United States of America v. Michael Ryan Coulson

United States Court of Appeals
For the Eighth Circuit
___________________________
No. 23-1690
___________________________
United States of America
Plaintiff - Appellee
v.
Michael Ryan Coulson
Defendant - Appellant
____________
Appeal from United States District Court
for the Northern District of Iowa - Central
____________
Submitted: October 26, 2023
Filed: November 20, 2023
____________
Before LOKEN, MELLOY, and KELLY, Circuit Judges.
____________
MELLOY, Circuit Judge.
Michael Ryan Coulson was convicted at a court martial for “forcible
pandering” in violation of Article 120c(b) of the Uniform Code of Military Justice
(UCMJ), codified at 10 U.S.C. § 920c(b) (2012). He later failed to register as a sex
offender in Iowa and pleaded guilty to violating the Sex Offender Registration and
Notification Act (SORNA). See 18 U.S.C. § 2250(a); see also, generally, 34 U.S.C.
§ 20901 et seq.

-- 1 of 9 --

-2-
The sentence for a SORNA conviction depends in part on the severity of the
defendant’s underlying sex offense as categorized by tier. Tier III is the most severe
category and Tier I is the least severe, serving as a catchall when Tiers II or III do
not apply. Tiers II or III apply when the SORNA defendant’s underlying offense is
“comparable to or more severe than” a listed offense. See 34 U.S.C. § 20911(2)–(4)
(defining Tiers I–III); see also U.S.S.G. § 2A3.5(a) (setting base offense levels).
Coulson argued below that he was a Tier I offender. The district court concluded he
was a Tier III offender and sentenced him accordingly.
On appeal, Coulson argues the “categorical approach” applies to his SORNA
tier analysis. He also argues that, pursuant to this approach, his prior statute of
conviction is “overbroad” in relation to SORNA’s Tier III listed comparator
offenses.1 We now hold for the first time, in line with a consensus of authority from
other circuits, that the categorical approach applies to SORNA’s tier analysis.
Further, we agree with Coulson that application of the categorical approach results
in his classification as a Tier I offender. Accordingly, we reverse and remand for
resentencing.
I.
Under the UCMJ, Coulson’s underlying offense of forcible pandering
amounts to forced prostitution defined in material part as compelling another person
to engage in “sexual abuse” or “sexual contact” for which anyone is paid. 10 U.S.C.
§ 920c(b) (defining pandering through reference to prostitution); 10 U.S.C.
§ 920c(d)(1) (defining prostitution through reference to “sexual abuse” or “sexual
contact” as further defined in 10 U.S.C. § 920(g)). For the present analysis, the
critical distinction between these two statutory terms boils down to the fact that
sexual contact may occur outside or over a victim’s clothing, whereas sexual abuse
may not.2
1 The parties present no arguments concerning Tier II.
2 10 U.S.C. § 920(g) provides:

-- 2 of 9 --

-3-
At sentencing for the present SORNA conviction, the district court faced the
question of whether the circumstance-specific approach, modified-categorical
approach, categorical approach, or some other approach applied to determine if
Coulson’s prior conviction was “comparable to or more severe than . . . aggravated
sexual abuse or sexual abuse (as described in sections 2241 and 2242 of Title 18).”
34 U.S.C. § 20911(4)(A)(i). The United States argued in its district court briefing
and at the sentencing hearing that the term “comparable” as used in § 20911 provided
an expansive degree of flexibility and allowed the court to move away from the rigid
structure of the categorical or modified-categorical approach. In support, the United
States relied on dicta from an unpublished Tenth Circuit opinion and argued,
“[R]egardless of which approach applies . . . even if the prior conviction statute is
slightly broader, this wider protective sweep is allowable under SORNA’s tier
regime.” In the alternative, the United States argued that even if the word
“comparable” did not provide an additional degree of flexibility, the modified-
categorical approach should apply. And in advocating for the modified-categorical
Definitions.–In this section:
(1) Sexual act. –The term “sexual act” means–
(A) the penetration, however slight, of the penis into the
vulva or anus or mouth;
(B) contact between the mouth and the penis, vulva,
scrotum, or anus; or
(C) the penetration, however slight, of the vulva or penis
or anus of another by any part of the body or any object,
with an intent to abuse, humiliate, harass, or degrade any
person or to arouse or gratify the sexual desire of any
person.
(2) Sexual contact. –The term “sexual contact” means touching,
or causing another person to touch, either directly or through the
clothing, the vulva, penis, scrotum, anus, groin, breast, inner
thigh, or buttocks of any person, with an intent to abuse,
humiliate, harass, or degrade any person or to arouse or gratify
the sexual desire of any person. Touching may be accomplished
by any part of the body or an object.
(Emphasis added).

-- 3 of 9 --

-4-
approach, the United States argued that permissibly viewable Shepard materials
showed Coulson’s underlying pandering conviction had involved sexual abuse—
compelled intercourse involving an adult woman with money paid to Coulson—
rather than mere “sexual contact.”3
The district court appears to have partially adopted the United States’s
arguments. The district court stated that the categorical approach applied but held
that the possibility of a pandering (prostitution) conviction arising from mere sexual
contact over the clothing was so unlikely as to be speculative or hypothetical. As a
result, the district court found the pandering conviction comparable to sexual abuse,
18 U.S.C. § 2242, resulting in a Tier III classification.
We interpret the district court’s conclusion, articulated generally as the
categorical approach but applied with some degree of flexibility, to reflect either: (1)
the partial adoption of the United States’s argument that the term “comparable”
broadens SORNA’s tier inquiry away from a traditional categorical or modified-
categorical approach; or (2) an application of the “realistic probability” qualifier that
courts sometimes apply to reject “fanciful” interpretations of unclear statutes under
the categorical approach when analyzing prior convictions. See, e.g., Gonzales v.
Duenas-Alvarez, 549 U.S. 183, 193 (2007) (“[T]o find that a state statute creates a
crime outside the generic definition of a listed crime in a federal statute requires
more than the application of legal imagination to a state statute’s language. It
requires a realistic probability, not a theoretical possibility, that the State would
apply its statute to conduct that falls outside the generic definition of a crime.”); but
see Gonzalez v. Wilkinson, 990 F.3d 654, 660 (8th Cir. 2021) (refusing to apply the
realistic probability test where the statute at issue was “clear on its face”).
On appeal, Coulson renews his arguments, but the United States’s position is
somewhat unclear. The United States does not argue clearly against the categorical
3Shepard v. United States, 544 U.S. 13, 20–21 (2005) (permitting courts to
review certain reliable judicial records to determine which subpart of a divisible
statute a defendant had been convicted of violating).

-- 4 of 9 --

-5-
approach but continues to urge the adoption of some degree of flexibility based on
the statutory term “comparable.” 34 U.S.C. § 20911(4)(A)(i). However, in briefing
and at oral argument, the government was unable to articulate standards or any
guiding principles as to how this “categorical light” approach would work in
practice.
II.
In general, we review de novo the classification of prior offenses or
convictions for sentencing purposes. See United States v. Bragg, 44 F.4th 1067,
1075 (8th Cir. 2022). De novo review applies to the questions before us because
determination of the proper framework for SORNA’s tier analysis, and application
of that framework, are questions of law. This remains true even if classification of
a prior conviction requires the court to review limited materials under the modified-
categorical approach, as urged by the United States. United States v. Myers, 928
F.3d 763, 765–67 (8th Cir. 2019). Such an exercise involves interpretation of the
legal consequences flowing from prior judicial records. It does not involve fact
finding in a traditional sense that might merit deference.
Several years ago, a panel of our court identified, but was not required to
resolve, the outstanding question of how to conduct SORNA’s tier analysis. See
United States v. Mulverhill, 833 F.3d 925, 930 (8th Cir. 2016) (applying plain error
review and finding it unnecessary to “wade into the quagmire of which approach
applies to the three tier classifications”). At that time, a few circuits had applied the
categorical approach. See, e.g., United States v. Berry, 814 F.3d 192, 197 (4th Cir.
2016); United States v. Morales, 801 F.3d 1, 5–6 (1st Cir. 2015); United States v.
White, 782 F.3d 1118, 1134–36 (10th Cir. 2015); United States v. Cabrera-
Gutierrez, 756 F.3d 1125, 1133 (9th Cir. 2014). Since then, a broad consensus has
grown, with no circuits holding to the contrary. See United States v. Walker, 931
F.3d 576, 579 (7th Cir. 2019); United States v. Barcus, 892 F.3d 228, 231–32 (6th

-- 5 of 9 --

-6-
Cir. 2018); United States v. Young, 872 F.3d 742, 746 (5th Cir. 2017).4 We now join
these other circuits.
In general, as applied in several different criminal- and immigration-law
contexts, the categorical approach does not permit a court to consider a defendant’s
actual underlying conduct. See, e.g., Moncrieffe v. Holder, 569 U.S. 184, 190 (2013)
(directing courts to consider “not . . . the facts of the particular prior case, but instead
. . . whether the . . . statute defining the crime of conviction categorically fits within
the generic federal definition” (internal citations omitted)). This approach permits
only an elements-to-elements comparison between a defendant’s prior offense and
either: (1) a general or traditional common law definition of a referenced offense,
e.g., “burglary” as referenced in 18 U.S.C. § 924(e); or (2) the elements of an offense
as identified with express reference to a particular statutory provision. SORNA’s
tier provisions involve the latter in that 34 U.S.C. § 20911(4)(A)(i) expressly
references the definitions of “aggravated sexual abuse” and “sexual abuse” from 18
U.S.C. §§ 2241 and 2242.
Like the circuits cited above, we find that textual support points almost
exclusively toward the categorical approach. Reference to a generic comparator
“offense,” to a specific statute, or to a “conviction”—all as contrasted with
references to conduct or to specific acts that a defendant previously committed—
strongly suggests courts must apply the categorical approach. See Nijhawan v.
Holder, 557 U.S. 29, 36–37 (2009) (distinguishing a Congressional directive to
review specific facts from language typically interpreted as permitting only the
review of elements; stating that reference to a generic crime or a “particular section
of the Federal Criminal Code” suggests application of the categorical approach); see
also United States v. White, 782 F.3d 1118, 1132 (10th Cir. 2015).
4 The unpublished Tenth Circuit opinion cited by the United States below is
not to the contrary because the key discussion referenced by the United States was
mere dicta. See United States v. Escalante, 933 F.3d 395, 402 n.9 (5th Cir. 2019)
(discussing United States v. Forster, 549 F. App’x 757 (10th Cir. 2013)
(unpublished)).

-- 6 of 9 --

-7-
In contrast, when courts have elected to focus on facts, acts, and conduct rather
than statutory elements, courts typically have identified statutory language
indicating a clear Congressional directive to consider such facts. In Nijhawan, for
example, the Court addressed different aspects of a prior offense inquiry in the
immigration context. There, the Court settled upon a hybrid approach using the
categorical approach generally but calling for a broader evidentiary inquiry and a
circumstance-specific approach where Congress had made immigration
consequences dependent on specific facts apart from the fact of conviction. See
Nijhawan, 557 U.S. at 38 (“The upshot is that [8 U.S.C. § 1101(a)(43)(M)(i)]
contains some language that refers to generic crimes and some language that almost
certainly refers to the specific circumstances in which a crime was committed.”).
In fact, our court found clear Congressional language mandating a
circumstance-specific approach under a different provision of SORNA itself. See
United States v. Hill, 820 F.3d 1003, 1005 (8th Cir. 2016) (applying a factual
analysis under 42 U.S.C. § 16911(7)(I) (now codified at 34 U.S.C. § 20911(7)(I))
where Congress specifically directed an examination of “conduct”). Several of the
circuits cited above have done the same, essentially adopting a “hybrid” approach
under SORNA depending upon the statutory subsection being applied. Under this
hybrid approach, courts use the categorical approach generally in the tier analysis
and employ a circumstance-specific approach only where Congress separately
identified “conduct” or a victim’s age as relevant to a SORNA determination. See,
e.g., United States v. Berry, 814 F.3d 192, 196–97 (4th Cir. 2016).
Apart from statutory language, we conclude that several prudential
considerations support the categorical approach. First, by refusing to reach beyond
the elements of a prior offense, we avoid possible encroachment on a defendant’s
Sixth Amendment jury rights. Descamps v. United States, 570 U.S. 254, 269–70
(2013). The fact of a conviction, whether by jury or plea, after all, establishes
“nothing more than the least of the acts criminalized.” Gonzalez, 990 F.3d at 658
(quoting Moncrieffe, 569 U.S. at 190–91)). Cabining our analysis to the elements
properly respects the limits of a previous jury’s conclusions. Second, the categorical

-- 7 of 9 --

-8-
approach respects the benefits of any earlier plea agreements or bargains that may
have driven a defendant’s decision to waive or exercise jury rights.
Third, and finally, it is prudent to avoid the general inefficiency and the
unfairness or unreliability that may exist in the factual review of prior offenses. See
Moncrieffe, 569 U.S. at 200–01 (“[T]he procedure the Government envisions would
require precisely the sort of post hoc investigation into the facts of predicate offenses
that we have long deemed undesirable. The categorical approach serves ‘practical’
purposes: It promotes judicial and administrative efficiency by precluding the
relitigation of past convictions in minitrials conducted long after the fact.” (citation
omitted)); see also United States v. Vasquez-Garcia, 449 F.3d 870, 873 (8th Cir.
2006) (discussing the modified categorical approach and stating, “The Court
declined to adopt ‘a more inclusive standard of competent evidence,’ citing the
‘practical difficulties and potential unfairness of a factual approach’ and the ‘respect
for congressional intent and avoidance of collateral trials.’” (quoting Shepard, 544
U.S. at 20, 23) (internal citations omitted)). Collateral inquiries into the details of
potentially “aged” prior acts necessarily raise concerns regarding the availability and
the quality or staleness of evidence. Descamps, 570 U.S. at 270. Moreover, even
where such evidence is available, it may be far afield and down separate paths from
the investigation and prosecutorial efforts that support the later, separate
prosecution.
The categorical approach largely resolves these concerns, thus protecting
investigatory and judicial resources as well as Sixth Amendment rights. It does so
without finesse in that sometimes-reliable evidence as to actual circumstances may
make the earlier criminal act appear different than the elements alone might suggest.
Like the several circuits cited above, however, we conclude this trade-off favoring
efficiency and the protection of jury rights is the correct choice.
Finally, turning to application of the categorical approach, the present case
leaves no room for the United States’s suggested and generalized broadening of the
analysis based on SORNA’s use of the term “comparable.” To broaden the inquiry

-- 8 of 9 --

-9-
beyond the elements would introduce a confusing double measure of subjectivity. Is
flexibility to be found surrounding the elements of the prior offense or the elements
of the SORNA comparators? How is comparability measured if not by the elements?
And what does comparability mean if it is distinct from relative severity?
Subjectivity and flexibility may come into play later in the imposition of a final
sentence through the application of 18 U.S.C. § 3553(a), after tier determination and
after application of the Sentencing Guidelines. They do not fit well into the tier
analysis itself.
Here, the offense of conviction is unambiguously broader in scope than the
SORNA comparators. The offense of “forcible pandering” employs the defined term
“prostitution,” which may be found based on “sexual contact.” 10 U.S.C. § 920c(b)
(forcible pandering); 10 U.S.C. § 920c(d)(1) (prostitution). “Sexual contact”
includes over-the-clothes touching. 10 U.S.C. § 920(g)(2). Such contact falls
outside the scope of SORNA’s Tier III comparator offenses as found in 18 U.S.C.
§§ 2241 and 2242, both of which require a “sexual act” as defined in 18 U.S.C.
§ 2246 and which does not include over-the-clothes touching.
To the extent the government argues there is no “realistic probability” that
prostitution would involve over-the-clothes touching, we find no lack of clarity in
the statutes as required to open the door for application of the “realistic probability”
exception. Gonzalez, 990 F.3d at 660.
We reverse Coulson’s sentence and remand for further proceedings consistent
with this opinion.
LOKEN, Circuit Judge, dissenting without opinion.
______________________________

-- 9 of 9 --