PRECEDENTIAL


     UNITED STATES COURT OF APPEALS
          FOR THE THIRD CIRCUIT
               _____________

                    No. 14-4476
                   _____________

              CARLTON BAPTISTE,
               a/k/a Carlton Baptist,

                                  Petitioner

                          v.

   ATTORNEY GENERAL UNITED STATES OF
               AMERICA,

                                Respondent
                   _____________

 On Petition for Review of a Decision of the Board of
                 Immigration Appeals
   (Immigration Judge: Margaret R. Reichenberg)
                   (A030-338-600)

               Argued: April 5, 2016
                  _____________

Before: GREENAWAY, JR., SCIRICA and RENDELL,
               Circuit Judges.
                (Filed: November 8, 2016)


Michael L. Foreman, Esq.
Penelope A. Scudder [ARGUED]
Pennsylvania State University
Dickinson School of Law
329 Innovation Boulevard
Suite 118
State College, PA 16802

      Attorneys for Petitioner



Jennifer J. Keeney, Esq.
Jesse M. Bless, Esq. [ARGUED]
Anthony C. Payne, Esq.
Colette J. Winston, Esq.
United States Department of Justice
Office of Immigration Litigation
P.O. Box 878, Ben Franklin Station
Washington, D.C. 20044

      Attorneys for Respondent
                   _________________

               OPINION OF THE COURT
                 __________________

GREENAWAY, JR., Circuit Judge.




                            2
       Carlton Baptiste petitions for review of a decision of
the Board of Immigration Appeals (“BIA”) ordering his
removal as an alien convicted of: (1) an “aggravated
felony” pursuant to 8 U.S.C. § 1227(a)(2)(A)(iii), which is
defined as, inter alia, a “crime of violence,” 18 U.S.C. § 16;
and (2) two or more crimes involving moral turpitude
(“CIMTs”) pursuant to 8 U.S.C. § 1227(a)(2)(A)(ii).

        Baptiste’s petition requires us to decide whether the
definition of a “crime of violence” provided in 18 U.S.C. §
16(b) is void for vagueness under the Due Process Clause of
the Fifth Amendment. Section 16(b) and similarly worded
statutes have come under attack in federal courts across the
country after the Supreme Court’s decision in Johnson v.
United States, 

(2015), which invalidated the
so-called “residual clause” of the Armed Career Criminal
Act (“ACCA”), 18 U.S.C. § 924(e)(2)(B)(ii), as
unconstitutionally vague.

       Although we initially conclude that Baptiste’s New
Jersey second-degree aggravated assault conviction was for
a crime of violence pursuant to § 16(b), we are persuaded
that the definition of a crime of violence in § 16(b) is
unconstitutionally vague after Johnson. We therefore
invalidate § 16(b) and hold that Baptiste was not convicted
of an aggravated felony. However, we conclude that
Baptiste is nonetheless removable because he was convicted
of two or more CIMTs.

       Accordingly, we will grant the petition in part as it
relates to the BIA’s aggravated felony determination, deny
the petition in part as it relates to the BIA’s CIMT
determination, and remand the case to the BIA for further
proceedings so that Baptiste may apply for any relief from




                              3
removal that was previously unavailable to him as an alien
convicted of an aggravated felony.

I.   FACTUAL        BACKGROUND         AND     PROCEDURAL
HISTORY

A.     Factual Background

       Petitioner Carlton Baptiste is a native of Trinidad and
Tobago who was admitted to the United States as a lawful
permanent resident in 1972. On December 15, 1978,
Baptiste was convicted of atrocious assault and battery
pursuant to former N.J. Stat. Ann. § 2A:90-1 (West 1969)
(the “1978 Conviction”). There is no indication from the
administrative record as to the facts underlying this
conviction. Baptiste was sentenced to a suspended twelve-
month term of imprisonment and placed on probation for
one year.

       Over thirty years later, on April 8, 2009, Baptiste
was convicted of second-degree aggravated assault pursuant
to N.J. Stat. Ann. § 2C:12-1b(1) (West 2005) (the “2009
Conviction”). 1 That statute provides that “[a] person is
guilty of aggravated assault if he . . . [a]ttempts to cause
serious bodily injury to another, or causes such injury
purposely or knowingly or under circumstances manifesting
extreme indifference to the value of human life recklessly
causes such injury.” N.J. Stat. Ann. § 2C:12-1b(1) (West
2005). As with his earlier conviction, there is no indication

       1
         We use the term “second-degree aggravated
assault” throughout this opinion to refer to the crime
defined at N.J. Stat. Ann. § 2C:12-1b(1) (West 2005).




                              4
from the administrative record as to the facts underlying
Baptiste’s 2009 Conviction. There is also no indication
from the administrative record as to whether Baptiste
pleaded guilty to the attempt crime in the statute, or, if he
pleaded guilty to the completed crime, to which mental state
in the statute Baptiste pleaded guilty to possessing—
purpose, knowledge or recklessness. See A.R. 334. He was
sentenced to a five-year term of imprisonment.

B.     Procedural History

       In June 2013, the Department of Homeland Security
(“DHS”) instituted removal proceedings against Baptiste.
DHS asserted that, based on his 2009 Conviction, Baptiste
was removable as an alien convicted of a crime of violence
pursuant to 18 U.S.C. § 16 and, therefore, an aggravated
felony pursuant to 8 U.S.C. § 1227(a)(2)(A)(iii). DHS later
asserted that Baptiste was also removable, based on both his
1978 Conviction and his 2009 Conviction, as an alien
convicted of “two or more crimes involving moral turpitude,
not arising out of a single scheme of criminal misconduct”
pursuant to 8 U.S.C. § 1227(a)(2)(A)(ii). On October 8,
2013, the Immigration Judge (“IJ”) sustained both charges
of removability. Baptiste appealed the IJ’s determinations
to the BIA.

        The BIA agreed with the IJ’s determination that the
2009 Conviction was for a crime of violence. It reasoned
that, in order to qualify as a crime of violence under § 16(b),
“the nature of [a] crime . . . must be such that its
commission ordinarily would present a risk that physical
force would be used against the person . . . of another,
irrespective of whether the risk develops or harm actually
occurs.” A.R. 4. Accordingly, the BIA determined that




                              5
“the relevant question . . . is whether the offense (whatever
its mens rea may be) is one that inherently involves a
person acting in conscious disregard of the risk that, in the
course of its commission, he may ‘use’ physical force
against the person of another.” A.R. 4. Under these
principles, the BIA concluded that:

      [A]n individual who undertakes to cause
      serious bodily injury to another under
      circumstances       manifesting         extreme
      indifference to human life necessarily
      disregards the substantial risk that in the
      course of committing that offense he will use
      physical force against another, either to effect
      the serious bodily injury that the statute
      requires or to overcome the victim’s
      resistance or both.

A.R. 4−5.

       The BIA also agreed with the IJ’s determination that
the 2009 Conviction was for a CIMT. 2 It examined the
manner in which New Jersey courts have construed the
recklessness crime in Baptiste’s statute of conviction and
observed that:

      New Jersey courts hold that an individual acts
      under circumstances manifesting an extreme
      indifference to the value of human life if he acts

      2
        Baptiste did not contest before the BIA, and does
not contest in his petition for review before this Court, the
IJ’s conclusion that his 1978 Conviction was for a CIMT.




                             6
       with conscious awareness of the fact that his
       conduct bears a substantial risk that he will kill
       another and he conducts himself with no regard
       to that risk.

A.R. 5. Based on that observation, the BIA concluded that
“an individual cannot form the culpable mental state and
commit the culpable acts required for conviction . . .
without acting in a base, vile or depraved manner and
without consciously disregarding a substantial risk that he
will kill another.” A.R. 6.

       Accordingly, the BIA dismissed Baptiste’s appeal.
Baptiste filed a timely petition for review with this Court on
November 14, 2014.

II.    JURISDICTION AND STANDARD OF REVIEW

        The BIA had appellate jurisdiction over the IJ’s order
of removal pursuant to 8 C.F.R. § 1003.1(b)(3). We have
jurisdiction over Baptiste’s petition for review of the BIA’s
dismissal of his appeal pursuant to 8 U.S.C. § 1252(a)(1).

       “Where, as here, the BIA issues a written decision on
the merits, we review its decision and not the decision of the
IJ.” Bautista v. Att’y Gen. of the U.S., 

, 57 (3d
Cir. 2014). Because an assessment of whether a crime
constitutes a crime of violence pursuant to 18 U.S.C. §
16(b) implicates the criminal provisions of the U.S. Code,
we exercise de novo review over the BIA’s determination
that the 2009 Conviction was for a crime of violence and,
therefore, an aggravated felony. Aguilar v. Att’y Gen. of the
U.S., 

, 695 (3d Cir. 2011). Similarly, we
review Baptiste’s due process challenge to the definition of




                              7
a crime of violence in § 16(b) de novo. Abdulrahman v.
Ashcroft, 

, 595−96 (3d Cir. 2003).

       Since the BIA’s determination that the 2009
Conviction was for a CIMT was made in an unpublished,
non-precedential decision issued by a single BIA member,
we do not accord that determination any deference, and it is
“[a]t most . . . persuasive authority.” Mahn v. Att’y Gen. of
the U.S., 

, 173 (3d Cir. 2014). We therefore
review the BIA’s CIMT determination de novo as well.

III.   ANALYSIS

A.     Baptiste’s 2009 Conviction was for a “crime of
violence” under § 16(b)

         An alien who is convicted of an “aggravated felony”
after his admission to the United States is removable
pursuant to 8 U.S.C. § 1227(a)(2)(A)(iii). The term
“aggravated felony” is defined as, inter alia, a “crime of
violence (as defined in [18 U.S.C. § 16], but not including a
purely political offense) for which the term of imprisonment
[is] at least one year.”3 8 U.S.C. § 1101(a)(43)(F). Thus, in
order to determine whether Baptiste’s 2009 Conviction was
for an aggravated felony, we must first examine the
definition of a “crime of violence” in 18 U.S.C. § 16.

. After having “ascertain[ed] the

       3
         Baptiste does not dispute that his 2009 Conviction
was for a crime for which the term of imprisonment is at
least one year.




                             8
definition of a ‘crime of violence,’” we must then compare
that definition to the statute of conviction to determine
whether the applicable crime defined in the statute of
conviction is categorically a crime of violence—an inquiry
known as the “categorical approach.” 

of a “crime of violence”

       A “crime of violence” is defined, in relevant part, as
an offense “that is a felony and that, by its nature, involves a
substantial risk that physical force against the person or
property of another may be used in the course of
committing the offense.” 18 U.S.C. § 16(b) (emphasis
added). 4 That definition requires “specific intent to use
force” or, in other words, “the intentional employment of . .
. force, generally to obtain some end.” Tran v. Gonzales,

, 470−71 (3d Cir. 2005); see Leocal v.
Ashcroft, 

, 9 (2004) (“‘[U]se’ requires active
employment.” (emphasis added)).            Thus, a crime of
violence under § 16(b) is one that involves a substantial risk
that force will be “actively employ[ed]” “in the furtherance
of the offense.” 

.

      Within this framework, we have distinguished
between those types of recklessness crimes that may be

       4
          Section 16(a) alternatively defines a “crime of
violence” as “an offense that has as an element the use,
attempted use, or threatened use of physical force against
the person or property of another.” 18 U.S.C. § 16(a).
However, the BIA did not address this alternative statutory
definition and so we similarly do not address it here. See Li
v. Att’y Gen. of the U.S., 

, 163 (3d Cir. 2005).




                               9
considered crimes of violence under § 16(b) and those that
may not be so considered. On the one hand, we have held
that “pure” recklessness crimes are generally not crimes of
violence under § 16(b). 

. Pure
recklessness exists when “the perpetrator runs ‘no risk of
intentionally using force in committing his crime.’” 

(quoting 

). For example, reckless
burning is not a crime of violence under § 16(b) because
“the risk [is] that the fire started by the offender will spread
and damage the property of another,” which “cannot be said
to involve the intentional use of force.” 

. Similarly, crimes that only “raise[] a substantial risk
that accidental, not intentional, force [will] be used,” such
as reckless vehicular homicide, are not crimes of violence
under § 16(b). 

. “The idea of
purposeful action, of actively employing a means to achieve
an end, is an essential component of both ‘use’ and ‘intent,’
and is absent from the concept of ‘recklessness.’” 

.5


       5
         The Supreme Court recently addressed the concept
of “using” force in the related context of 18 U.S.C. §
922(g)(9). See Voisine v. United States, 

(2016). Section 922(g)(9) “prohibits any person convicted
of a ‘misdemeanor crime of domestic violence’ from
possessing a firearm.” 

(quoting 18 U.S.C. §
922(g)(9)). The phrase “misdemeanor crime of domestic
violence” is defined “to include any misdemeanor
committed against a domestic relation that necessarily
involves the ‘use . . . of physical force.’” 

original) (emphasis added) (quoting 18 U.S.C.
§ 921(a)(33)(A)). The question before the Court was




                              10
whether reckless assaults fell within that definition. 

       In answering that question in the affirmative, the
Court observed that an actor who is reckless “with respect
to the harmful consequences of his volitional conduct” can
“use” force within the meaning of § 921(a)(33)(A). 

    To illustrate its point, the Court posited a
hypothetical situation in which “a person throws a plate in
anger against a wall near where his wife is standing.” 

counts as a ‘use’ of force even if the husband did
not know for certain (or have as an object), but only
recognized a substantial risk, that a shard from the plate
would ricochet and injure his wife.” 

not stretch the imagination to see that
applying the Court’s formulation in Voisine to the § 16(b)
context might sweep into the provision’s ambit the pure
recklessness and accidental force recklessness crimes
described above. Both reckless burning and reckless
vehicular homicide involve volitional acts “undertaken with
awareness of their substantial risk of causing injury.” 

“differences in [the] contexts and
purposes” of § 921(a)(33)(A) and § 16, the Court went out
of its way to make clear that its decision in Voisine “does
not resolve whether § 16 includes reckless behavior.” 

n.4. Since we conclude Baptiste’s 2009 Conviction
falls within our more-circumscribed interpretation of
§ 16(b), we need not examine to what extent the reasoning
of Voisine applies in the § 16(b) context to broaden our
existing interpretation of the provision. We leave that
question for another day.




                             11
        However, in contrast to those types of recklessness
crimes, we have recognized that some recklessness crimes
“raise a substantial risk that the perpetrator will resort to
intentional physical force in the course of committing the
crime” and so are crimes of violence under § 16(b).

. In Aguilar v. Attorney General,
we held that the Pennsylvania crime of reckless sexual
assault is a crime of violence under § 16(b). 

Although a defendant may act with a reckless state of mind
in committing the offense, we observed that the defendant’s
actions create a “substantial risk . . . that . . . the offender
will intentionally use force to overcome the victim’s natural
resistance against participating in unwanted intercourse.”

2.     The categorical approach

       In determining whether Baptiste’s 2009 Conviction
was for a crime of violence under the foregoing principles,
we must use the “categorical approach” set forth by the
Supreme Court in Taylor v. United States, 

(1990). The categorical approach is used in a variety of
contexts to determine whether a criminal conviction meets
the requirements of a federal statute triggering some form of
sentencing or immigration consequence. See Rojas v. Att’y
Gen. of the U.S., 

, 214 (3d Cir. 2013) (en
banc); see, e.g., United States v. Tucker, 

, 209
(3d Cir. 2012) (“serious drug offense” requirement in the
ACCA triggering sentencing enhancement); Restrepo v.
Att’y Gen. of the U.S., 

, 791 (3d Cir. 2010)
(“sexual abuse of a minor” requirement in the INA




                              12
triggering removability). Under this approach, we do not
consider the facts underlying Baptiste’s conviction (i.e., the
conduct giving rise to his conviction). See 

. Instead, we “compare [the] federal definition
[of a crime of violence] to the statute of conviction” itself to
determine whether the applicable crime defined in the
statute of conviction is categorically a crime of violence.

of conviction at issue here provides that
“[a] person is guilty of aggravated assault if he . . .
[a]ttempts to cause serious bodily injury to another, or
causes such injury purposely or knowingly or under
circumstances manifesting extreme indifference to the value
of human life recklessly causes such injury.” N.J. Stat. Ann.
§ 2C:12-1b(1) (West 2005). The parties agree that, since
the administrative record does not reveal to which crime in
the statute of conviction Baptiste pleaded guilty, we should
look to the recklessness crime in the statute—recklessly
causing serious bodily injury to another under
circumstances manifesting extreme indifference to the value
of human life. Thus, the question we must answer is
whether recklessly causing serious bodily injury to another
under circumstances manifesting extreme indifference to the
value of human life is categorically a crime of violence
under § 16(b).

       However, the foregoing formulation begs the
question: what does it mean to say that a crime defined in a
statute of conviction is categorically a crime of violence
under § 16(b)?

      Baptiste and the Attorney General advocate opposing
approaches to this question. Baptiste points us to our




                              13
decision in Aguilar, in which we observed without further
exposition that only if the “least culpable conduct necessary
to sustain conviction under [a] statute” constitutes a crime
of violence can the applicable crime defined in the statute
be deemed categorically a crime of violence under § 16(b).

(emphasis added) (internal
quotation marks omitted) (quoting Denis v. Att’y Gen. of the
U.S., 

, 206 (3d Cir. 2011))..6 Baptiste argues
that the least culpable conduct for which there is a
possibility of conviction for reckless second-degree
aggravated assault is drunk driving manifesting extreme
indifference to the value of human life and resulting in
serious bodily injury to another. See, e.g., State v.
Kromphold, 

, 646 (N.J. 2000); State v.
Sweeney, No. 12-08-1429, 

, at *1–*2
(N.J. Super. Ct. App. Div. Oct. 26, 2015). Thus, under
Baptiste’s view, only if that least culpable conduct meets
the definition of a crime of violence in § 16(b) can the
recklessness crime in his statute of conviction be deemed
categorically a crime of violence pursuant to § 16(b).

       The Attorney General counters that we must instead
look to the conduct associated with the “ordinary case” of
reckless second-degree aggravated assault—not the least
culpable conduct. The ordinary case inquiry finds its roots

      6
         Although we have not had occasion to interpret the
“least culpable conduct” language in the § 16(b) context, we
have interpreted it in the CIMT context to mean that “the
possibility of conviction for non-turpitudinous conduct,
however remote, is sufficient to avoid removal.” Jean-
Louis v. Att’y Gen. of the U.S., 

, 471 (3d Cir.
2009).




                             14
in the Supreme Court’s opinion in James v. United States,

(2007), which addressed the operation of the
categorical approach in the related ACCA residual clause
context.    In James, the Court examined whether a
defendant’s conviction in Florida for attempted burglary fell
within the ACCA residual clause definition of a “violent
felony.” The residual clause defines “violent felony” in
relation to a list of enumerated offenses, such as burglary
and extortion, as a crime that “otherwise involves conduct
that presents a serious potential risk of physical injury to
another.” 18 U.S.C. § 924(e)(2)(B)(ii). The defendant
argued that, under the categorical approach, all cases of
attempted burglary under his statute of conviction had to
present a serious potential risk of physical injury to another
before attempted burglary could be deemed categorically a
violent felony. 

.

       The Court concluded that the defendant’s argument
“misapprehend[ed] Taylor’s categorical approach.” 

“[E]very conceivable factual offense covered by a
statute” need not “necessarily present a serious potential
risk of injury before the offense can be deemed a violent
felony.” 

Court concluded that the “proper
inquiry” under the categorical approach is “whether the
conduct encompassed by the elements of the offense, in the
ordinary case, presents a serious potential risk of injury to
another.” 7 

see United States v.
Stinson, 

, 466 (3d Cir. 2010).


       7
        This past year, the Supreme Court re-affirmed the
applicability of the ordinary case inquiry from James to the
categorical approach in the ACCA residual clause context.

. However, it later held the




                             15
       Although James was decided several years before
our opinion in Aguilar, we did not consider in Aguilar
whether the James ordinary case inquiry from the ACCA
residual clause context should displace the least culpable
conduct inquiry in the § 16(b) context. 8 However, since
James, nearly all of our sister circuits have adopted the
ordinary case inquiry in the § 16(b) context. See United
States v. Vivas-Ceja, 

, 722−23 (7th Cir. 2015);
United States v. Keelan, 

, 871 (11th Cir. 2015);
United States v. Avila, 

, 1107 (4th Cir. 2014);
United States v. Fish, 

, 13 (1st Cir. 2014);
Rodriguez-Castellon v. Holder, 

, 854 (9th Cir.
2013); United States v. Echeverria-Gomez, 

,
978 (5th Cir. 2010) (per curiam); Van Don Nguyen v.
Holder, 

, 530 (6th Cir. 2009); United States v.
Sanchez-Garcia, 

, 1213 (10th Cir. 2007).
Additionally, the BIA reached the same conclusion last

residual clause unconstitutionally vague due, in part, to the
indeterminacy of the required ordinary case inquiry. 

Aguilar did not decide this question or
address the Supreme Court’s precedent in James, we may
decline to use Aguilar’s least culpable conduct inquiry if we
determine that the ordinary case inquiry is the correct
analytical approach. See United States v. Tann, 

, 542 (3d Cir. 2009). Moreover, the Supreme Court’s
recent decision in Johnson, in which it re-affirmed the
applicability of the ordinary case inquiry, see supra note 7,
constitutes an intervening Supreme Court decision, which is
also a “sufficient basis” for us to reevaluate our precedent in
Aguilar. Leb. Farms Disposal, Inc. v. County. of Lebanon,

, 249 n.16 (3d Cir. 2008).




                              16
year. See In re Mario Francisco-Alonzo, 26 I. & N. Dec.
594, 601 (B.I.A. 2015).

        We are persuaded that the ordinary case inquiry is
the correct analytical approach in the § 16(b) context.
Section 16(b) requires courts to ask whether a crime “by its
nature” presents a substantial risk of the use of force.
Accordingly, in Leocal v. Ashcroft—the Supreme Court’s
only § 16(b) case—the Court stated that § 16(b) “covers
offenses that naturally involve a person acting in disregard
of the risk that physical force might be used against another
in committing an 

(emphasis
added). As a matter of plain language, asking whether the
least culpable conduct sufficient to support a conviction for
a crime presents a certain risk is inconsistent with asking
whether that crime “by its nature” or “naturally” presents
that risk. See Perez-Munoz v. Keisler, 

, 364
(5th Cir. 2007) (noting that every violation of a state
criminal statute “need not be violent” for the crime “to be a
crime of violence by its nature” (emphasis added)); United
States v. Lucio-Lucio, 

, 1204 n.2 (10th Cir.
2003) (“We do not take the phrase ‘by its nature’ as an
invitation to search for exceptional cases.”).

       By contrast to the least culpable conduct inquiry, the
Supreme Court’s ordinary case inquiry is aligned with the
“by its nature” inquiry that the text of § 16(b) requires.
Asking whether the “ordinary case” of a crime presents a
certain risk is the equivalent of asking whether that crime
“by its nature” presents that same risk. The Court’s
description of the ordinary case inquiry as asking whether
“an offense is of a type that, by its nature” presents a certain




                              17
risk 9 demonstrates the equivalence of the two inquiries. 10

(emphasis added). Accordingly, we

       9
        Although the residual clause does not include the
“by its nature” language in its text, it is clear from this
statement that the Court has read the same “by its nature”
requirement as exists in § 16(b) into the residual clause. See
Shuti v. Lynch, 

, 446—47 (6th Cir. 2016);

.
       10
          We are mindful that the Supreme Court used a
“least of the acts criminalized” inquiry when undertaking
the categorical approach in Moncrieffe v. Holder, 

, 1684−85 (2013). See also Mellouli v. Lynch, 135 S.
Ct. 1980, 1986 (2015). This inquiry asks whether “a
conviction of the state offense ‘necessarily involved . . .
facts equating to [the] generic [federal offense],’”

(alterations in original)
(internal quotation marks omitted) (quoting Shepard v.
United States, 

, 24 (2005)), and so we view it as
synonymous with the least culpable conduct inquiry from
Aguilar. However, we conclude that this inquiry is not
applicable in the § 16(b) context.

       Moncrieffe involved a determination of whether a
predicate crime met the definition of a specific federal
generic offense—“illicit trafficking in a controlled
substance,” 

8 U.S.C. § 1101(a)(43)(B). Other
specific federal generic offenses include a “theft offense,” 8
U.S.C. § 1101(a)(43)(G), “burglary offense,” 

abuse of a minor,” 

The
specific federal generic offense analysis is different in kind
from the analysis required by § 16(b).




                             18
adopt the ordinary case inquiry as part of the categorical
approach in § 16(b) cases.

3.     Application of the categorical approach

        Given our adoption of the ordinary case inquiry in
the § 16(b) context, we now must determine how to
ascertain the ordinary case of reckless second-degree
aggravated assault.      The first step in making this
determination is defining the term “ordinary.” Black’s Law
Dictionary defines “ordinary” as “[o]ccuring in the regular
course of events,” “normal,” and “usual.” Black’s Law
Dictionary 1273 (10th ed. 2014). Other circuits have
defined the ordinary case in a way consistent with this
definition. See 

(looking to the “usual” violation of a statute); United States
v. Sonnenberg, 

, 366 (7th Cir. 2010) (looking

       A specific federal generic offense provision requires
a court to determine whether a predicate crime is, for
example, a “theft offense.” By contrast, § 16(b) requires a
court to determine whether a predicate crime, by its nature,
poses a certain risk. This linguistic distinction explains why
the least of the acts criminalized inquiry is appropriate for
specific federal generic offense cases, but the ordinary case
inquiry is appropriate for § 16(b) cases. See Rodriguez-

(“[A] court considering whether
a state statute meets the definition of ‘sexual abuse of a
minor’ must consider cases ‘at the margins of the statute,’
but a court performing an analysis of ‘substantial risk’
under § 16(b) may not do so.” (quoting Delgado-Hernandez
v. Holder, 

, 1129 (9th Cir. 2012))); In re
Mario Francisco-Alonzo, 26 I. & N. Dec. at 599−600.




                             19
to the “typical case”); Van Don 

(looking to “the mainstream of prosecutions brought under
the statute”); see also Sykes v. United States, 

, 40
n.4 (2011) (Kagan, J., dissenting) (defining the ordinary
case of a crime as the “most common form” of that crime).
Therefore, in ascertaining the ordinary case of reckless
second-degree aggravated assault, we will look to the
conduct associated with the normal or usual commission of
the crime.

        There is little guidance as to how we should go about
identifying that conduct. See 

.
Indeed, during oral argument, neither advocate was able to
articulate the ordinary case of reckless second-degree
aggravated assault. “How does one go about deciding what
kind of conduct the ‘ordinary case’ of a crime involves? ‘A
statistical analysis of the state reporter? A survey? Expert
evidence? Google? Gut instinct?’” 

States v. Mayer, 

, 952 (9th Cir. 2009)
(Kozinski, C.J., dissenting from denial of rehearing en
banc)).      Although we ultimately conclude that the
indeterminate nature of the ordinary case inquiry
contributes to § 16(b)’s unconstitutionality, we must first
undertake the analysis as best we can to determine whether
Baptiste’s 2009 Conviction was for a crime of violence.
See Egolf v. Witmer, 

, 109 (3d Cir. 2008) (“We
have          a          longstanding       practice       of
avoiding constitutional questions in cases where we can
reach a decision upon other grounds.”).

      In the absence of any empirical analysis of
convictions for reckless second-degree aggravated assault,
we are limited to examining New Jersey case law to
determine what conduct is associated with the ordinary case




                             20
of the crime. Our review of case law is complicated in this
case because the statute of conviction at issue includes
several crimes (an attempt crime and a completed crime
phrased with several disjunctive mental states) and the
conviction documents of defendants prosecuted under the
statute often do not specify which crime in the statute the
defendant was convicted of committing. See United States
v. Garcia-Jimenez, 

, 1081 (9th Cir. 2015);
see, e.g., State v. Watkins, No. 12-02-0369, 

, at *2 (N.J. Super. Ct. App. Div. Jan. 4, 2016)
(verdict sheet for second-degree aggravated assault did not
differentiate mental states). This lack of specificity makes it
impossible in many cases to determine whether a defendant
was convicted of the crime at issue in this case—reckless
second-degree aggravated assault—or the other crimes
specified in the statute.11

       However, based on our review of pertinent case law,
we observe that there is a wide array of conduct for which a
defendant can be convicted for reckless second-degree
aggravated assault. For purposes of our analysis, we group
this conduct into three categories: (1) conduct that itself
constitutes an intentional use of force; (2) conduct that
presents a substantial risk of the intentional use of force;
and (3) conduct that presents no risk of the intentional use
of force.

       11
          Given the dearth of New Jersey cases that make
clear a defendant was convicted of the recklessness crime in
the statute, we are forced to depart from our typical practice
and cite to unpublished New Jersey opinions.




                              21
a)     Intentional use of force

       A defendant can be convicted for reckless second-
degree aggravated assault if he intentionally uses force
against a victim and is reckless as to whether that force will
cause “serious bodily injury.” See State v. Jaramillo, No.
04-01-0140, 

, at *11 (N.J. Super. Ct.
App. Div. Aug. 25, 2008) (per curiam) (noting that a jury
was entitled to find the defendant guilty of reckless second-
degree aggravated assault for punching the victim); State v.
Battle, 

, 299 (N.J. Super. Ct. App. Div. 1986)
(observing that a thief’s forceful snatching of a victim’s
purse, which leads to her serious bodily injury, could
constitute reckless second-degree aggravated assault). A
recent case from the New Jersey courts addressing the
closely-related crime of reckless third-degree aggravated
assault12 is illustrative.



       12
          We use the term “third-degree aggravated assault”
here to refer to the crime defined at N.J. Stat. Ann. § 2C:12-
1b(7) (West 2005). Reckless third-degree aggravated
assault is in all material respects identical to reckless
second-degree aggravated assault with the exception that
reckless third-degree aggravated assault results in
“significant bodily injury” as opposed to “serious bodily
injury.” Compare N.J. Stat. Ann. § 2C:12-1b(7) (West
2005) with N.J. Stat. Ann. § 2C:12-1b(1) (West 2005).




                             22
       In State v. Steffen, No. 09-11-2753, 

, at *1−*2 (N.J. Super. Ct. App. Div. Aug. 6, 2012)
(per curiam), the defendant was convicted of reckless third-
degree aggravated assault after using a “choke slam” to
subdue the victim. As a result of the choke slam, the victim
suffered a hematoma and temporary loss of sight. 

The trial court determined that the defendant had “acted
‘recklessly under circumstances manifesting extreme
indifference to the value of human life,’” 

and the
reviewing court affirmed the trial court’s verdict, 

       Such conduct, which involved choke slamming the
victim, itself involves the intentional use of force and so
clearly meets the requirements of § 16(b).13 See Jimenez-

       13
          In addition, there are examples of second-degree
aggravated assault convictions in New Jersey for conduct
clearly involving the intentional use of force for which it is
unclear with what mental state the defendant was convicted
of acting. As we alluded to above, in such cases, the
defendant pleads guilty, or the judge or jury returns a
verdict of guilty, to the general offense of causing serious
bodily injury purposely or knowingly or recklessly under
circumstances manifesting extreme indifference to the value
of human life. See, e.g., Watkins, 

, at *1–
*2 (defendant kicked an elderly man and was convicted
without designation of mental state); State v. Fowlkes, No.
05-09-1271, 

, at *1–*3 (N.J. Super. Ct.
App. Div. Jan. 12, 2010) (per curiam) (defendant punched
victim and hit victim with a broom and was convicted
without designation of mental state).

      It stands to reason that some of these convictions,
which involve the intentional use of force and do not




                             23
Gonzalez v. Mukasey, 

, 561 (7th Cir. 2008)
(examining cases holding that recklessness crimes are
crimes of violence under § 16(b) as involving
“intentional conduct exhibiting a reckless disregard to the
likelihood of injury”); Blake v. Gonzales, 

, 161
n.6 (2d Cir. 2007) (finding a crime to be a crime of violence
under § 16(b) where, under one theory of violation, “the
perpetrator intends the conduct, and . . . recklessness is
the mens rea with respect to the likelihood of physical
harm” (alteration in original) (internal quotation marks
omitted)).

b)     Substantial risk of intentional use of force

       A defendant can also be convicted for conduct that,
while itself not constituting an intentional use of force,
presents a substantial risk that he will intentionally use
force. For example, in State v. Colon, 

,
1361−62 (N.J. Super. Ct. App. Div. 1997), the defendant’s
friend was being battered by a group of men outside a bar.
The bar’s bouncer testified that he had grabbed hold of one
of the men and was pulling him off of the defendant’s friend
when that man was shot. 

The jury found that
the defendant had shot the victim, but acquitted him of
purposeful or knowing aggravated assault; instead, it
convicted him only of reckless second-degree aggravated
assault. 

n.3, 1364. Although several theories of
the crime could have supported the jury’s verdict, relevant

designate a mental state, are based on a reckless mental
state whereby the defendant, as in Steffen, intentionally used
force but was reckless as to the possibility of serious bodily
injury.




                             24
for our purposes is the court’s comment that the verdict
could have been the result of a jury finding that the
defendant “recklessly fired [his] weapon.” 

        As we explained above, we determined in Aguilar
that a reckless sexual assault is a crime of violence because
there is a substantial risk that the defendant will encounter
resistance from the victim and then decide to intentionally
use force to “overcome” the victim. See 

−02. Similarly, in Colon, once the defendant
recklessly fired his weapon and hit the victim, there was a
substantial risk that the victim would fight back and that the
defendant would then decide to intentionally fire his
weapon (i.e., intentionally use force against the victim).
Although not a certainty, the reckless firing of the weapon
created a substantial risk of that result, which is all that §
16(b) requires.14


       14
           Although this analysis considers conduct and
events taking place after the recklessness crime has
technically been completed, it is consistent with our prior
interpretations of the “in course of committing the offense”
language in § 16(b). See 18 U.S.C. § 16(b) (defining a
crime of violence as “a felony . . . that, by its nature,
involves a substantial risk that physical force against the
person or property of another may be used in the course of
committing the offense” (emphasis added)).

       For example, we observed in Aguilar, in dicta, that
burglary is a crime of violence under § 16(b). 

; see 

(observing that
burglary is the “classic example” of a crime of violence
under § 16(b)). The crime of burglary—breaking and




                             25
entering a dwelling at night to commit a felony—is
technically complete as soon as the defendant has entered
the dwelling. However, we observed that burglary is a
crime of violence under § 16(b) because “burglary creates a
substantial risk that the burglar will have to use physical
force to overcome the desire of home occupants to protect
themselves and their property.” 

.
This risk only materializes after the defendant has entered
the dwelling and thus after the crime of burglary has been
completed. See 

risk of the use of force
as being “created by an unlawful entry into a victim’s
home”); Henry v. Bureau of Immigration & Customs Enf’t,

, 310 (3d Cir. 2007) (“[T]he requisite elements
of a burglary are complete once the burglar enters and
possesses the necessary mental intent. However, the
substantial risk that the burglar will use force comes from
the possibility that the burglar will encounter another during
the course of the burglary; it is irrelevant that the technical
elements have already been accomplished.”); cf. 

(“[A] risk of injury arises . . . because the
burglar might confront a resident in the home after breaking
and entering.”).

       Similarly, we observed in Ng v. Attorney General
that the use of interstate commerce facilities in the
commission of a murder-for-hire is a crime of violence
under § 16(b). Ng v. Att’y Gen. of the U.S., 

,
397 (3d Cir. 2006). That crime is technically complete after
mere solicitation to commit a murder-for-hire and so
“proscribes conduct that may never pose a risk of violence.”

observed that it is a crime of violence under
§ 16(b) because, even if “some violations . . . will never




                              26
c)      No risk of intentional use of force

        Finally, a defendant can be convicted for conduct
that presents no risk that he will intentionally use force.
Specifically, in accordance with Baptiste’s suggested least
culpable conduct, a defendant can be convicted for reckless
second-degree aggravated assault for drunk driving
manifesting extreme indifference to the value of human life
and resulting in serious bodily injury to another. See, e.g.,

; Sweeney, 

,
at *1–*2. Common to such drunk driving cases is that the
defendant did not intend to cause harm to the victim and so
is not “actively employ[ing]” force in committing the crime.

; see Oyebanji v. Gonzales, 

, 264 (3d Cir. 2005). Moreover, such conduct does not
present a “risk that the reckless[] offender will step in and
commit an intentional act of violence.” 

−73.

                        *      *       *

      Our task is to determine, based on the foregoing
review of case law, what conduct is associated with the

culminate in . . . the commission of a murder[,] . . . the
natural consequence of [the commission of the crime] is that
physical force will be used upon another.” 

United States v. Hull, 

, 140 (3d Cir. 2006)
(“[M]ere possession of a pipe bomb holds no risk of
the intentional use of force. . . . [T]he relevant inquiry is not
whether possession makes it more likely that a violent crime
will be committed, but instead whether there is a risk that in
committing the offense of possession, force will be used.”).




                               27
ordinary case of reckless second-degree aggravated assault.
Unsurprisingly, the Attorney General urges us to focus on
conduct in the first two categories and Baptiste urges us to
focus on conduct in the third category. In the absence of
any concrete guidance as to how to make this determination,
see 

−58, we must rely on our
common sense and judicial experience, see 

; 

.

        We recognize that it is impossible in this case to
determine with precision what specific conduct is associated
with the ordinary case of the crime. The crime at issue in
this case covers a wide array of conduct—more than, say,
burglary. A defendant can be convicted of the crime for
conduct as dissimilar as an intentional act of physical
violence (first category of conduct) and drunk driving
causing accidental injury (third category of conduct). With
a crime that covers such a wide array of conduct, we begin
with the common sense proposition that the conduct
associated with the ordinary case of a conviction
presumptively lies at or near the middle of the culpability
spectrum15—here, the second category of conduct we have
identified.



      15
          We use the term “culpability spectrum” here to
refer to conduct that, on one end of the spectrum, presents
no risk of the intentional use of force (third category of
conduct) and, on the other end of the spectrum, involves an
intentional use of force (first category of conduct).




                            28
       Baptiste’s single factual scenario to the contrary in
which there is no risk of the intentional use of force—a
drunk driver—is not enough to overcome this presumption.
We have seen nothing in our foregoing review of case law
that persuades us that the normal or usual commission of the
crime involves the actions of a drunk driver (third category
of conduct). Rather, we view such conduct as being
associated with a narrow subset of convictions and thus
insufficient to render the crime categorically not a crime of
violence under the ordinary case inquiry. Cf. Van Don

(“[A]n unsubstantiated risk of
physical force in some small subset of cases is [in]sufficient
to classify [an] offense as a ‘crime of violence.’”). We
reach the same conclusion with respect to the first category
of conduct we have identified.

        We therefore conclude that the conduct associated
with the ordinary case of reckless second-degree aggravated
assault lies somewhere within the second category of
conduct we have identified, which falls within the definition
of a crime of violence in § 16(b).16 See 

(referring to the ordinary case as a “judge-imagined

       16
         If this conclusion is unsatisfying, it is the result of
the indeterminacy of the ordinary case inquiry, which
requires us to determine what conduct is associated with the
normal conviction of the crime despite the broad swath of
disparate conduct it covers. See 

(“How does common sense help a federal court discern
where the ‘ordinary case’ of vehicular flight in Indiana lies
along th[e] spectrum [of culpable conduct]?”). We address
this indeterminacy in the next section. See infra section
III.B.




                              29
abstraction”). Because we conclude that reckless second-
degree aggravated assault does, in the ordinary case, present
a substantial risk of the intentional use of force, reckless
second-degree aggravated assault in New Jersey is
categorically a crime of violence pursuant to § 16(b).

       Given our conclusion that Baptiste was convicted of
a crime of violence pursuant to § 16(b), we now turn to the
constitutional question presented in this case—is § 16(b)
void for vagueness under the Due Process Clause of the
Fifth Amendment?

B.    Section 16(b) is void for vagueness under the Due
Process Clause of the Fifth Amendment

       The Due Process Clause precludes the government
from taking away a person’s life, liberty, or property under
a statute “so vague that it fails to give ordinary people fair
notice of the conduct it punishes, or so standardless that it
invites arbitrary enforcement.” 

.
Baptiste argues that his 2009 Conviction was not for an
aggravated felony because the incorporated definition of a
crime of violence in 18 U.S.C. § 16(b) is unconstitutionally
vague. 17 Baptiste bases his argument on the Supreme

       17
          The Attorney General wisely does not contest
Baptiste’s assertion that he has a right under the Fifth
Amendment’s Due Process Clause to bring a void for
vagueness challenge to the definition of a crime of violence
in § 16(b). See Jordan v. De George, 

, 231
(1951) (considering whether the phrase “crime involving
moral turpitude” was void for vagueness due to the “grave
nature of deportation”); Golicov v. Lynch, --- F.3d ----, No.
16-9530, 

, at *2−*3 (10th Cir. Sept. 19,




                             30
Court’s recent decision in Johnson v. United States, 135 S.
Ct. 2551 (2015), which invalidated the residual clause of the
ACCA.

        The ACCA provides for a sentence enhancement for
certain defendants who have three or more prior convictions
for a “violent felony.” 

The Act defines
“violent felony” as, inter alia, a crime that is “burglary,
arson, or extortion, involves use of explosives, or otherwise
involves conduct that presents a serious potential risk of
physical injury to another.” 18 U.S.C. § 924(e)(2)(B)(ii)
(emphasis added). The emphasized language is known as
the “residual clause.” As we explained above, prior to
Johnson’s holding that the residual clause is
unconstitutionally vague, courts assessing whether a crime
fell within the residual clause were required to use the same

2016); 

(“[B]ecause deportation strips
a non-citizen of his rights, statutes that impose this penalty
are subject to vagueness challenges under the Fifth
Amendment.”); Dimaya v. Lynch, 

, 1112−14
& n.4 (9th Cir. 2015), cert. granted, --- S. Ct. ----, No. 15-
1498, 

(U.S. Sept. 29, 2016) (concluding
that an alien “may bring a void for vagueness challenge to
the definition of a ‘crime of violence’ in the INA” and
collecting cases from other circuits permitting similar
challenges).    “It is well established that the Fifth
Amendment entitles aliens to due process of law in
deportation proceedings.” Demore v. Kim, 

,
523 (2003) (internal quotation marks omitted) (quoting
Reno v. Flores, 

, 306 (1993)); see, e.g., 

−19 (entertaining an alien’s procedural due
process challenge).




                             31
categorical approach that courts use in the § 16(b) context.

section III.A.2. Thus, in “[d]eciding whether the
residual clause covers a crime,” a court had to “picture the
kind of conduct that the crime involves in ‘the ordinary
case,’ and . . . judge whether that abstraction presents a
serious potential risk of physical injury.” Johnson, 135 S.
Ct. at 2557 (quoting 

).

        The majority in Johnson observed that two features
of the residual clause “conspire[d] to make it
unconstitutionally vague”—the ordinary case inquiry and
the serious potential risk inquiry. 

First, the
majority observed that there are many different conceptions
of what the ordinary case of a crime involves. 

“does the ordinary instance of witness tampering
involve offering a witness a bribe? Or threatening a witness
with violence?” 

The majority concluded that
“[t]he residual clause offers no reliable way to choose
between . . . competing accounts of what [an] ‘ordinary’
[case] involves.” 

Second, the majority
observed that the clause left “uncertainty about how much
risk it takes for a crime to qualify as a violent felony.” 

majority concluded that the combination of
“indeterminacy about how to measure the risk posed by a
crime with indeterminacy about how much risk it takes for
the crime to qualify as a violent felony . . . produces more
unpredictability and arbitrariness than the Due Process
Clause tolerates.” 

this conclusion, the majority
examined the residual clause precedents of both the
Supreme Court and the Courts of Appeals and determined
that “repeated failures to craft a principled and objective
standard out of the residual clause confirm its hopeless




                             32
indeterminacy.” 

addressed several arguments
penned by the dissent. First, it rejected as inconsistent with
the Court’s precedents the dissent’s view that “a statute is
void for vagueness only if it is vague in all its applications.”

Second, the majority dismissed the dissent’s
concern that the invalidation of the residual clause for
vagueness would cast constitutional doubt over laws similar
to the residual clause that use terms such as “substantial
risk.” 

reasoned that such laws do not link
the phrase “substantial risk” to a “confusing list of
examples,” and, “[m]ore importantly . . . require gauging
the riskiness of conduct in which an individual defendant
engages on a particular occasion.” 

majority rejected the dissent’s invitation to abandon the
ordinary case inquiry and interpret the residual clause to
“refer to the risk posed by the particular conduct in which
the defendant engaged.” 

       In addressing whether Johnson compels the
invalidation of § 16(b), we do not write on a blank slate.
The Sixth, Seventh, Ninth, and Tenth Circuits have
considered the question and concluded that Johnson does
render § 16(b) void for vagueness. See Shuti v. Lynch, 

(6th Cir. 2016); United States v. Vivas-Ceja, 

(7th Cir. 2015); Dimaya v. Lynch, 

(9th Cir. 2015), cert. granted, --- S. Ct. ----, No. 15-1498,

(U.S. Sept. 29, 2016); Golicov v. Lynch,
--- F.3d ----, No. 16-9530, 

(10th Cir.
Sept. 19, 2016). By contrast, the en banc Fifth Circuit has
concluded that § 16(b) is not unconstitutionally vague after
Johnson, and the Second and Eighth Circuits have
concluded that 18 U.S.C. § 924(c)(3)(B), which contains




                              33
nearly identical language to § 16(b), 18 survives Johnson.
See United States v. Prickett, --- F.3d ----, No. 15-3486,

(8th Cir. Oct. 5, 2016); United States v.
Gonzalez-Longoria, 

(5th Cir. 2016) (en banc);
United States v. Hill, 

(2d Cir. 2016). We
enter the fray with the benefit of these considered opinions
on § 16(b)’s constitutionality.

       The two features of the residual clause that the
Supreme Court concluded “conspire[d] to make [the
residual clause] unconstitutionally vague” were the ordinary
case inquiry and the serious potential risk inquiry. 

−58; see United States v. Calabretta, 

, 133 (3d Cir. 2016). Given that the ordinary case
inquiry, as used in the § 16(b) context, is derived from the
residual clause context, we can be certain that the ordinary
case inquiry is identical in both contexts. As we described
above, in the § 16(b) context, a court must ask “whether the
conduct encompassed by the elements of the offense, in the
ordinary case, presents a [substantial risk of the intentional
use of force].” 

(emphasis added).

       18
         Before the Sixth Circuit’s decision in Shuti holding
§ 16(b) to be vague, a panel of the Sixth Circuit had
concluded that § 924(c)(3)(B) was not unconstitutionally
vague after Johnson. See United States v. Taylor, 

(6th Cir. 2016). However, in Shuti, the Sixth Circuit
distinguished Taylor, noting that “[u]nlike the ACCA and
INA, which require a categorical approach to stale predicate
convictions, 18 U.S.C. § 924(c) is a criminal offense that
requires an ultimate determination of guilt beyond a
reasonable doubt—by a jury in the same proceeding.”

.




                             34
Because § 16(b) “offers no reliable way to choose between .
. . competing accounts of what” that “judge-imagined
abstraction” of the crime involves, 

, the ordinary case inquiry is as indeterminate in the §
16(b) context as it was in the residual clause context. See
Golicov, 

, at *6; 

;

−23; 

−16.

       This conclusion holds true for the second feature of
each statute as well—the risk inquiry. Whereas the residual
clause asks how much risk it takes for a crime to present a
“serious potential risk” of physical injury, § 16(b) asks how
much risk it takes for a crime to present a “substantial risk”
of the intentional use of force. The phrases have two
linguistic differences: § 16(b) replaces the residual clause’s
“serious” with the word “substantial” and replaces the
residual clause’s “potential risk” with “risk.”

       A “serious risk” is equally as vague as a “substantial
risk.” See Golicov, 

, at *6. To be sure, a
“potential risk” encompasses more conduct than a simple
“risk.” See 

−08 (“[T]he combination
of the two terms suggests that Congress intended to
encompass possibilities even more contingent or remote
than a simple ‘risk.’”). However, in our view, this minor
linguistic distinction is insufficient to bring § 16(b) outside
of the reasoning of Johnson. See 

; 

n.9. The critical feature of
the “serious potential risk” inquiry that rendered it
indeterminate in Johnson was not that the risk was
“potential,” but that the residual clause required the use of a
vague “serious risk” inquiry. The majority confirmed as
much when, in response to the dissent’s suggestion that the




                              35
majority opinion would cast constitutional doubt on statutes
using a “substantial risk” inquiry, it did not draw any
vagueness distinction between the phrases based on the
word “potential.” See 

.

        The Attorney General directs our attention to an
additional linguistic distinction between the statutes that she
views as meaningful. She argues that the scope of crimes
that present a substantial risk of the use of force is narrower
than the scope of crimes that presents a serious potential
risk of physical injury. See Prickett, 

, at
*2; 

; 

. This is so because there is undoubtedly a class of
conduct that presents a risk that a victim will be injured
without presenting a risk that force will intentionally be
used against that victim. See 

n.7
(noting that § 16(b) “plainly does not encompass all
offenses which create a ‘substantial risk’ that injury will
result from a person’s conduct”). One example of such
conduct is arson with intent to destroy a building, which
runs the risk of a victim being injured without any risk of
the arsonist using intentional force against that victim. The
Attorney General argues that the § 16(b) inquiry therefore
“falls short of the wide-ranging thought experiment
previously required by the [residual clause].” Resp’t Br. 44
(internal quotation marks omitted) (quoting United States v.
Doe, 

, 182 (E.D.N.Y. 2015)).

       While the Attorney General is correct that fewer
crimes fall within § 16(b) than within the residual clause,
we do not view the scope of crimes covered by each
provision as integral to the vagueness analysis. The
Attorney General cannot point us to any language in
Johnson that suggests otherwise because the Court’s




                              36
vagueness holding in Johnson was focused on the “serious
potential risk” inquiry required by the residual clause. See

(“[T]he residual clause leaves
uncertainty about how much risk it takes for a crime to
qualify as a violent felony. It is one thing to apply an
imprecise ‘serious potential risk’ standard to real-world
facts; it is quite another to apply it to a judge-imagined
abstraction.” (emphasis added)); Welch v. United States,

, 1262 (2016) (“The residual clause failed
not because it adopted a ‘serious potential risk’ standard but
because applying that standard under the categorical
approach required courts to assess the hypothetical risk
posed by an abstract generic version of the offense.”
(emphasis added)). As such, we focus here in our
vagueness analysis on the “substantial risk” inquiry required
by § 16(b).

       In applying those indeterminate risk inquiries,
whether fewer or more cases fall within each respective
statutory provision because of the modifiers “physical
injury” and “use of force” does not affect the indeterminacy
of the “serious potential risk” or “substantial risk” inquiries
themselves. See 

(Thomas, J.,
dissenting) (observing that the residual clause was held to
be vague because it requires courts to “judge whether [the
ordinary case of a crime] presents a serious potential risk of
some result” (emphasis added) (internal quotation marks
omitted)). In short, the distinction the Attorney General
draws between the two statutes is a distinction without a




                              37
difference within the reasoning of Johnson.19 See 

.


       19
         The Fifth Circuit in Gonzalez-Longoria identified
another linguistic distinction between the residual clause
and the language of § 16(b), which contributed to its
conclusion that § 16(b) is not unconstitutionally vague. It
pointed to the requirement in § 16(b) “that the risk of
physical force arise ‘in the course of committing’ the
offense” and observed that the § 16(b) inquiry is narrower
than the residual clause inquiry because it “does not allow
courts to consider conduct or events occurring after the
crime is complete.” 

(emphasis added).

        However, as we explained supra note 14, we have
not always interpreted § 16(b) in such a restrictive manner
as we have sometimes considered conduct occurring after
the offense has technically been completed in our
substantial risk inquiry. See, e.g., 

;
see also 

(White, J., concurring in
part and dissenting in part) (“[T]he cases demonstrate that
the phrase ‘in the course of committing the offense’ has not
consistently been interpreted to exclude consideration of the
risk of force after the offense has technically been
completed.”); 

(observing that the
Ninth Circuit has similarly not interpreted § 16(b) in such a
restrictive manner).

       Moreover, the Supreme Court’s observation that
burglary is the “classic example,” of a crime of violence
within the meaning of § 16(b), 

,
suggests that it similarly does not so restrictively interpret




                             38
        The Attorney General next asserts that § 16(b) does
not fall within the reasoning in Johnson because, “unlike the
list of exemplar crimes preceding the residual clause, . . .
§ 16(b) . . . do[es] not rely [on] a unique list of enumerated
crimes to complicate the assessment of risk.”20 Resp’t Br.
46; 18 U.S.C. § 924(e)(2)(B)(ii) (defining “violent felony”
as a crime that is “burglary, arson, or extortion, involves
use of explosives, or otherwise involves conduct that
presents a serious potential risk of physical injury to

the “in the course of committing the offense” language in §
16(b). See 

. As the Court explained
in Johnson, “[t]he act of . . . breaking and entering into
someone’s home does not, in and of itself, normally cause
physical injury. Rather, risk of injury arises . . . because the
burglar might confront a resident in the home after breaking
and entering.” 

.
       20
           Section 4B1.2 of the Sentencing Guidelines
previously contained a residual clause defining a “crime of
violence” that was both identically worded to the residual
clause in the ACCA and preceded by a list of exemplar
crimes. Accordingly, we recently held the residual clause
that was in § 4B1.2 to be void for vagueness after Johnson.
See 

. In invalidating that
residual clause, we noted that “we need not consider — and
so leave for another day — whether a similar residual clause
without an exemplary list of offenses would be subject to
the same degree of due process concern that the Supreme
Court identified in Johnson.” 

n.9. Today is that
day. As we explain herein, we find § 16(b), which does not
contain an exemplary list of offenses, to be
unconstitutionally vague.




                              39
another” (emphasis added)); see Prickett, 

, at *2; 

; 

. It is true that the majority in Johnson
commented on the confusion engendered by the list of
exemplar crimes preceding the residual clause.             See

, 2561. In responding to the
dissent’s argument that holding the residual clause
unconstitutional would place numerous provisions of
federal and state law that use terms like “substantial risk” in
constitutional doubt, the majority retorted:

       Almost none of the cited laws links a phrase
       such as “substantial risk” to a confusing list of
       examples.      “The phrase ‘shades of red,’
       standing alone, does not generate confusion or
       unpredictability; but the phrase ‘fire-engine red,
       light pink, maroon, navy blue, or colors that
       otherwise involve shades of red’ assuredly does
       so.”

(quoting 

n.7 (Scalia, J.,
dissenting)).

        However, in the very next sentence of the opinion, in
response to the dissent’s same argument, the majority
stated:

       More importantly, almost all of the cited laws
       require gauging the riskiness of conduct in
       which an individual defendant engages on a
       particular occasion. As a general matter, we do
       not doubt the constitutionality of laws that call
       for the application of a qualitative standard such
       as “substantial risk” to real-world conduct; “the




                              40
      law is full of instances where a man’s fate
      depends on his estimating rightly . . . some
      matter of degree[.]”       The residual clause,
      however, requires application of the “serious
      potential risk” standard to an idealized ordinary
      case of the crime.

in original) (first emphasis added)
(internal citation omitted) (quoting Nash v. United States,

, 377 (1913)); see 

.

        We read Johnson to mean that the confusing list of
examples preceding the residual clause only added to the
residual clause’s already-existing vagueness. Indeed, the
language in Johnson by no means suggests that the list of
examples was an integral component of the Court’s finding
that the residual clause was unconstitutionally vague. See
Golicov, 

, at *7; 

;

−18. Rather, as the Supreme
Court made clear, the vagueness was the product of “[t]wo
features of the residual clause”—the ordinary case inquiry
and the risk inquiry—which, as we explained above, are
present in the § 16(b) analysis as well.21 Johnson, 135 S.
Ct. at 2557; see 

−23.


      21
          The Supreme Court’s discussion in Johnson about
its “repeated failures to craft a principled and objective
standard out of the residual clause” does not change our
analysis. 

. The Court’s
difficulty in interpreting the residual clause on multiple
occasions merely provided further “evidence of vagueness,”

, that the Court had already
found in the provision as a result of the “[t]wo features of




                            41
        In fact, the lack of examples in § 16(b) introduces at
least as much vagueness into the provision as the presence
of confusing examples introduced into the residual clause.
See 

n.13. “The specific offenses
[preceding the residual clause] provide [a] baseline from
which to measure whether other similar conduct ‘otherwise
. . . presents a serious potential risk of physical injury.’”

(third alteration in original). This
baseline “provide[s] at least some guidance as to the sort of
offenses Congress intended for the [residual clause] to
cover.” 

n.13. Such guidance is
absent from § 16(b), which contains no example offenses.
As a result, courts are left to undertake the § 16(b) analysis
guided by nothing more than other judicial decisions that
can lay no better claim to making sense of the
indeterminacy of the analysis in a principled way than we
have today. 

section III.A.3.


the residual clause [that] conspire[d] to make it
unconstitutionally vague,” 

Thus, that difficulty
only served to “confirm [the residual clause’s] hopeless
indeterminacy.” 

(emphasis added); see 

−62 (distinguishing between the Court’s
difficulty in interpreting the residual clause and its
vagueness analysis); 

; 

. Moreover, the fact that the Supreme Court
has only taken and decided one § 16(b) case, see 

, and so has not experienced repeated failures
in interpreting the provision, is probative only of the Court’s
composition of its docket—not absence of vagueness in the
provision. See 

; 

.




                              42
                       *      *       *

       Seemingly lost in these nuanced arguments about the
scope and import of Johnson is the fact that the Supreme
Court expressly anticipated the effect its holding would
have on statutes with the language contained in § 16(b). In
addressing the applicability of its holding to those statutes,
the Court stated: “As a general matter, we do not doubt the
constitutionality of laws that call for the application of a
qualitative standard such as ‘substantial risk’ to real-world
conduct.” 

(emphasis added);
see 

. Section 16(b) is not such a
law. Rather, § 16(b) calls for the exact analysis that the
Court implied was unconstitutionally vague—the
application of the “substantial risk” inquiry to the “idealized
ordinary case” of a crime. 

(emphasis added).

       Thus, because the two inquiries under the residual
clause that the Supreme Court found to be indeterminate—
the ordinary case inquiry and the serious potential risk
inquiry—are materially the same as the inquiries under
§ 16(b), § 16(b) is unconstitutionally vague. See Golicov,

, at *6; 

; Vivas-

−23; 

. “By
combining indeterminacy about how to measure the risk
posed by a crime with indeterminacy about how much risk
it takes for the crime to qualify as” a crime of violence,
§ 16(b) “produces more unpredictability and arbitrariness
than the Due Process Clause tolerates.” 

.

      Because § 16(b) is invalid, Baptiste’s 2009
Conviction was not for an aggravated felony pursuant to 8




                              43
U.S.C. § 1227(a)(2)(A)(iii). However, since Baptiste does
not contest that his 1978 Conviction was for a CIMT, he is
still removable if his 2009 Conviction was for a CIMT. We
now turn to that question.

C.     Baptiste’s 2009 Conviction was for a CIMT

       An alien who is convicted of “two or more crimes
involving moral turpitude, not arising out of a single scheme
of criminal misconduct” after his admission to the United
States is removable pursuant to 8 U.S.C. §
1227(a)(2)(A)(ii). Baptiste argues that the BIA erred in
concluding that his 2009 Conviction was for a CIMT. In
determining whether that conviction was for a CIMT, we
must again follow the categorical approach. Mehboob v.
Att’y Gen. of the U.S., 

, 275 (3d Cir. 2008). As
with our crime of violence determination, the parties agree
that, in undertaking the categorical approach, we should
look to the recklessness crime in the statute of conviction.
Thus, the question we must answer is whether recklessly
causing serious bodily injury to another under
circumstances manifesting extreme indifference to the value
of human life is categorically a CIMT.

       In the CIMT context, our cases make clear that “we
look to the elements of the statutory offense to ascertain the
least culpable conduct hypothetically necessary to sustain a
conviction under the statute.” 

(internal quotation marks omitted) (quoting 

). Thus, the “possibility of conviction for non-
turpitudinous conduct, however remote, is sufficient to
avoid removal.” 

marks omitted)
(quoting 

). Under these
dictates, if there is any non-turpitudinous conduct that could




                             44
sustain a conviction for reckless second-degree aggravated
assault, then that crime is categorically not a CIMT.

       We have in the past defined morally turpitudinous
conduct as “inherently base, vile, or depraved, contrary to
the accepted rules of morality and the duties owed other
persons.” Hernandez-Cruz v. Att’y Gen. of the U.S., 

, 284 (3d Cir. 2014) (internal quotation marks
omitted) (quoting Knapik v. Ashcroft, 

, 89 (3d
Cir. 2004)). Such conduct can “inhere in serious crimes
committed recklessly, i.e., with a conscious disregard of a
substantial and unjustifiable risk that serious injury or death
would follow.” Partyka v. Att’y Gen. of the U.S., 

, 414 (3d Cir. 2005). Specifically, a recklessness crime
can constitute a CIMT “if certain statutory aggravating
factors are present.” 

; see Idy v.
Holder, 

, 118−19 (1st Cir. 2012) (recklessness
coupled with “serious bodily injury” aggravating factor).

        In Knapik, the BIA concluded that first-degree
reckless endangerment under New York law was a 

. New York law provided that a “person is
guilty of reckless endangerment in the first degree when,
under circumstances evincing a depraved indifference to
human life, he recklessly engages in conduct which creates
a grave risk of death to another person.” 

(quoting
N.Y. Penal Law § 120.25 (McKinney 2009)).                We
concluded that the BIA had acted reasonably in concluding
that the New York crime constituted a CIMT. 

       In so concluding, we observed that the New York
statute at issue defined a recklessness crime that
“contain[ed] aggravating factors, requiring that a defendant
create a ‘grave risk of death to another person’ ‘under




                              45
circumstances evincing a depraved indifference to human
life.’” 

on to observe that “the BIA could
reasonably conclude that the elements of depravity,
recklessness and grave risk of death, when considered
together, implicate accepted rules of morality and the duties
owed to society.” 

recklessness crime
defined in the statute of conviction in this case uses
nominally different wording, it is in all material respects the
same as the New York crime in Knapik that we found the
BIA reasonably classified as morally turpitudinous.

       First, both crimes are recklessness crimes and the
mental state of recklessness is virtually identical under New
York and New Jersey law. In New York, “[a] person acts
recklessly . . . when he is aware of and consciously
disregards a substantial and unjustifiable risk” that is “of
such nature and degree that disregard thereof constitutes a
gross deviation from the standard of conduct that a
reasonable person would observe in the situation.” N.Y.
Penal Law § 15.05(3) (McKinney 2009). In New Jersey,
“[a] person acts recklessly . . . when he consciously
disregards a substantial and unjustifiable risk” that is “of
such a nature and degree that . . . its disregard involves a
gross deviation from the standard of conduct that a
reasonable person would observe in the actor’s situation.”
N.J. Stat. Ann. § 2C:2-2b(3) (West 2005).

        Second, the aggravating factors in both crimes are
virtually identical. As to the first aggravating factor, the
New York crime required that the defendant act “under
circumstances evincing a depraved indifference to human
life,” N.Y. Penal Law § 120.25 (McKinney 2009), whereas
the New Jersey crime at issue here requires that the
defendant act “under circumstances manifesting extreme




                              46
indifference to the value of human life,” N.J. Stat. Ann. §
2C:12-1b(1) (West 2005).        There is no meaningful
difference between those two phrases.

       As to the second aggravating factor, the New York
crime required that the defendant engage in conduct that
“creates a grave risk of death to another person.” N.Y.
Penal Law § 120.25 (McKinney 2009). Similarly, the New
Jersey crime at issue here requires conduct that results in
“serious bodily injury.” N.J. Stat. Ann. § 2C:12-1b(1)
(West 2005). And the New Jersey courts have required that
the defendant be aware that “his conduct [bears] a
substantial risk that he will kill or seriously injure” others.

(alteration in original). This risk
must be so great that it constitutes a “probability as opposed
to the mere possibility of serious bodily injury.” State v.
Pigueiras, 

, 1096 (N.J. Super. Ct. App. Div.
2001); see 

(concluding
Pennsylvania’s reckless endangerment crime is not a CIMT
because it “only requires conduct that may put a person in
danger”). Again, the aggravating factor in each crime is
materially the same.

        Thus, the New Jersey crime of reckless second-
degree aggravated assault, which requires recklessly causing
serious bodily injury to another under circumstances
manifesting extreme indifference to the value of human life,
falls squarely within our opinion in Knapik as a recklessness
crime with two aggravating factors. Reckless second-
degree aggravated assault is a CIMT.22


       22
         In arguing for a contrary result, Baptiste points us
to   reported convictions for reckless second-degree




                              47
Because Baptiste’s 2009 Conviction was for a CIMT,23 the
BIA correctly determined that, together with his 1978


aggravated assault for drunk driving and cites our statement
in Knapik that “drunk driving . . . almost certainly does not
involve moral turpitude.”       

.
However, we were careful in Knapik not to foreclose the
possibility that some egregious forms of drunk driving
could involve moral turpitude. We were merely referring in
that case to a “simple DUI offense,” 

marks omitted) (quoting In re Lopez-Meza, 22 I. & N. Dec.
1188, 1194 (B.I.A. 1999)), and not drunk driving as
prosecuted under the statute at issue here, which results in
serious bodily injury to another person and evinces extreme
indifference to the value of human life. Such egregious
conduct is undoubtedly turpitudinous.

        Baptiste also argues that our decision in Partyka
compels the conclusion that his 2009 Conviction was not
for a CIMT. However, in Partyka, we concluded that
negligently assaulting a law enforcement officer was not a
CIMT so the holding in that case is not applicable to the
more culpable recklessness crime at issue here. 

. Moreover, we expressly stated in Partyka
that, if the petitioner was convicted of recklessly assaulting
a law enforcement officer, we would agree with the BIA’s
conclusion that the crime involved moral turpitude. 

holding today is limited to the New Jersey
crime of reckless second-degree aggravated assault, which
requires recklessly causing serious bodily injury to another
under circumstances manifesting extreme indifference to the
value of human life. We express no opinion on whether an




                             48
Conviction, Baptiste is removable as an alien convicted of
two or more CIMTs pursuant to 8 U.S.C. §
1227(a)(2)(A)(ii).

IV.   CONCLUSION

        For the foregoing reasons, we will grant the petition
in part as it relates to the BIA’s aggravated felony
determination, deny the petition in part as it relates to the
BIA’s CIMT determination, and remand the case to the BIA
for further proceedings.




assault crime involving “ordinary” recklessness would
constitute a CIMT.




                             49