In the
United States Court of Appeals
For the Seventh Circuit
____________________
Nos. 15-3589 & 15-3601
UNITED STATES OF AMERICA,
Plaintiff-Appellee,
v.
ANNA F. NOVAK AND JOHN C. MORRISON,
Defendants-Appellants.
____________________
Appeals from the United States District Court for the
Western District of Wisconsin.
No. 14-cr-121 — James D. Peterson, Judge.
____________________
ARGUED SEPTEMBER 28, 2016 — DECIDED NOVEMBER 9, 2016
____________________
Before POSNER, FLAUM, and MANION, Circuit Judges.
FLAUM, Circuit Judge. Defendants-appellants John Morri-
son and Anna Novak both pled guilty to distributing a con-
trolled substance analog and to tax fraud. The district court
accepted their guilty pleas and later sentenced Morrison to
forty-eight months of incarceration and Novak to ninety-six
months. They now appeal, challenging the constitutionality
of the Controlled Substances Analogue Act, 21 U.S.C. § 813
2 Nos. 15-3589 & 15-3601
(the Analogue Act), the district court’s acceptance of their
guilty pleas, and their sentences. We affirm.
I. Background
From approximately November 2009 through September
2013, Morrison and Novak sold substances they referred to as
“herbal incense” from their retail store, JC Moon, in Ashland,
Wisconsin. These substances included XLR-11, UR-144, PB-
22, and 5F-PB-22, which at the time were not listed on the con-
trolled substance schedules but were similar to scheduled
controlled substances. Between November 2012 and July
2013, undercover law enforcement officers made twenty-
eight controlled purchases of various substances from JC
Moon.
Through undercover interviews with Novak, the govern-
ment learned that appellants hid a significant portion of their
cash income from the IRS. Novak directed JC Moon employ-
ees to take all $50 and $100 bills from the cash registers and
put them in Novak and Morrison’s safe. These bills never
went into JC Moon’s business bank account. This artificially
low business income information was provided to the appel-
lants’ tax preparer; as a result, the “skimmed” cash was not
included on the returns and not reported to the IRS. From
2010 to 2012, appellants under-reported JC Moon’s business
income by approximately $575,752 for a tax loss of $186,095.
On December 10, 2014, a federal grand jury returned a
thirty-five-count indictment against Novak and Morrison.
These counts included the Analogue Act and tax fraud counts
to which Novak and Morrison would eventually plead guilty.
Nos. 15-3589 & 15-3601 3
On May 5, 2015, Novak and Morrison moved to dismiss
the thirty indictment counts alleging either conspiracy or sub-
stantive distribution charges. Appellants argued before the
district court that the Analogue Act was unconstitutionally
vague as applied to XLR-11 and several other analogues in-
volved in the case. On July 24, 2015, the court denied Novak
and Morrison’s motion to dismiss.
On August 27, 2015, Novak and Morrison pled guilty to
one count each of distributing a controlled substance ana-
logue—XLR-11—and to tax fraud. The plea agreements were
unconditional and did not reserve the right to appeal the pre-
viously-denied motion to dismiss.
At Morrison’s change of plea hearing, the district court re-
viewed the maximum penalties and asked whether Morrison
understood the proceedings. The government then summa-
rized the evidence it would submit at trial on the Analogue
Act charge, including employee and customer testimony that
appellants sold synthetic marijuana as “herbal incense” from
a back room at JC Moon; a Facebook post indicating that ap-
pellants sold XLR-11 after they knew it was scheduled to be
treated as a controlled substance;* undercover agent testi-
mony regarding a May 2, 2013 XLR-11 purchase; expert testi-
mony demonstrating the physiological-effects and chemical-
* The May 2, 2013 post read: “The federal government is banning the
current herbal incense on May 13, 2013. What that means for us is every-
thing we are selling right now will be banned as of May 13. Our inventory
is limited. We may run out before then. We will keep you posted.” The
government said it also would have submitted evidence that XLR-11 was
to be officially treated as a controlled substance on May 16, 2013, but that
the notice of this change had been publicized earlier.
4 Nos. 15-3589 & 15-3601
structural similarities between XLR-11 and JWH-18, a con-
trolled substance; and lay witness testimony from JC Moon
customers regarding the effects of XLR-11. Morrison then
agreed that the government could have proved all of the
above facts.
When questioned, both appellants vacillated between
knowing and not knowing that XLR-11 was similar to con-
trolled substances while they were selling it. When the court
first asked Morrison to explain the distribution offense in his
own words, he said,
I didn’t realize I was being charged with any-
thing because I didn’t think there was anything
wrong with [the XLR-11] at the time. Because
every time that something came down, we al-
ways stopped and … I didn’t know. I’m sorry. I
really didn’t know it was illegal.
Morrison then said that he knew that people used the
XLR-11 he sold to get high, he “thought it was like mariju-
ana,” and he knew marijuana was a controlled substance.
However, when questioned further, Morrison stated that he
did not know that XLR-11 was a controlled substance when
he sold it and that he stopped selling it once he believed it
became a scheduled controlled substance.
In response, the government said that XLR-11 was in fact
not a scheduled controlled substance as of the date of the
charged offense, May 2, 2013. Rather, the evidence showed
that Morrison knew that XLR-11 would be treated as a con-
trolled substance later that same month, yet continued to sell
it after having that knowledge. The government continued,
Nos. 15-3589 & 15-3601 5
Mr. Morrison’s forthright answers that he knew
people were smoking, it means he knew it was
for human consumption. He knew they were
using it to get high, which we had a lot of proof
of that and I knew he would say that, and then
his acknowledgement that he thought it was
like marijuana, a Schedule 1 controlled sub-
stance, I think is sufficient.
One of the things McFadden [v. United States,
— U.S. —, 135 S. Ct. 2298 (2015)] tells us is you
can prove knowingly … by showing that the de-
fendant knew the characteristics that would
make it illegal even if he did not know its legal
status as an analogue.… I would submit … that
there are facts, circumstantial facts that he knew
it produced a high. It could give rise to an infer-
ence. The Court could infer … that he then knew
enough—that it would be an analogue even
though he doesn’t know its legal status as an an-
alogue.
The district court concluded,
I’m satisfied that there’s a factual basis for the
plea, despite Mr. Morrison’s denials of the spe-
cific knowledge of the chemical involved or its
specific legal status. But I think on the theory
that [the government] described that … Mr.
Morrison was familiar with the features of the
products that he was selling that made them
controlled substance analogues, … there is in-
deed a factual basis for the plea.
6 Nos. 15-3589 & 15-3601
Morrison also initially seemed unsure about how his tax
reporting worked. He was not aware of how much money JC
Moon made in 2012 or how much business income it reported.
He also said the he was unaware of the cash-skimming oper-
ation underlying the tax fraud. However, after speaking with
his attorney, Morrison said, “We put the fifties and hundreds
in the safe as a separate, and … the taxes were made out the
way they were, I signed them and I’m guilty of it.” Morrison
also said that he and Novak gave their business income infor-
mation—without including the $50 and $100 bills—to their
tax preparer, and that Morrison signed the return. The court
questioned Morrison on his ability to raise lack of knowledge
as a defense to the tax fraud charge. Morrison responded that
he knew and that he nevertheless wanted to plead guilty.
The court concluded,
despite Mr. Morrison’s denial, I think there is a
factual basis … because of his knowledge of the
treatment of the fifties and hundreds at the
store, that he knew that the tax return that he
signed was not an accurate tax return and that
it understated the income ….
The district court conditionally accepted Morrison’s guilty
pleas pending a review of his presentence report.
The district court then proceeded with Novak’s change of
plea hearing. The court reviewed the maximum penalties and
ensured that Novak understood the proceedings. The govern-
ment noted that evidence from Morrison’s change of plea
hearing applied to Novak. Novak was present at the defense
table during that recitation and waived a proffer of the same
Nos. 15-3589 & 15-3601 7
evidence at her hearing, which took place directly after Mor-
rison’s. Novak agreed that the government could prove eve-
rything from Morrison’s proffer. Then, like Morrison, Novak
expressed some hesitation about the distribution charge. No-
vak initially said she thought XLR-11 was supposed to be to-
bacco, but then later realized it was not. Novak also stated
that she did not know people smoked the substances sold at
JC Moon.
After speaking with her attorney, Novak said she later
learned that people were smoking the herbal incense, but that
people were interested in its “relaxing effect.” When asked if
she knew that people used the substances to get high, Novak
responded, “I don’t like to use that terminology. [I]t affects
everybody differently. Some it would relax; some would go
to sleep; some would buzz ….” Novak later admitted that JC
Moon sold XLR-11 after Novak knew it had effects similar to
marijuana.
When questioned about the May 2, 2013 Facebook post,
Novak said she was out of the state and did not create the post
but that she was “aware of a date coming up, this particular,
and our wholesalers would say [XLR-11] is going to be
banned and so we just posted that instantly, not knowing it
was going to come back at us.” The district court then found
that there was a sufficient factual basis for accepting Novak’s
guilty plea on the distribution charge.
Regarding the tax fraud charge, Novak admitted that she
knew the $50 and $100 bills were not reflected in the infor-
mation she gave to the tax preparer, and that therefore the tax
returns reflected fraudulently low business income. The court
found that there was a sufficient factual basis for Novak’s
plea. The court conditionally accepted Novak’s pleas pending
8 Nos. 15-3589 & 15-3601
a review of the presentence report. Neither appellant objected
to the factual basis for the guilty pleas or attempted to with-
draw the pleas. Novak and Morrison did not orally reserve
their rights to appeal the district court’s earlier denial of their
motion to dismiss.
On November 6, 2015, the district court formally accepted
appellants’ guilty pleas and sentenced them. At the hearing,
the government called an expert who testified that XLR-11,
UR-144, PB-22, and 5F-PB-22 had pharmacological effects
substantially similar to THC and were at least as potent as
THC. The expert also provided reasons why the substances
were not most similar to marijuana. Five lay witnesses, all
customers at JC Moon, testified to the effects of the sub-
stances. The court then found that THC was the closest con-
trolled substance to the substances at issue. THC had a 1:167
marijuana equivalency ratio, used to calculate the base of-
fense level under U.S.S.G. § 2D1.1. The district court applied
the 1:167 ratio to the substances in this case.
The court sentenced Morrison to forty-eight months in
prison on the Analogue Act charge and thirty-six months in
prison on the tax fraud charge, with the sentences to run con-
currently. The court sentenced Novak to ninety-six months in
prison on the drug charge and thirty-six months in prison on
the tax charge, with these sentences also running concur-
rently.
II. Discussion
A. Analogue Act Vagueness Challenge
Typically, we review de novo the district court’s legal con-
clusions regarding the constitutionality of a statute. United
States v. Morris, 821 F.3d 877, 879 (7th Cir. 2016). However, an
Nos. 15-3589 & 15-3601 9
unconditional guilty plea waives appellate review of all non-
jurisdictional pretrial issues. United States v. Adams, 125 F.3d
586, 588 (7th Cir. 1997). “An exception exists if the plea is con-
ditioned on preserving specified issues for appeal,” and alt-
hough “[t]hat conditional plea should be in writing, … we
have held that the conditions can also be sufficiently indicated
in a transcript of the sentencing hearing.” United States v. Rob-
inson, 20 F.3d 270, 273 (7th Cir. 1994) (citing Fed. R. Crim. P.
11(a)(2); United States v. Yasak, 884 F.2d 996, 999 (7th Cir.
1989)).
The threshold question is therefore whether Novak and
Morrison’s vagueness challenge alleges jurisdictional, pre-
plea defects that this Court may consider even after an uncon-
ditional plea. This Court has spoken directly on the issue:
“While a facial attack on a statute’s constitutionality is juris-
dictional, an as-applied vagueness challenge is not.” United
States v. Phillips, 645 F.3d 859, 863 (7th Cir. 2011). To the extent
that appellants make non-jurisdictional as-applied challenges
to the Analogue Act, those arguments are waived. Their
guilty pleas did not reserve the right to appeal the district
court’s denial of their motion to dismiss, and nothing in the
record indicates that they preserved the issue during the hear-
ings before the district court. To the extent that appellants fa-
cially attack the Analogue Act on vagueness grounds, the Su-
preme Court resolved this issue in the government’s favor in
McFadden v. United States, — U.S. —, 135 S. Ct. 2298, 2307
(2015) (holding that the Analogue Act is not unconstitution-
ally vague due to its scienter requirement).
B. Guilty Pleas
Appellants next say they unknowingly and involuntarily
entered their guilty pleas and argue the district court accepted
10 Nos. 15-3589 & 15-3601
the pleas without a sufficient factual basis. Where, as here, a
defendant makes no attempt to withdraw his plea in the dis-
trict court, we review the claim for plain error. United States v.
Pineda-Buenaventura, 622 F.3d 761, 770 (7th Cir. 2010). Like-
wise, we conduct plain-error review for the sufficiency of a
factual basis to support a plea where the defendant failed to
make such an objection below. United States v. Arenal, 500 F.3d
634, 637 (7th Cir. 2007).
1. Voluntariness
Rule 11 requires that before the court accepts a
plea of guilty or nolo contendere the court must
address the defendant personally in open court
and inform the defendant of, and determine
that the defendant understands[,] the nature of
each charge to which the defendant is plead-
ing.… To determine whether a defendant in fact
understands the nature of a charge, we take a
totality-of-the-circumstances approach and con-
sider (1) the complexity of the charge; (2) the de-
fendant’s intelligence, age, and education;
(3) whether the defendant was represented by
counsel; (4) the district judge’s inquiry during
the plea hearing and the defendant’s own state-
ments; and (5) the evidence proffered by the
government.
Pineda-Buenaventura, 622 F.3d at 770 (internal citations, quota-
tion marks, brackets, and alterations omitted).
The balance of these factors leads to the conclusion that
Morrison and Novak understood the nature of the charges
Nos. 15-3589 & 15-3601 11
against them and that they knowingly and voluntarily en-
tered their guilty pleas. First, while the complexity of Ana-
logue Act charges can be difficult for defendants to grasp, the
district court can alleviate this burden through careful ques-
tioning and explanation of the charges. See United States v. Fer-
nandez, 205 F.3d 1020, 1026 (7th Cir. 2000); United States v. Le-
Donne, 21 F.3d 1418, 1423 (7th Cir. 1994). The district court in
this case reviewed the elements of the Analogue Act and Con-
trolled Substances Act with appellants and thoroughly ques-
tioned each on his or her knowledge of XLR-11’s effects.
Second, Morrison was seventy-one years old at the time of
the plea, had served in the Army, and had earned a GED. He
had been operating businesses like JC Moon around Wiscon-
sin for many years. English is his native language. Though
there is some suggestion in the change of plea hearing that
Morrison was suffering from the early stages of Alzheimer’s,
there is no indication that any such complications impacted
his ability to understand the nature of the proceedings. Novak
was fifty-eight years old when she pled guilty. She had grad-
uated high school and had taken some college courses. She
ran the operations of JC Moon. Previously, she had managed
her own restaurant. English is also Novak’s native language.
Nothing about appellants’ personal history indicated that ei-
ther was unable to understand the proceedings or the nature
of the charges.
Third, both Novak and Morrison were represented by
counsel. Fourth, as explained above, the district court re-
viewed the nature of the charges with each appellant and
dwelled on what each knew about XLR-11 and at what time.
Finally, the government summarized, in plain language, the
expert and lay witness testimony it expected to elicit at trial,
12 Nos. 15-3589 & 15-3601
and both appellants agreed that the government could prove
all of the facts mentioned. These factors, and the totality of the
circumstances surrounding Novak’s and Morrison’s guilty
pleas, indicate that they entered into those pleas voluntarily
and with sufficient understanding of the charges against
them.
2. Factual Basis
Appellants next argue the district court lacked a sufficient
factual basis on which to accept their distribution pleas and
Morrison’s tax fraud plea. As to the Analogue Act charges,
appellants say the government did not proffer, and the district
court did not elicit, sufficient evidence of their knowledge
about the chemical structure and effects of XLR-11. Alterna-
tively, they contend that if the court found sufficient evidence
to satisfy the knowledge requirement, it did so by impermis-
sibly relying on a now-disfavored Turcotte inference. See
United States v. Turcotte, 405 F.3d 515, 527 (7th Cir. 2005) (cre-
ating the “provisional remedy” of a rebuttable, permissive in-
ference of knowledge in Analogue Act cases, such that a de-
fendant knew about the chemical structure of an analogue if
the government can show the defendant knew the analogue
had similar physiological effects to those of a controlled sub-
stance).
The Analogue Act calls for analogues to be treated as con-
trolled substances. 21 U.S.C. § 813. Though this statute does
not itself contain any scienter requirements, courts have inter-
preted the Analogue Act to incorporate the knowledge re-
quirement from federal laws prohibiting the distribution of
scheduled controlled substances. That is, a defendant must
have known that the substance in question was a controlled
Nos. 15-3589 & 15-3601 13
substance or analogue, just as someone who distributes a con-
trolled substance must have known that he was dealing with
a controlled substance. McFadden, 135 S. Ct. at 2305 (citing 21
U.S.C. § 841(a)(1)). McFadden sets forth two ways in which the
government can satisfy this knowledge requirement:
First, [knowledge] can be established by evi-
dence that a defendant knew that the substance
with which he was dealing is some controlled
substance—that is, one actually listed on the
federal drug schedules or treated as such by op-
eration of the Analogue Act—regardless of
whether he knew the particular identity of the
substance. Second, it can be established by evi-
dence that the defendant knew the specific ana-
logue he was dealing with, even if he did not
know its legal status as an analogue.
Id.
To satisfy the knowledge requirement the second way, the
Supreme Court further reasoned that the government must
show that a defendant knew the substance he is charged with
distributing had (1) a chemical structure substantially similar
to that of an already-scheduled controlled substance and (2) a
physiological effect substantially similar to or greater than the
effect of an already-scheduled controlled substance. Id.
Appellants are correct that after McFadden, courts can no
longer rely on the Turcotte inference to demonstrate a defend-
ant’s chemical-structure and physiological-effects knowledge.
Turcotte effectively collapsed the second requirement (that the
defendant know the analogue had physiological effects simi-
14 Nos. 15-3589 & 15-3601
lar to those of a controlled substance) into the first require-
ment (that the defendant know that the analogue had a chem-
ical structure substantially similar to that of a controlled sub-
stance) in a way that is unsupported by McFadden’s reasoning.
However, the government may still prove either through cir-
cumstantial evidence. McFadden, 135 S. Ct. at 2304, n.1 (“Cir-
cumstantial evidence could include, for example, a defend-
ant’s concealment of his activities, evasive behavior with re-
spect to law enforcement, knowledge that a particular sub-
stance produces a ‘high’ similar to that produced by con-
trolled substances, and knowledge that a particular substance
is subject to seizure at customs.”); id. at 2306, n.3.
In the case at hand, considering the government’s prof-
fer—which both appellants admitted could be proven at
trial—and appellants’ admissions at the change of plea hear-
ings, the district court had sufficient circumstantial evidence
to satisfy the first McFadden knowledge requirement: appel-
lants knew the substance in question was a controlled sub-
stance or analogue.
As explained above, the parties agreed the government
could prove at trial that appellants sold XLR-11 as “herbal in-
cense” from the back of their store, JC Moon. While selling it,
appellants learned from customers and employees that XLR-
11 would give users a high. Novak admitted that she knew
XLR-11 got people buzzed, and Morrison believed it was like
marijuana. In other words, appellants knew XLR-11 was not
meant to be used as incense; rather, they knew it was for
smoking and that it would give a buzz or high similar to ma-
rijuana. Most tellingly, JC Moon’s May 2, 2013 Facebook post
said, “The federal government is banning the current herbal
incense on May 13, 2013. What that means for us is everything
Nos. 15-3589 & 15-3601 15
we are selling right now will be banned as of May 13. Our
inventory is limited. We may run out before then. We will
keep you posted.” Appellants knew XLR-11 was going to be
treated as a controlled substance, yet they continued to sell it.
In sum, there was enough evidence before the district court to
find that appellants were aware of the specific analogue they
were dealing with and its status as a controlled substance an-
alogue. The court committed no plain error in concluding that
it had a sufficient factual basis to accept Novak’s and Morri-
son’s guilty pleas on the Analogue Act charges.
The district court likewise had a sufficient factual basis on
which to accept Morrison’s guilty plea for tax fraud. Morrison
initially hesitated to admit he knew about the cash-skimming
scheme; however, he later admitted to knowing that the $50
and $100 bills went into his and Novak’s safe, that the taxes
were prepared without including income from the skimmed
bills, and that he knowingly signed off on the artificially low
tax returns. Furthermore, the district court informed Morri-
son that not knowing the submitted tax information was false
would be a defense to the tax fraud charge. Morrison still pled
guilty. Thus, the district court did not err in accepting Morri-
son’s plea to the tax fraud charge.
C. 1:167 Marijuana Equivalency Ratio
Finally, appellants argue the district court erred in finding
that THC was the most closely related controlled substance to
XLR-11, UR-144, PB-22, and 5F-PB-22 for purposes of calcu-
lating the offense level under U.S.S.G. § 2D1.1 cmt. n.6. The
district court’s conclusion that THC is the most closely related
controlled substance to XLR-11, UR-144, PB-22, and 5F-PB-22
is a finding of fact we review for clear error. United States v.
16 Nos. 15-3589 & 15-3601
Malone, 828 F.3d 331, 337 (5th Cir. 2016) (collecting cases);
United States v. Hall, 101 F.3d 1174, 1176 (7th Cir. 1996).
Where a substance, such as XLR-11, is not listed in the Sen-
tencing Guidelines, the Guidelines require application of the
marijuana equivalent of the most closely related controlled
substance. U.S.S.G. § 2D1.1 cmt. n.6. To make this finding, the
Guidelines instruct the district court to consider (1) whether
the unlisted substance has a chemical structure substantially
similar to that of a controlled substance listed in the Guide-
lines; (2) whether the unlisted substance produces a pharma-
cological effect that is substantially similar to the effect pro-
duced by a controlled substance in the Guidelines; and (3)
whether a lesser or greater amount of the unlisted substance
is required to produce a pharmacological effect that is sub-
stantially similar to the effect produced by a controlled sub-
stance in the Guidelines. Id.
The government conceded at sentencing that none of the
analogues involved in this case had a chemical structure sim-
ilar to any controlled substance referenced in the Guidelines.
The district court acknowledged that this factor was of no
help in finding the most closely-related controlled substance
for XLR-11. However, missing the first factor does not end the
analysis. The government produced evidence on the latter
two factors. At sentencing, the government called an expert
witness and lay witnesses who testified that the substances
had similar or stronger physiological effects to those of THC.
The expert also elaborated on why the substances were like
THC, and not like marijuana, which contains THC as one of
its many chemicals. The expert also opined that less of the
substances was needed to produce an effect compared to a
given amount of THC. The district court was entitled to credit
Nos. 15-3589 & 15-3601 17
that testimony and use it to find that the substances were most
like THC. It follows that the district court did not err in ap-
plying the 1:167 marijuana equivalency ratio for THC to the
substances involved in this case, or in calculating appellants’
offense levels accordingly.
III. Conclusion
For the foregoing reasons, we AFFIRM the judgment of the
district court.