way: it applies when a defendant not only knew the
victim was a government employee but the motivation for the
criminal action also arose from the duties of the government
employee victim, as here. 26 Absent the mail carrier’s status as

       26
         United States v. Feeback, 

 (8th
Cir. 2022) (official victim enhancement applied to defendant
who made violent threats against Veterans Affairs employees
over the non-payment of benefits he thought owed to him);
United States v. Ball, 

 (4th Cir. 2021) (same
for defendant who fatally shot a police officer after his texts to




                               30
a postal service employee, she (and, thus, Chandler) would
have lacked access to the USPS mailvan and the packages
inside of it. The guideline provision is “designed to protect
government officers in the performance of their official


his girlfriend indicated that he would “kill any police officer
attempting to take him into custody”); United States v. Dávila-
Bonilla, 

 (1st Cir. 2020) (same for defendant
who illegally intimidated and threatened his probation officer
after being asked to take a random drug test); United States v.
Sulik, 

 (6th Cir. 2019) (same for
defendant who sent threatening emails to a member of
Congress because he was upset with the content of the
Representative’s statements); United States v. Watts, 

 (7th Cir. 2015) (same for defendant who, after losing
his excessive force civil rights case against a police officer,
threw a forty-four pound chair at the defendant immediately
after the verdict was read; the police officer “aroused the
defendant’s wrath by the exercise of police authority over
him”); United States v. Polk, 

 (5th Cir.
1997) (same for defendant who was convicted of plotting to
blow up an Internal Revenue Service building because, even
though no one was killed and the defendant did not know the
names of his intended victims, the defendant intended to “kill,
injure, or maim federal employees in the [building] solely
because those persons worked for the IRS”); United States v.
Rue, 

 (10th Cir. 1993) (same for prisoner
who attacked a prison guard attempting to take away his
“homemade hypodermic syringe”); United States v. Sanchez,

 (9th Cir. 1990) (same for a defendant
who rammed a border patrol agent with his car and physically
assaulted the agent after later being confronted).




                              31
duties[,]” United States v. Watts, 

 (7th Cir.
2015), and this is such a scenario. Thus, the court did not
clearly err in finding motive based on the status of the USPS
worker whom Chandler robbed and kidnapped.

III.   CONCLUSION

        Criminal defendants have long been on notice that using
an imitation of a dangerous weapon to achieve criminal ends is
the equivalent in the eyes of the law of using the real thing.
The fear instilled in victims is just as real and there remains a
real risk of violence. The District Court did not err either in
treating Chandler’s replica handgun as a dangerous weapon or
in accepting his guilty plea to armed postal robbery. Nor did
the Court err when it found that Chandler was motivated to
kidnap the mail carrier because she was a government
employee. We will therefore affirm the District Court’s
judgment in all respects.




                               32
United States v. Chandler, No. 22-1786

BIBAS, Circuit Judge, dissenting.
   A dangerous weapon must be both dangerous and a weapon.
Fake guns are neither. Today, the majority holds otherwise.
    James Chandler used a fake gun to kidnap and rob mail car-
riers. At sentencing, the District Court enhanced his sentence
twice for (1) using a “dangerous weapon” and (2) targeting a
government employee.
   To affirm the first enhancement, my colleagues water down
the ordinary meaning of “dangerous weapon” by including
non-weapons that only seem dangerous. Yet text, history, and
precedent are to the contrary. And even if the phrase were
ambiguous, we should turn not to the Guidelines commentary
but to the rule of lenity.
    To affirm the second one, my colleagues supply what the
District Court should have said. True, the District Court found
that Chandler knew the mailwoman was a government employee.
But it did not make the necessary link: that he had robbed and
kidnapped her because she worked for the government. So I
respectfully dissent on both sentencing issues.
   I.     UNDER THE GUIDELINES, A FAKE GUN IS NOT
              A “DANGEROUS WEAPON”

   A. The enhancement’s text requires a weapon that can
      physically harm someone
   We should start and end with the text. Both the kidnapping
and robbery Guidelines enhance sentences “[i]f a dangerous
weapon was used.” U.S.S.G. § 2A4.1(b)(3) (kidnapping);
accord § 2B3.1(b)(2)(D) (robbery). Yet neither those Guide-
lines nor any others define “dangerous weapon,” so we must
dig into the phrase’s ordinary meaning.
   A “weapon” is “something (as a club, sword, gun, or gre-
nade) used in destroying, defeating, or physically injuring an
enemy.” Weapon (def. 1), Webster’s Third New International
Dictionary (1966). And a “dangerous weapon” must be “dan-
gerous to life; one by the use of which a fatal wound may prob-
ably or possibly be given.” Dangerous Weapon, Black’s Law
Dictionary (5th ed. 1979); accord Dangerous Weapon (def. 1),
Ballentine’s Law Dictionary (3d ed. 1969) (“An instrument
which, when used in the ordinary manner contemplated by its
design and construction, will, or is likely to, cause death or
great bodily harm.”). So a dangerous weapon must, at a mini-
mum, be able to kill or gravely harm another person.
   That clear definition should resolve this case. Yet my col-
leagues do not mention it until several pages into their discus-
sion. There, they admit “that definition of course fits.” Maj.
Op. 9. And later, they concede that mine is the “literal[ ]”
“read[ing].” Id. at 17. But they call the plain meaning “narrow”
and claim that “it is not the only meaning” of the term. Id. at 9,
17. So what does “dangerous weapon” mean? My colleagues
never quite say. Nor do they offer examples of ordinary English
usage that broaden “dangerous” to “apparently dangerous.”
    Instead, they suggest that the phrase has a broader meaning
as a “legal term of art.” Id. at 10. But their evidence does not
support that bold claim. The post-1989 cases that they cite




                                2
reflexively deferred to the commentary. Id. at 6 n.4, 15–16 &
n.11. Those cases do not parse the phrase itself.
    Plus, this legal-term-of-art argument falls flat. Surveying
state laws and cases from around the Guidelines’ enactment,
Judge Eric Murphy found that “dangerous weapon” lacked any
“well-established legal meaning that could cover pretending to
possess a dangerous weapon.” United States v. Tate, 

 (6th Cir. 2021) (Murphy, J., concurring in the judg-
ment). On the contrary, many state courts at the time rejected
my colleagues’ reading. State legislatures thus had to amend
their armed-robbery statutes to reach fake guns. 

 at 390–91.
The Sentencing Commission could have done the same but
chose not to. So there is no basis to stretch the text.
   B. History shows that the Commission did not codify
      McLaughlin
    My colleagues claim that after the Supreme Court in
McLaughlin defined “dangerous weapon” to include some-
thing that just looked like a dangerous weapon, the Sentencing
Commission adopted that understanding. Maj. Op. 10–13 (cit-
ing McLaughlin v. United States, 

 (1986)). But
history shows just the opposite.
    When the Commission first defined “dangerous weapon”
the year after McLaughlin, it gave the phrase its ordinary mean-
ing: “an instrument capable of inflicting death or serious bodily
injury.” U.S.S.G. § 1B1.1 cmt. n.1(d) (1987). As the Commis-
sion explained, this definition excluded toy guns. U.S. Sent’g
Comm’n Answers Questions Most Frequently Asked About the
Sent’g Guidelines, 1 Fed. Sent’g Rep. 423, 425–26 (1989)




                               3
(Question 36). So at first, the Commission rejected McLaugh-
lin’s definition.
   Not until two years later did the Commission amend the
commentary: “Where an object that appeared to be a danger-
ous weapon was brandished, displayed, or possessed, treat the
object as a dangerous weapon.” U.S.S.G. § 1B1.1 cmt. n.1(d)
(1989) (emphases added).
    The Commission claimed its comment was just “clar-
ify[ing] the definition of dangerous weapon.” U.S.S.G. app. C,
amend. 71 (1989). My colleagues take this claim at face value.
Maj. Op. 14. But calling a Guideline amendment a “clarifica-
tion” does not make it so. United States v. Menon, 

 (3d Cir. 1994) (rejecting a “clarifying amendment” that
punished a defendant “more harshly” than the original Guide-
line). Rather, we look to what the Commission did, not just to
what it said it did. United States v. Nasir, 

 (3d
Cir. 2021) (en banc) (Bibas, J., concurring).
    By amending the commentary, the Commission impermis-
sibly expanded the Guidelines’ scope. See Tate, 

 (Murphy, J., concurring in the judgment). The original
comment followed the plain meaning. But this new comment
stretched the text to also reach objects that looked like danger-
ous weapons. In a later amendment, the Commission admitted
doing so. U.S.S.G. app. C, amend. 601 (Supp. 2000) (specify-
ing when “an object that is not an actual, dangerous weapon
should be treated as one” (emphases added)). And that is how
we understood this change when we recognized that the Guide-
lines commentary “equates the image of a ‘dangerous weapon’
with its reality.” United States v. Dixon, 

, 121–24




                                4
(3d Cir. 1992) (grounding its holding in pre-Kisor deference).
In short, as even the Guidelines’ author admits, the commen-
tary sets forth a legal fiction, not a fair reading of the text.
   C. Precedent confirms the text’s plain meaning
   Without text or history on their side, my colleagues rest on
a single, inapt precedent—McLaughlin. McLaughlin read a
bank-robbery statute to encompass unloaded guns in part because
a gun is a weapon and in part because it is dangerous when
used as a bludgeon. 476 U.S. at 17–18. But unlike an empty
gun, a fake one can never fire bullets, so it is not “typically and
characteristically dangerous.” Id. at 17. And the government
never bore its burden to prove that Chandler’s replica was
heavy enough to club someone. Contra Maj. Op. 20 & n.16
(putting the burden to disprove the gun’s weight on the crimi-
nal defendant).
    The majority seizes on McLaughlin’s other rationale: that
an unloaded gun is a “dangerous weapon” because guns pro-
voke fear and violence. 476 U.S. at 17–18. Yet McLaughlin
never tried to ground that claim in the text. Instead, it cobbled
together that functional meaning from the bank-robbery stat-
ute’s legislative history. Id. at 18 n.3. But this rationale shows
that McLaughlin did not lay down a sweeping rule, but just
construed a specific statute. After all, the specific legislative
history of an unrelated statute is irrelevant to these Guidelines’
meaning. It cannot universally justify stretching the meaning
of “dangerous weapon” to reach fake guns. Plus, legislative
history no longer carries the weight that it once did. And it can-
not muddy clear text. Milner v. Dep’t of Navy, 

,
572 (2011). So I would cabin McLaughlin to the bank-robbery




                                5
statute and decline to extend its atextual approach to these
Guidelines.
    The more instructive precedent deals with the enhance-
ment’s text before the 1989 commentary (upheld by bare pre-
Kisor deference) distorted it. Those cases confirm the text’s
plain meaning: As two of our sister circuits understood, the
enhancement “did not provide for an increase in the offense
level based upon the apparent possession of a dangerous
weapon.” United States v. Dzielinski, 

 (7th
Cir. 1990); accord United States v. Mahler, 

, 76–
77 (4th Cir. 1989) (“[T]he Sentencing Commission did not
contemplate the use of a handgun replica in a robbery.”). Both
courts recognized that the proper response is not to stretch the
enhancement’s text, but to depart upward when warranted. See

 (both sources). So the relevant precedent confirms the text’s
plain meaning, refutes my colleagues’ broad meaning, and dis-
pels their claim of ambiguity. The only ambiguity here comes
from the commentary and improper deference to it.
   D. If there were any ambiguity, the rule of lenity would
      trump Kisor deference
   Even if the Guidelines were ambiguous, the key to the case
would be lenity, not deference. We get to Kisor’s second step
only when the regulation is ambiguous. But before deferring, we
must first “empty” the “legal toolkit.” Kisor, 

.
As I have explained elsewhere, “[a] key tool in that judicial
toolkit is the rule of lenity.” Nasir, 

 (Bibas, J.,
concurring). And “[t]here is no compelling reason to defer to a
Guidelines comment that is harsher than the text.” 

 So when




                                6
we read ambiguous Guidelines, the tie should go to liberty.
Here, though, the majority gives the tie to the agency.
    Doubling down, my colleagues argue that applying the rule
of lenity here does not “make sense.” Maj. Op. 22. Yet they
understate lenity’s benefits and overstate its risks. “[T]he rule
of lenity serves our nation’s strong preference for liberty.”
Nasir, 

 (Bibas, J., concurring). And when used
to interpret ambiguous Guidelines, it checks over-punishment,
ensuring that “criminal punishment … represents the moral
condemnation of the community.” 

 (alteration in orig-
inal) (quoting United States v. Bass, 

 (1971)).
That benefit is no less present here than elsewhere.
    My colleagues fear that lenity would make all “commen-
tary … superfluous.” Maj. Op. 22. Not so. The rule applies
only when the commentary “has a clear tilt toward harshness.”
Nasir, 

 (Bibas, J., concurring). So it never dis-
places neutral or defendant-friendly commentary.
    They also worry that the rule of lenity “would seem to
trump the traditional tools of interpretation” and “render
[them] irrelevant.” Maj. Op. 21. That fear is baseless. The rule
allows for all other “traditional canons of statutory construc-
tion” to be applied first. Shular v. United States, 

 (2020) (Kavanaugh, J., concurring) (internal quotation
marks omitted). But deference to the Sentencing Commission
is not a traditional interpretative tool. Lenity is. So it must
come before deference.
   The majority mistakenly finds the text ambiguous, then
chooses the harsher reading—the very thing barred by the rule




                               7
of lenity. But our Court has long favored liberty over commen-
tary that “punish[es] the defendant more harshly than he would
have been” under our own “independent interpretation” of the
Guidelines. Menon, 

. If these Guidelines were
ambiguous, I would follow our preference for liberty and apply
lenity to foreclose the harsher reading.
II.    THE DISTRICT COURT NEVER FOUND THAT CHAN-
       DLER KIDNAPPED AND ROBBED THE MAILWOMAN
       BECAUSE SHE WAS A GOVERNMENT EMPLOYEE
    My colleagues also go astray on the other sentencing issue.
Though they read the government-employee enhancement cor-
rectly, they misapply it. As they note, it is not enough for a
defendant to know that the victim is a government employee.
That knowledge must also be part of the reason why the defend-
ant committed the crime or targeted the victim. Maj. Op. 26.
   But the District Court said nothing about Chandler’s motive.
As my colleagues concede, “[t]he record with respect to the
District Court’s finding on Chandler’s motive is admittedly
scant.” 

 at 25–26. Even that is an overstatement—the court
made no such finding. All it said was that the “postal workers
… had their uniforms on and Mr. Chandler knew exactly what
he was doing.” Id. at 26 (alteration in original) (quoting JA 43–
44). That line suggests that it thought the enhancement requires
only knowledge, not motive. So it never found that the mail-
woman’s government employment motivated Chandler.
    Rather than affirm this enhancement, we should remand for
resentencing. The District Court might well choose to canvass
the record evidence as my colleagues do, reading it as proof of
motive. Id. at 27. If it had done so, I would have voted to affirm




                                8
this enhancement. But it did not, and we should not speak on
its behalf.
                           *****
    Text is the touchstone of interpretation. Yet my colleagues
skirt the plain meaning of “dangerous weapon,” misread prec-
edent, find nonexistent ambiguity, and even then eschew len-
ity. And they justify the government-employee enhancement
by reading speech into the District Court’s silence. I respect-
fully dissent.




                              9