United States Court of Appeals For the First Circuit

No. 20-1311

UNITED STATES OF AMERICA,

Appellee,

v.

RAFAEL SOLER-MONTALVO,

Defendant, Appellant.

APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF PUERTO RICO

[Hon. Gustavo A. Gelpí, U.S. District Judge]

Before

Thompson and Howard, Circuit Judges, and Woodcock,* District Judge.

Andrew S. McCutcheon, Assistant Federal Public Defender, with whom Eric Alexander Vos, Federal Public Defender, Franco L. Pérez- Redondo, Assistant Federal Public Defender, Supervisor, Appeals Division, and Kevin E. Lerman, Research & Writing Specialist, were on brief, for appellant. Ross B. Goldman, with whom W. Stephen Muldrow, United States Attorney, Mariana Bauza and Ginette L. Milanés, Assistant United States Attorneys, Nicholas L. McQuaid, Acting Assistant Attorney General, and Robert A. Zink, Acting Deputy Assistant Attorney General, were on brief, for appellee.

* Of the District of Maine, sitting by designation. August 2, 2022 THOMPSON, Circuit Judge. After a four-day trial, a

federal jury in Puerto Rico found Rafael Soler-Montalvo guilty of

attempting to persuade, induce, or entice a minor to engage in

criminal sexual activity. Appealing, he flags three areas of

error. He says the evidence was constitutionally insufficient to

convict him, that the district court erred in limiting the

testimony of his expert witness, and that the prosecution engaged

in a string of misconduct that ultimately discolored the jury's

view of the trial. Finding the evidence sufficient, but the trial

tainted by the erroneous limitation of Soler's expert's testimony,

we vacate and remand for a new trial.

BACKGROUND

The story begins in March 2017. Soler was a 64-year-

old retiree from the NYPD living in Guánica, Puerto Rico. After

separating from his spouse at some point prior to 2017, he became

lonely and had trouble socializing. So he began using some dating

websites. One of the websites he used for dating was Craigslist,

specifically the "Casual Encounters" section, where people posted

personal ads.

In late March or early April 2017, one of the ads up on

Craigslist in Puerto Rico was titled: "In Mayagüez for a few

weeks." Opening up the ad that was posted as a "69" year-old woman

(the poster) seeking a man (the responder), it said: "Hey, I'm

visiting Mayagüez for a little bit, looking for a cool guy to spend

- 3 - some time with." One of the people who responded to the posting

was Soler. After that, someone responded to Soler's email saying:

"Hey, what's up can you send me a message on Kik Messenger at

JanisN666. If you're into young thin girls say hi." "Janis,"

though, was actually Special Agent Ryan Sieg from Homeland Security

Investigations, posing as a young girl.

From there, a Kik user -- later identified as Soler --

going by the username "4Real4U2Day" sent a message to Janis. And

a conversation sparked over the next days or weeks. We'll get

into much more detail later on, so we'll just give the highlights

now.

Soler introduced himself (including telling Janis his

background, where he lived, and about his children), asked Janis

questions about her experience in Puerto Rico, and asked about

where she was from. Quite early on in the conversation, Janis

told Soler that she was 13 years old. Although Soler chuckled

that remark off and said he thought Janis was joking, Janis doubled

down that she was only 13. And although Soler expressed surprise

that a 13 year old was posting on Craigslist Casual Encounters

(which requires the poster to verify they are over 18), he

acknowledged there was no way for Craigslist to verify that

information.

From there, the conversation turned sexually explicit.

Soler began asking Janis for photos of herself and making

- 4 - suggestive comments about her appearance. Eventually, Janis

shared two photographs of "herself," which were actually childhood

photos of a female law-enforcement officer used with her consent.

The photos, which clearly reflected an underage girl, generated

more suggestive comments from Soler about Janis's physical

appearance. And the conversation became very sexual, with Soler

telling Janis extensively -- and in some detail -- about the sexual

things he wanted to do to her. He even sent a sexually explicit

photo of himself.

Soler also discussed meeting up with Janis. The two

discussed logistics, including for how long Janis could get away

from her family, whether family members would be suspicious, where

they should meet, and how Janis would get there. Throughout their

conversation, Soler revealed concern that they would be caught and

repeatedly sought assurances from Janis that their meeting would

be their secret and that she wouldn't tattle to her family or the

authorities. The two ultimately agreed to meet at the Walmart at

the mall in Mayagüez. But on the day of the meet, "Janis" got

scared and refused to come outside -- though not before figuring

out what kind of car Soler was driving. After trying to convince

Janis to come outside, Soler eventually drove away, leaving Janis

with a reminder that everything should remain their secret. Soler

was soon pulled over and arrested, and the officers pulled Soler's

- 5 - cell phone -- with the messages with Janis still on it -- out of

the truck.

Following his arrest, a Puerto Rico federal grand jury

handed down an indictment charging Soler with one count of attempt

to persuade, induce, or entice a minor to engage in criminal sexual

activity, in violation of

(b). And so began the

criminal proceedings that eventually brought Soler to us. We'll

offer more detail on all the procedural history later as it becomes

relevant to the analysis. But, at a high level, here's what went

down. At the district court, Soler and the government jostled

over a host of issues before, at, and after the trial. Pre-trial,

their spars included motion practice over the admission of Soler's

proposed expert, with the district court ultimately issuing four

separate, short orders on the subject, seemingly changing its mind

(again, more on that later). After going to trial, at which Soler

testified, a jury convicted Soler of the sole count against him.

Soler moved for both a judgment of acquittal (claiming insufficient

evidence to convict) and a new trial (claiming a litany of trial

errors, including alleged evidentiary misfires, jury-instruction

errors, and prosecutorial misconduct). The district court denied

both motions, and Soler timely appealed.

- 6 - DISCUSSION

I. Sufficiency of the Evidence

We begin with Soler's argument that there was

insufficient evidence to find him guilty beyond a reasonable doubt,

and that the district court therefore should have ordered an

acquittal. See Fed. R. Crim. P. 29. Because Soler preserved his

challenge to the sufficiency of the evidence below, our review is

de novo. See United States v. Maldonado-Peña,

(1st

Cir. 2021), cert. denied sub nom. Rivera-George v. United States,

(2022). So we look at the issues with fresh eyes

and without any deference to the district court's assessment.

In testing the evidentiary sufficiency, we must

"determine whether 'any reasonable jury could find all the elements

of the crime [proven] beyond a reasonable doubt.'" United States

v. Seary-Colón,

(1st Cir. 2021) (quoting United

States v. Santos-Soto,

(1st Cir. 2015)), cert.

denied,

(2021). The question is not whether "no

verdict other than a guilty verdict could sensibly be reached,"

but only whether "the guilty verdict finds support in a plausible

rendition of the record."

(quoting United States v. Hatch,

(1st Cir. 2006)). To affirm, we need not be satisfied

that "the government succeeded in eliminating every possible

theory consistent with the defendant's innocence."

(citation omitted).

- 7 - To conduct our analysis, we review the record in the

light most favorable to the verdict. United States v. Clough,

(1st Cir. 2020). In doing so, "we do not view each

piece of evidence separately, re-weigh the evidence, or second-

guess the jury's credibility calls." Seary-Colón,

.

Rather, we "giv[e] the prosecution the benefit of all sensible

inferences and credibility choices." United States v. Cruz-Ramos,

(1st Cir. 2021). Indeed, it is not our role to

"decide 'which witness to credit,'" for we must assume that the

jury "credited those witnesses whose testimony lent support to the

verdict."

at 38 (quoting United States v. Lara,

(1st Cir. 1999)).

In the end, "[w]e will only reverse on a sufficiency

challenge if, 'after viewing the evidence and reasonable

inferences in the light most flattering to the prosecution, [we

conclude that] no rational jury could have found [the defendant]

guilty beyond a reasonable doubt.'" Seary-Colón,

(quoting United States v. Acosta-Colón,

(1st

Cir. 2013)). But if we do reverse, then that seals the deal

because we must order acquittal, which then precludes a second

trial. See Maldonado-Peña,

.

Soler was charged with attempted coercion and enticement

of a minor, in violation of

(b). To prevail, the

government had to prove, beyond a reasonable doubt, that Soler (1)

- 8 - used a facility of interstate commerce (2) to attempt to, or to

knowingly, persuade, induce, or entice (3) someone younger than

eighteen years old (4) to engage in criminal sexual activity. See

United States v. Dávila-Nieves,

(1st Cir. 2012). To

prove its attempt theory, the government had to show that Soler

"inten[ded] to commit the substantive offense" and took "a

substantial step towards its commission." United States v. Berk,

(1st Cir. 2011). As most pertinent here, the

parties agree that the government had to prove -- again, beyond a

reasonable doubt -- that Soler believed Janis was a minor.1

According to Soler, there was evidence that "undermined"

the idea that he knew he was communicating with a minor. Chiefly,

Soler points out that the encounter with Janis originated from a

website designed specifically for adults, which required that the

poster affirm she is over the age of 18. He contends that there

were points in the conversation with Janis that showed that it was

a role-play, including Janis's supposed understanding of what a

"daddy relationship" entails. And he points to his "candid"

1 We note the circuits have divided concerning whether a defendant must know the individual being persuaded, induced, or enticed was a minor. See United States v. Montijo-Maysonet,

(1st Cir. 2020) (describing the split and collecting cases). It does not appear that we have decided whether mistake of age is an available defense to a prosecution under § 2422(b). See id. But, in any event, the government itself conceded in the district court that it had to prove that Soler believed Janis was a minor. So we need not decide that question here.

- 9 - testimony in which he "categorically rejected" that he thought he

was doing anything but role-playing.

The problem with Soler's arguments, though, is that

there was plenty of other circumstantial evidence from which a

reasonable jury could have surmised that Soler believed he was

speaking with a minor. For example, in his messages with Janis,

Soler expressed surprise that someone so young posted on Craigslist

after Janis said she was 13. But Janis responded: "Ya so? Anyone

can[.]" And Soler agreed with her: "I realize that, they have no

way of knowing." Plus, Janis out and out told Soler that she was

13 years old. And though Soler replied with a snicker that he was

100 years old and later said that he thought Janis was "joking"

that she was 13, Soler continued the conversation after Janis

repeated that she was only 13. And she brought it up again, asking

Soler twice if he was "cool with [her] being 13," to which Soler

one time responded: "Why shouldn't I be? You sound very mature

[] and from your pic very nice also."

What's more, Janis sent two fake photographs of herself,

and those photos were quite clearly of an underage girl. See

United States v. Montijo-Maysonet,

(1st Cir. 2020)

(noting the jury could find that photographs depicted a minor).

Soler also reacted to the photographs, calling Janis "gorgeous"

and her (clearly underage) body "beautiful," later telling her

- 10 - that he was "admiring [her] sexyness [sic]," and even asking for

more photos on multiple occasions.

Additionally, the conversation between Soler and Janis

could also be viewed as reflecting Soler's belief that Janis was

young and sexually inexperienced. For example, Janis asked if sex

was "going to hurt[.]" Janis told Soler that she "d[oes]n[']t

really know much about sex" and that she worried that she "won[']t

be good at sex probably like older girls and [she] will feel

bad[.]" But Soler told her: "I won't let you feel you bad, your

[sic] so beautiful I will make you feel very good[.]" Janis also

expressed inexperience with condoms, oral sex, and lubricants, and

asked if certain things were "what girls do . . . . [l]ike older

girls?" Soler acknowledged Janis's lack of experience, too, asking

her if she had "ever hear[d] of doggy style" and telling her,

"[y]ou['ll] probably be nervous but I'll try to keep you calm and

happy[.]"

Soler also repeatedly expressed concern about getting

caught. He pressed again and again that everything should be kept

his and Janis's "secret," seeking assurances that she wouldn't

tattle. In arranging their meeting, Soler persistently worried

that Janis's cousin, who would have to drive Janis to the mall,

would not think anything amiss if Janis disappeared for a couple

of hours and cautioned Janis not to let anyone see her texting on

the day of the meet-up. He said it would be best if they could go

- 11 - to his home instead of a motel because "it's a lot safer here at

home. No nosey people around us." Later, he acknowledged it could

be "too risky" for them to go to the beach or a movie. And he

told Janis that once she got in the car she could "put the seat

back . . . [so] nobody can spot [her]." Soler contends that all

of this was part of the role-play. But given that the jurors knew

he was a retired detective, and just given the jurors' common

sense, see Montijo-Maysonet,

(noting "[j]urors don't

have to check 'common sense' or 'mature experiences' at the

courthouse door" (citation omitted)), they could have inferred

that Soler was saying all of this because he knew the risks

involved and needed to cover his tracks.2

Soler took even more steps that a jury could have seen

as beyond a simple role-play. Soler discussed with Janis using a

unique type of condom, which he said was only sold at a store

called "Condom World." So he told her he would call the store

near the mall where they were going to meet to see if they had any

in stock. And, in fact, Soler actually called that store like he

said he would. Soler also discussed a strawberry lubricant with

Janis, and he in fact had that same strawberry lubricant in his

2 Soler also testified that his risk concerns arose from his fear that the person who he was communicating with, and who he believed to be another adult, might release all of the personal information Soler had shared and embarrass him. Again, as with his role-play testimony, it was the jury's call on whether to believe that explanation or not.

- 12 - home. The jury reasonably could have drawn the inference that

someone wouldn't have taken those steps had it been a mere role-

play.3

To be sure, Soler testified that he never believed Janis

was a minor, he was merely role-playing (though he never clearly

defined the contours of what role-playing involved for him, whether

it was merely cyber-based or possibly in-person). And, in

explaining why he assumed it was role-play, Soler testified about

3Soler's expert (who we'll get to in more detail later) told the jury about so-called "de-masking" activities that occur from role-playing online, in which one party tries to identify who the other party they're talking to really is. So (Soler's argument goes) he went to the meet-up to try to de-mask his conversation partner, knowing full-well that a meet-up "most likely wasn't going to happen." And, apparently trying to account for the real-life steps Soler took here, his counsel seemed to insinuate at closing argument that Soler might have left open the possibility that his meet-up could end in a consensual sexual encounter, perhaps continuing the role-play in-person, with that adult. Problem is, Soler didn't testify that he was preparing for such a real-life sexual encounter. Instead, Soler testified that he tried to meet up merely because he "was extremely curious to whom [he] was speaking." And he never testified that he intended to continue his role-play fantasy in-person with another adult, instead testifying only that "going to meet up and have sex" was one of his undefined "fantasies" in the chat. Nor was Soler asked at trial why he took these real-life steps. Moreover, Soler's expert Dr. Kraft did not testify about de-masking habits leading to consensual, in-person, role-playing sexual encounters. Dr. Kraft testified only that de-masking, which "doesn't happen that often," usually leads to the two individuals "laugh[ing]" about the fantasy, or is borne out of one party's "frustration" with people misrepresenting themselves. Given Soler's and Dr. Kraft's testimonies, the jury could have reasonably rejected Soler's closing-argument insinuation that he was preparing for a consensual adult sexual encounter with the steps he actually took outside of the role-play chat.

- 13 - holes he saw in Janis's story that he said led him to believe she

was not actually the 13-year-old girl she claimed to be. Soler

noted that it made no sense that a girl that young would have been

on a three-to-four-week April school vacation, and that it was odd

that she was unfamiliar with the geography of Virginia even though

that's where she said she lived.

Ultimately though, it was up to the jury to decide

whether to credit Soler's testimony in the face of this ample

evidence questioning Soler's claimed belief that he was only role-

playing with a consenting adult. "[S]ifting through conflicting

testimony and determining where the truth lies is the sort of work

that falls squarely within the jury's province, not ours." Cruz-

Ramos,

(cleaned up) (quoting United States v.

Nascimento,

(1st Cir. 2007)). And the jury "was,

of course, free to reject [Soler's] defense" that "he had no

interest in actually having sex with [a] minor[], but was instead

engaging in a form of 'role playing.'" Berk,

n.8;

see also United States v. Gomez-Villamizar,

(1st

Cir. 1992) (noting the jury is not obligated to credit a

defendant's testimony that she didn't know there was cocaine in

the suitcase); United States v. Ocampo-Guarin,

(1st Cir. 1992).

Nor was the evidence, as Soler contends, in equipoise

concerning his guilt and innocence. It is true that we "must

- 14 - reverse a conviction on the grounds of evidentiary insufficiency

where an equal or nearly equal theory of guilt and a theory of

innocence is supported by the evidence viewed in the light most

favorable to the verdict." United States v. Rodríguez-Martinez,

(1st Cir. 2015) (quoting United States v.

Woodward,

(1st Cir. 1998)). But "this equal-

evidence rule takes hold only after [we] ha[ve] drawn all

reasonable inferences in favor of the verdict." United States v.

Simon,

(1st Cir. 2021) (emphasis in original)

(quoting Magraw v. Roden,

(1st Cir. 2014)), cert.

denied,

,

(2022); see also

Rodríguez-Martinez,

(explaining the equipoise rule

comes into play where "[w]ithout additional circumstantial

evidence from which the jury could rationally infer that one was

more supportable than the other, . . . [the] evidence . . . permits

two equally plausible inferences"). Here, to find equipoise, we

would have to credit Soler's testimony inconsistent with the

verdict. That we cannot do, especially given the plentiful

evidence from which the jury could reasonably infer that Soler's

explanation was not credible.

The district court did not err in denying Soler's motion

for a judgment of acquittal.

- 15 - II. Exclusion of Dr. Kraft's Testimony

Soler contends the district court erroneously excluded

part of the testimony of his proposed expert, Dr. Chris Kraft. He

accordingly demands a new trial.

Dr. Kraft, who holds a Ph.D. in clinical psychology, is

a Board-certified psychologist working at the Johns Hopkins

University specializing in internet sexual behaviors. Before

trial, Soler proffered Dr. Kraft to opine on two subjects. First,

Dr. Kraft would (and ultimately did) testify on the psychology

behind internet communication and messaging, and specifically the

frequency of role-play, imagination, or exaggeration in online

messaging. Second (and this one is important for us here), Soler

wanted him to testify about "the difference between a desire to

actually engage in sexual activity with a minor and mere fantasy

and role-playing related to sexual contact with children." As

part of that category of testimony, Soler proffered that Dr. Kraft

would testify that "the chats here . . . may be consistent with

the conduct and behavior of consenting adults using forums for

fun, play, fantasy and cybersex," but they are not consistent with

the patterns of child sexual predators.

A. Preservation

On appeal, Soler complains that the district court,

siding with the government's pre-trial motion, precluded Dr. Kraft

from testifying on the second subject, the modus operandi of

- 16 - predators and how his behavior differed. But the government throws

in a new appellate wrinkle: It now says that Dr. Kraft's testimony

was not at all limited by the district court as Soler claims, and

since no such limitation was imposed upon him, the government

declines on appeal to speculatively respond as to why Soler chose

not to probe the issue at trial.

To evaluate the government's claim, a brief (but

detailed) procedural recap is in order so the reader can follow

our analysis that is to come. Initially, the district court

granted the government's request to exclude Dr. Kraft's testimony

in its entirety. In a six-sentence order, the court explained

that Dr. Kraft's testimony "ultimately . . . involves the

defendant's state of mind." As the court understood the proffer,

"Dr. Kraft . . . would review the evidence and conclude that the

same is probative of a desire to engage in fantasy role playing

with children rather than in actual sexual contact," further

opining that Soler "is not a typical internet sexual predator . . .

[and] possesses none of the characteristics of one." Thus, the

district court said, the testimony was not proper.

Unhappy, Soler moved for reconsideration of that ruling.

In his motion, Soler explained that the district court did not

recognize the two distinct areas of Dr. Kraft's proposed testimony:

(1) testimony concerning "psychology of the internet, internet

messaging, fantasy[,] and role-play"; and (2) testimony about the

- 17 - patterns of sexual predators and analysis that Soler's actions

here do not line up with those patterns.

Denying the motion for reconsideration, the court

stated: "The Court notes that in the Joseph case from the Second

Circuit, defendant himself testified and that the excluded

evidence would have been relevant to defendant's credibility.

Here, the proffered evidence seeks to be presented in a vacu[u]m

and not in addition to any other evidence." But less than two

hours later the district court walked things back. It amended its

reconsideration order to add that "[s]hould [Soler] testify at

trial, following said testimony the Court will allow the testimony

of Dr. Kraft. See US v Joseph,

(2nd Cir. 2008)."

Finally, when the time came for Dr. Kraft to testify at

trial, the district court added another note: "[Y]ou have to

remember the expert is just going to testify in general terms.

He's not going to testify -- he hasn't interviewed him [meaning

Soler], he hasn't -- he's not going to come to a conclusion about

him. He can't talk about him."

Staring at these four terse, quite ambiguous orders,

we're left to piece together whether the district court's rulings

meant that Dr. Kraft could not testify about the patterns of sexual

predators (i.e., the typical-predator testimony) and/or how

Soler's actions did not fit those patterns (i.e., the not-a-

typical-predator testimony). Reviewing the record, we think Soler

- 18 - was correct to understand the court's orders as precluding Dr.

Kraft from offering his not-a-typical-predator testimony.4

First, we think the district court's mid-trial

qualification made clear that it was, in fact, limiting Dr. Kraft's

ability to testify concerning Soler's actions in this case.5

Contrary to the government's appellate position, Soler did not

have free rein from the district court.6 And the government does

4 At oral argument, Soler was not clear whether he believed he was also precluded from introducing Dr. Kraft's opinion on the general pattern of sexual predators, separate from his analysis of how that pattern lined up with the facts here. His appellate briefing seems to make clear that he treated the typical-predator testimony and not-a-typical-predator testimony as one cohesive unit, such that he would not introduce one segment without the other. And his post-trial briefing did not suggest that he thought the typical-predator portion of the testimony was precluded -- only the not-a-typical-predator portion. We thus do not understand Soler to argue that he was precluded from introducing the limited segment of the testimony concerning the general pattern of sexual predators. Yet it was completely reasonable for Soler to treat Dr. Kraft's testimony as one integrated unit composed of both the typical-predator profile and the not-a-typical-predator analysis. The two were inextricably intertwined, and offering only the profile of an online predator, without the ability to demonstrate why he didn't fit that profile, would have been extremely prejudicial to Soler. 5 The government, in its brief, fails to recognize that the district court ordered mid-trial that Dr. Kraft "can't talk about [Soler]." Instead, it claims that the district court's "final word" on the scope of Dr. Kraft's testimony was the amended order on the reconsideration motion, even though Soler's post-trial briefing pointed specifically to -- and quoted directly from -- the court's mid-trial order on the subject. 6 The district court's citation to Joseph also further suggests that it "w[ould] allow the testimony of Dr. Kraft" only on the first category of evidence: the psychology of internet fantasy and role-playing. Indeed, Joseph involved only expert evidence of that category. See 542 F.3d at 21–22. It did not

- 19 - not explain how Soler could have, consistent with that mid-trial

order, elicited testimony about how his actions did not fit a

predator's m.o. if Dr. Kraft had to "testify in general terms" and

"[could]n't talk about [Soler]." Testifying about Soler's actions

would, of course, have violated that limitation.

Second, the only apparent confusion about whether the

not-a-typical-predator testimony was excluded has come from the

government's appellate counsel. Indeed, the government's trial

counsel below agreed in its briefing on Soler's new-trial motion

that the district court had excluded Dr. Kraft's testimony

concerning whether Soler's actions fit the pattern of an internet

predator. As the government put it below: "[T]he Court properly

determined that the proposed testimony was irrelevant and, even if

relevant, it should be excluded." That was so, said the

government, because "Dr. Kraft never evaluated, examined, or

treated Soler so he could not . . . try to draw a comparison [from

Soler] to other types of sexual predators he may have treated or

studied."

involve proposed expert testimony concerning the patterns of sexual predators or whether the defendant in that case's actions fit the bill of a typical predator. See

If the district court's re-reconsideration order was meant to swing open the door to expert testimony concerning how Soler's actions lined up with those typically expected of a predator, then Joseph would be an odd citation.

- 20 - To tie a bow on it, the district court, post-trial,

agreed with everyone's understanding of its orders. In denying

Soler's motion for a new trial, the district court explained:

"[T]he exclusion of part of Dr. Kraft's expert testimony was

warranted. It is undisputed that he did not evaluate, interview,

meet[,] nor treat [Soler]. . . . Thus, Dr. Kraft could not testify

as to [Soler]'s state of mind." In fact, prior to ruling on the

new-trial motion, the district court ordered Soler to identify

where in the record each of his new-trial arguments was preserved

by timely objection. Responding, Soler pointed to both the pre-

trial orders and the court's mid-trial can't-talk-about-him order,

after which the district court did not correct anyone's

understanding of those orders.7

7 At oral argument, the government argued that the district court's in-trial statement was merely a "rehash" of the agreement Soler made that he would not ask Dr. Kraft to testify directly about his state of mind. But Soler thrice explained that he would not ask Dr. Kraft to testify so directly, rather he would ask Dr. Kraft to testify only as to a comparison of the activities here to the modus operandi of sexual predators. Yet the district court's post-trial ruling apparently concluded that the proffered testimony, even considering that self-imposed limitation, would still have violated Rule 704(b), and the court thus concluded it did not err in imposing the limitation about which Soler complained. That makes the government's reading highly implausible. To buy it, we would have to assume that the district court either ignored or repeatedly misunderstood Soler's explanation -- and also the government's similar understanding -- that the issue presented was whether Dr. Kraft could testify as to whether Soler's actions fit the pattern of a typical predator without testifying directly about Soler's state of mind.

- 21 - Given the parties' -- and the district court's -- clear

understanding below of what was and wasn't excluded, we conclude,

contrary to the government's contention here, that Soler is

correct: The district court precluded Dr. Kraft's proposed not-

a-typical-predator testimony. And so the evidentiary ruling is

properly before us.

B. Merits

We turn now to examining if the district court was

correct in barring Dr. Kraft from testifying about whether Soler's

actions fit the mold of a sexual predator. In excluding the

evidence, the district court appeared to base its decision on two

rules of evidence: Rule 704(b) and Rule 403. We take each in

turn.

As we do so, we bear in mind that evidentiary calls are

generally left to the district court's discretion. See Maldonado-

Peña,

; United States v. Galíndez,

(1st Cir. 2021). Still, abuse-of-discretion review is not

toothless. Jodoin v. Toyota Motor Corp.,

(1st

Cir. 2002). A district court abuses its discretion "when a

relevant factor deserving of significant weight is overlooked, or

when an improper factor is accorded significant weight, or when

the court considers the appropriate mix of factors, but commits a

palpable error of judgment in calibrating the decisional scales."

- 22 - United States v. Taylor,

(1st Cir. 2017) (quoting

United States v. Jiménez,

(1st Cir. 2005)).

1. Rule 704(b)

Admission of expert testimony is governed by Federal

Rule of Evidence 702. See Montijo-Maysonet,

. But

that's not the only hurdle expert testimony must clear. Expert

testimony is still subject to the rigors of other rules, including

in criminal cases Rule 704(b). That rule provides that "an expert

witness must not state an opinion about whether the defendant did

or did not have a mental state or condition that constitutes an

element of the crime charged or of a defense," since "[t]hose

matters are for the trier of fact alone." Fed. R. Evid. 704(b);

see United States v. Valle,

(1st Cir. 1995)

(recounting the history of Rule 704(b)).

Applying Rule 704(b), we have held time and again that

although "Rule 704(b) bars a witness from characterizing the

defendant's intent, . . . it 'does not . . . apply to predicate

facts from which a jury might infer such intent.'" United States

v. Henry,

(1st Cir. 2017) (quoting United States v.

Peña-Santo,

(1st Cir. 2015)); see Valle,

; United States v. Lamattina,

(1st

Cir. 1989).

Identifying where the line lies in the sand, we have

held that "a qualified expert does not violate Rule 704(b) by

- 23 - expressing an opinion as to whether predicate facts are consistent

with drug distribution rather than mere possession." Henry,

. Accordingly, expert testimony that "explained that

the quantity of crack found at the search site was consistent with

distribution, as opposed to personal use" in a case concerning

intent to distribute drugs did not violate Rule 704(b). Valle,

. Nor did an expert's opinion that the drugs at issue

in a similar case were "packaged for sale" since "the expert

grounded his opinion that the drugs were packaged for sale on his

general knowledge of criminal practices and the circumstantial

evidence bearing on the issue of intent that was produced during

the trial." Henry,

. Similarly, an expert's

testimony that "recorded conversations involved loansharking" did

not cross the line into testimony on the defendant's state of mind,

but rather only "may have provided the jury with some basis for an

inference as to defendant's state of mind." Lamattina,

(emphasis added). In sum, we have held repeatedly that

testimony that a defendant's actions are consistent with the modus

operandi of illegal activity, though allowing the jury to infer

the defendant's state of mind, does not violate Rule 704(b)'s

ultimate-issue prohibition.

Dr. Kraft's not-a-typical-predator testimony fits

comfortably within that rubric. Dr. Kraft would have compared the

conversations and surrounding circumstances of this case to the

- 24 - patterns of online predators and identified inconsistencies, thus

suggesting that Soler's actions did not accord with those of a

typical predator. We see no distinction between that testimony

and a government-offered expert's testimony that the manner in

which drugs were packaged were consistent with the m.o. of drug

distributors where the issue was whether the defendant intended to

distribute drugs. See Henry,

; Valle,

. Nor is there any daylight between the testimony here and the

testimony in Lamattina concerning whether the recorded

conversations "involved loansharking" in a prosecution for

loansharking.

. The only difference is that the

testimony here was that Soler's actions were not consistent with

illegal activity.

In this precise scenario, one of our sister circuits has

drawn the same conclusion, just on the side of the government. In

United States v. Romero,

(7th Cir. 1999), the Seventh

Circuit faced a defendant's challenge to a government-proffered

expert who testified that "hypothetical" facts mirroring the

defendant's actions were consistent with the patterns of child

molesters,

at 585–86. The defendant there was charged with

various counts related to his transportation of a minor with the

intent to engage in criminal sexual activity.

.

Objecting that the expert's analysis of the supposed hypotheticals

violated Rule 704(b), the defendant argued that the testimony was

- 25 - tantamount to an opinion on his intent to molest the victim.

. The Seventh Circuit rejected that argument.

Relying

on case law in that circuit similar to ours in Henry and Valle

concerning drug distribution, the court said that the testimony

passed Rule 704(b) comfortably since the expert "never directly

opined as to [the defendant's] mental state."

Nor did his

explanation of "what types of actions might distinguish the actual

molester from the mere collector of child pornography" violate

Rule 704(b), even where the defendant admitted an interest in child

pornography but disclaimed any intent to act on that interest.

Id.; see

.

So too here. Dr. Kraft's not-a-typical-predator

testimony, as proffered by Soler, did not cross the line into the

territory of Rule 704(b). Rather, his testimony was limited only

to whether certain facts were consistent with the pattern typically

seen with individuals who were interested in having sex with

minors. Just as an expert may opine on whether the packaging of

drugs is consistent with distribution, Dr. Kraft could have

permissibly testified about whether Soler's actions here were

consistent with role-play rather than the pattern of those who

entice minors into sexual activity. The district court abused its

discretion in precluding this testimony under Rule 704(b).

- 26 - 2. Rule 403

The district court also appeared to conclude that Dr.

Kraft's not-a-typical-predator testimony should be excluded under

Rule 403, noting (without citing the rule) that "any probative

value of the [testimony] is substantially outweighed by confusion

of the issue at bar, to wit, [Soler]'s state of mind."8 Rule 403

allows a court to "exclude relevant evidence if its probative value

is substantially outweighed by," among other reasons, "a danger of

. . . unfair prejudice, confusing the issues, [or] misleading the

jury . . . ." Fed. R. Evid. 403.

Even though Dr. Kraft's testimony passed the strictures

of Rule 702's gatekeeping of expert evidence, it was still subject

to the Rule 403 balancing test. See United States v. Pires,

(1st Cir. 2011). As we have explained before, there is

a particular worry with expert testimony that "jurors may assign

more weight to [it] than it deserves."

Expert testimony "can

8The district court's reliance on Rule 403 is further muddled by its subsequent orders on the issue. In denying Soler's motion for reconsideration, it did cite Rule 403, but not for the same reason it suggested in its initial order. Rather, the reconsideration order spotted a risk of prejudice or confusion in the fact that the evidence would be presented in a vacuum and not to support the credibility of Soler's testimony, since Soler to that point had not said that he planned to testify in his own defense. Of course, Soler ultimately testified. And later on when denying Soler's motion for a new trial, the district court did not again recite its "substantially outweighed" language, but merely said that Dr. Kraft's testimony would have offered an opinion on Soler's state of mind, thus signaling reliance on Rule 704(b) -- not Rule 403.

- 27 - carry with it an unwarranted 'aura of special reliability and

trustworthiness,'"

(quoting United States v. Fosher,

(1st Cir. 1979)), so "courts must guard against letting

it intrude in areas that jurors, by dint of common experience, are

uniquely competent to judge without the aid of experts,"

"Because of this risk, the judge in weighing possible prejudice

against probative force under Rule 403 . . . exercises more

control over experts than over lay witnesses." Daubert v. Merrell

Dow Pharms., Inc.,

(1993) (quoting Jack B.

Weinstein, Rule 702 of the Federal Rules of Evidence Is Sound; It

Should Not Be Amended,

, 632 (1991)).

We review district courts' Rule 403 rulings under the

abuse-of-discretion framework detailed above, with a few

additional notes. We have said specifically to Rule 403 that

district courts have "wide discretion in steadying the . . .

seesaw." Pires,

. Reflecting our deference for the

district court's battlefield judgment, it is only in the rarest

and most compelling cases that "'we, from the vista of a cold

appellate record,' [will] reject a judge's on-the-scene Rule 403

ruling." United States v. Rodríguez-Soler,

(1st

Cir. 2014) (quoting DiRico v. City of Quincy,

(1st Cir. 2005)).

Still, the standard for exclusion under Rule 403 is a

high one. See United States v. Fahey,

(1st Cir.

- 28 - 1985). When conducting the Rule 403 balancing test, courts must

heed that the default rule is that relevant evidence will be

admitted. See United States v. Jones,

(1st Cir.

2012). And it is not enough for Rule 403 that the evidence's

dangers of unfair prejudice or confusion somewhat outweigh the

probative value of the evidence. See United States v. Trenkler,

(1st Cir. 1995) (concluding prior-bad-acts evidence

was admissible under Rule 403 even though there was "some danger"

that the jury would use it to assume a defendant's propensity for

criminal behavior). Rule 403 permits exclusion not when the

evidence is merely outweighed by the dangers of its admission, but

only when it is "substantially outweighed." See Fed. R. Evid. 403

(emphasis added).

With that framework in mind, and mindful of our bounded

review, we think the district court -- to the extent it relied on

Rule 403 in excluding Dr. Kraft's testimony -- erred in determining

both the probative value of the evidence and, relatedly, that the

probative value was substantially outweighed by a risk of confusing

or misleading the jury.

Probative value. Dr. Kraft's not-a-typical-predator

testimony was highly relevant to and probative of the key issue in

the case. Soler's defense effectively revolved around his argument

that he was merely role-playing and did not actually believe Janis

was a minor. Dr. Kraft's excluded testimony bore directly on the

- 29 - credibility of Soler's testimony concerning his state of mind and

sought to explain (just in the converse of oft-admitted government-

expert testimony) how seemingly sinister conduct could be part of

innocent sexual fantasy. Cf. United States v. Long,

,

666–68 (D.C. Cir. 2003) (finding government expert's child-

molester m.o. testimony probative under Rule 403 because it helped

show "how seemingly innocent conduct could be part of a seduction

technique" (cleaned up)).

The government's arguments made below (again, only below

because the government does not defend, even in the alternative,

the district court's exclusion on appeal since it thinks there was

no exclusion to begin with) do not demonstrate the testimony's

minimal probative value. First, although the jury may have been

familiar with "sexting" and sexual role-play fantasies, we find it

hard to believe that the jury would be familiar with prototypical

grooming behavior by an individual seeking out sex with minors and

thus how that behavior compared to Soler's.9

Second, we disagree that the testimony's relevance was

minimal because Soler "was not accused of being a pedophile or

child molester and he was not charged with possession of child

9 Nor do we have any information to confirm this speculation about the jurors' knowledge of sexting, sexual roleplay, or grooming behaviors. Although Soler proposed certain voir dire questions that may have revealed some jurors' familiarity with these concepts, the district court declined to ask those questions at jury selection.

- 30 - pornography." The government charged Soler with attempting to

entice a minor to engage in unlawful sexual activity. Testimony

that Soler's actions were inconsistent with the typical m.o. of

one attempting to entice a minor -- as opposed to engaging in role-

play with a consenting adult -- is highly relevant to that charge.

Moreover, Dr. Kraft was not proffered to testify as to whether

Soler was a pedophile or a child molester. He was proffered to

testify only as to whether Soler's actions were consistent with

patterns of known pedophiles or child molesters, similar to

testimony on whether a defendant's actions are consistent with the

patterns of other drug dealers in a case charging intentional

distribution of drugs.

The testimony was also relevant to respond to the

government's attempt to show that Soler acted like a typical

predator in an effort to demonstrate his knowledge. At trial, the

government elicited testimony from Agent Sieg about his use of

codewords in the Craigslist post.10 It then asked Agent Sieg:

"[I]n your experience, who are the people that generally use code

words[?]" To which Agent Sieg responded: "Well, anyone wanting

to keep something secret or their intentions masked." The

The codewords Agent Sieg referred to were "young," "thin," 10

and "girls" in his initial email to Soler saying: "If you're into young thin girls say hi." Sieg testified that he used those words because they "are code words among those people who would seek to engage in illegal sexual activity with minors."

- 31 - government, of course, was trying to suggest that since Soler

responded to codewords like others who actually intended to keep

their intentions masked (which includes those who try to engage in

sex with minors), the jury should not buy Soler's feigned innocence

and role-play defense. Dr. Kraft's testimony would have done the

same thing, just delivering the opposite inference.

Thus, to the extent the district court found Dr. Kraft's

not-a-typical-predator testimony not probative, we believe it

erred. See Rubert-Torres ex rel. Cintron-Rupert v. Hosp. San

Pablo, Inc.,

(1st Cir. 2000) (finding abuse of

discretion in assessment of probative value where we concluded the

evidence was "highly relevant" to a "central issue" in the case);

Espeaignnette v. Gene Tierney Co.,

(1st Cir. 1994)

(similar). The testimony was, in fact, highly relevant to and

probative of the central issue in this case: Soler's belief that

Janis was a minor.11

11 Our decision in Pires is not to the contrary. There, a defendant sought to introduce expert evidence that he did not have any mental illnesses, sexual deviances, or a prurient interest in children as evidence that he lacked a motive to possess child pornography.

. Yet we said that evidence was of "diminished relevance" because the only question in that case was whether the defendant knowingly received and possessed the child pornography, not whether he received and possessed the images related to a sexual interest in children.

at 10–12. Here, though, Dr. Kraft's testimony was highly probative of whether Soler's actions were consistent with role-playing or typical grooming behavior. The evidence was thus not "peripheral" as it was in Pires. See

.

- 32 - Balancing act. Given the high probative value of the

evidence, we expect the countervailing interests weighing against

the admission to be great to exclude the evidence. "When proffered

evidence relates to the central issue in a case, it is a difficult

matter indeed to show that the prejudicial effect of that evidence

substantially outweighs its highly probative nature, as Rule 403

requires." Rubert-Torres,

.

But the district court's reasoning, once corrected, is

not so weighty. Critically, the district court's assessment of

the danger of confusion was bound up in its erroneous analysis of

whether Dr. Kraft would opine on an ultimate issue in the case.

As the district court put it, "any probative value of the

[testimony] is substantially outweighed by confusion of the issue

at bar, to wit, [Soler]'s state of mind." Yet as we've explained,

Dr. Kraft's testimony would not opine on Soler's state of mind --

it would only offer evidence to the jury from which it could infer

that Soler did not believe Janis was 13 given that he did not

operate like a typical predator. Thus, we think the potential

danger of confusion posed by the testimony was substantially

limited when viewed in its proper light.12

12Nor were the district court's concerns about Dr. Kraft's testimony being presented in an evidentiary "vacuum" still relevant. At trial, Soler testified and spoke directly to his own state of mind. As the district court admitted, Dr. Kraft's testimony could be probative of whether Soler's testimony was credible, if he chose to testify.

- 33 - To be sure, we share the district court's concern that

expert testimony brushing up against -- but not directly touching

on -- the defendant's state of mind could weigh heavily on the

jurors' minds. We have often worried that jurors may be smitten

by an expert's "aura of special reliability and trustworthiness"

and therefore "assign more weight to expert testimony than it

deserves." E.g., Pires,

.

Yet even recognizing that risk, there are "less

restrictive means to minimize the prejudice than entirely

excluding" testimony. Rubert-Torres,

("Because

the Federal Rules of Evidence favor the admissibility of evidence,

less intrusive measures to minimizing the prejudicial effect of

evidence are preferred to excluding evidence."). For example, the

district court instructed the jury that it should not deem Dr.

Kraft credible merely because he is an expert. See United States

v. Encarnacion,

(1st Cir. 2022) (noting that

careful jury instructions on the weight to be given to an expert's

testimony help mitigate risks of unfair prejudice from the expert's

special stature). Rather, the district court told the jury that

credibility calls should be made the same way whether the jury was

judging Soler, Dr. Kraft, or Agent Sieg. The district court also

could have crafted a specific jury instruction that Dr. Kraft's

testimony should not be interpreted as an opinion on whether Soler

knew that Janis was a minor, thus blunting the potential danger.

- 34 - See United States v. Morris,

(7th Cir. 2009)

(noting that limiting instructions, as well as careful

construction of the examinations, can help alleviate risks of

confusion from expert-criminal-m.o. testimony). And it could have

-- as it did on Dr. Kraft's first subject of testimony -- given

the government an adequate opportunity to cross-examine Dr.

Kraft's not-a-typical-predator testimony. See Encarnacion,

.

* * *

Though the district court had "wide discretion in

steadying the Rule 403 seesaw," Pires,

, it still

needed to explain why the highly probative value of the evidence

was so "substantially outweighed" by the risks of confusing or

misleading the jury that the testimony should have been excluded

wholesale, Fed. R. Evid. 403; see Rubert-Torres,

.

And the need was more pronounced considering the court's allowance

of Agent Sieg's testimony about the use of codewords. The district

court failed to do so, thus creating the rare and most-compelling

case in which we conclude -- and do not do so lightly -- that the

district court abused its discretion in excluding this testimony

under Rule 403, too.

3. Harmlessness

Even if we find error in the district court's evidentiary

ruling, we must affirm if we find the error nonetheless harmless.

- 35 - See Rodríguez-Soler,

. Non-constitutional errors

in admitting or excluding evidence are harmless "unless the

evidence 'likely affected' the trial's outcome." United States v.

Correa-Osorio,

(1st Cir. 2015) (quoting United

States v. Landrón-Class,

(1st Cir. 2012)). Thus,

we will not reverse "if we can say with fair assurance, after

pondering all that happened without stripping the erroneous action

from the whole, that the errors did not substantially sway the

jury's verdict." United States v. Rivera-Carrasquillo,

(1st Cir. 2019) (cleaned up).

We're not so assured here. At the end of the day, the

key issue in this case was whether Soler believed that Janis was

13. And though there was substantial evidence from which the jury

could have concluded that he did, Soler's role-play defense turned

almost exclusively on the credibility of his testimony. Dr.

Kraft's not-a-typical-predator testimony could have provided

substantial support to Soler's credibility in claiming that he

truly believed Janis was just an assumed identity of another

consenting adult. Whether the jury ultimately buys that story is

for it to decide. But we could only speculate as to whether the

additional evidence could have swayed the jury's credibility

determination on that key element, leaving the error not

- 36 - harmless.13 See Rubert-Torres,

("On such a central

issue as causation, . . . it would be speculation at best, much

less with fair assurance, to say that the jury verdict was

unaffected by the error."); United States v. Ouimette,

(1st Cir. 1985) (error not harmless where the excluded

evidence went to "the core of the defendant's case"); see also

United States v. Kilmartin,

(1st Cir. 2019)

(finding erroneously admitted testimony not harmless to the jury's

assessment of the defendant's intent since "[i]ntent is inherently

difficult to demonstrate" and though there was ample

circumstantial evidence of intent, the defendant offered

alternative innocent explanations).14

CONCLUSION

For these reasons, we vacate the judgment of the district

court and remand for a new trial.

The government suggests that the testimony Dr. Kraft 13

actually gave was good enough since the "basic thrust" was to suggest that some people merely role-play without acting on it. Yet the government glosses over the difference between Dr. Kraft's proposed role-play-based testimony and not-a-typical-predator- based testimony, which were two distinct categories of testimony. The latter builds implicitly on the first (i.e., there is a difference between a true predator and a mere role-player), but the government has not directed us to anywhere in the record that Dr. Kraft actually testified that Soler's actions were inconsistent with the modus operandi of child sex predators. Given our conclusion, we need not address whether a new 14

trial is required because of alleged prosecutorial misconduct -- though we do caution the government to take greater care in its in-court statements before the jury on remand.

- 37 - - CONCURRING OPINION FOLLOWS -

- 38 - WOODCOCK, District Judge, concurring. I write

separately because I am not convinced that the record establishes

that Soler properly preserved at the trial court the error he

presents here as required by Federal Rule of Evidence 103 or that

the district court ever excluded the expert's evidence about the

characteristics of sexual predators and their typical patterns of

behavior and use of the internet.

Although the district judge changed his mind about the

admissibility of Dr. Kraft's testimony in prior orders, his final

ruling just before trial was:

The order at Docket No. 95 is hereby amended as follows. Should defendant testify at trial, following said testimony the Court will allow the testimony of Dr. Kraft. See US v Joseph,

(2nd Cir. 2008).

Nothing in this final order intimated that the defendant would not

have been allowed to present Dr. Kraft's full testimony as Soler

proposed it, including his testimony about sexual predators and

the internet. I am unable to conclude that the district judge

would have disallowed Dr. Kraft's testimony on sexual predation

because, following this order, Soler never moved its admission and

never received a definitive ruling against its admission.

Tracking the way this issue developed at trial, I see

the district judge's first concern turning on whether Dr. Kraft

would be allowed to testify if Soler did not take the stand, an

offshoot of the government's argument that Dr. Kraft's opinion

- 39 - testimony should not be a backdoor way of admitting Soler's

testimony. In my view, the district judge was rightfully concerned

about the prospect of Dr. Kraft testifying about Soler and then

Soler declining to testify, leaving Dr. Kraft's testimony - to the

extent it relied upon Soler's testimony - being without foundation.

Once defense counsel informed the jury in the opening statement

that Soler intended to take the stand, the district judge's concern

on this point was alleviated, but the district judge still had the

obligation to enforce Federal Rule of Evidence 704(b), which

prohibited Dr. Kraft from expressing an opinion "about whether the

defendant did or did not have a mental state or condition that

constitutes an element of the crime charged or a defense." Fed.

R. Evid. 704(b).

I see Soler's argument that he preserved this

evidentiary issue by giving the prior Notice of Expert as falling

flat. It is true that a definitive ruling on a pretrial motion in

limine may preserve an objection to an evidentiary ruling without

requiring the party to renew the objection at trial. Fed. R. Evid.

103(b); United States v. Grullon,

, 30 (1st Cir. 2021);

Zachar v. Lee,

(1st Cir. 2004) ("The 2000 Amendment

to Federal Rule of Evidence 103 specifically provides that once

the district court 'makes a definitive ruling on the record

admitting . . . evidence, either at or before trial, a party need

not renew an objection . . . to preserve a claim of error for

- 40 - appeal.'" (quoting Fed. R. Evid. 103(a)(2), now denominated as

Fed. R. Evid 103(b))). The problem here is that, as just noted,

the district judge's last ruling before trial allowed the expert's

testimony without qualification. As a result, in my view, the

defendant's motion in limine and Notice of Expert did not preserve

this issue on appeal because the district court's superseding

ruling allowed the expert's testimony to come in at trial (under

the condition that Soler testify, which he did).

It appears to me that the defense confused the district

judge's repeated rulings about Federal Rule of Evidence 704(b)

with a ruling excluding the expert's testimony on sexual predation.

The line between proper and improper expert testimony in this area

is subtle and it strikes me that the district judge's concern that

the expert not "talk about [Soler]" was a reminder that Rule 704(b)

prohibits an expert from expressing opinions about a defendant's

state of mind. My view is buttressed by the fact that the district

judge emphasized both during trial and in the post-trial order

that Dr. Kraft never interviewed Soler and therefore could not

testify about him.

To the extent that the defense was confused about the

district judge's rulings and the line between proper and improper

expert testimony, the burden was on defense counsel to obtain a

definitive ruling and, if the evidence were disallowed, to make an

offer of proof. Fed. R. Evid. 103(a)(2) ("[I]f the ruling excludes

- 41 - evidence, a party informs the court of its substance by an offer

of proof, unless the substance was apparent from the context.");

see Kelley v. Airborne Freight Corp.,

(1st Cir.

1998); Earle v. Benoit,

(1st Cir. 1988) ("[I]t

is precisely for such a situation, where a court refuses to receive

evidence and yet the same is needed to elucidate proponent's claim

for admissibility, that the offer of proof device exists.").

We do not know what would have happened at trial if

defense counsel had complied with Federal Rule of Evidence 103,

approached sidebar, cited United States v. Romero,

(7th Cir. 1999), and informed the judge that the defense would

like to explore Dr. Kraft's expert views about the significance of

already admitted evidence, namely the online chats, and to

elucidate Dr. Kraft's views of the hallmarks of an online predator,

such as the presence of collections of child pornography, the

grooming of children, and employment or activities that involve

children, and how those hallmarks compared to Soler's background

and behavior as already set forth on the record or to be supplied

by Soler's later testimony. The defense could have reminded the

judge that none of this proffered evidence would have suggested

what the defendant himself thought, only what Dr. Kraft in his

expert opinion would have expected to find in a case of online

sexual predation and that any opinion as to how Soler's behavior

compared with typical sexual predators was based on the evidence

- 42 - presented at trial and not a clinical examination. If the defense

had done so, we would know for certain whether the trial judge

would have excluded all, part, or none of Dr. Kraft's proposed and

unpresented testimony. But we do not know how the district judge

would have ruled because, after the judge ruled all of Dr. Kraft's

testimony admissible, the district judge was never squarely asked

to make a definitive ruling excluding any portion of it.

Finally, my reading of the record is further supported

by the fact that on cross-examination of Dr. Kraft, the district

judge allowed the government, over the defense's objection, to

extensively discuss sexual predation. Even after the government

elicited this testimony, the defense made no effort to present Dr.

Kraft's sexual predation testimony on the ground that the

government had opened the door. Instead, the defense asked no

further questions of Dr. Kraft.

Before we assign error to a district court's evidentiary

ruling, we should be clear that the trial court actually excluded

the admissible evidence and typically the burden is on the losing

party to demonstrate that he clarified his objection and obtained

a definitive ruling excluding the evidence. See Grullon, 996 F.3d

at 30-31 ("[W]hen a judge issues a preliminary, conditional, or

'tentative' ruling that 'clearly invites the party to offer the

evidence at trial,' then the party has an obligation to raise it

again to preserve the claim." (quoting United States v. Almeida,

- 43 -

(1st Cir. 2014)); United States v. Takesian,

(1st Cir. 2019) ("Rule 103 requires the objecting

party . . . 'to clarify whether an in limine or other evidentiary

ruling is definitive when there is doubt on that point.'" (quoting

Crowe v. Bolduc,

(1st Cir. 2003))).

Despite my reservations about whether the error was

preserved and whether the district judge committed any error, I am

concurring, not dissenting, because after the trial, the defendant

and the government both treated Dr. Kraft's evidence on sexual

predation as having been excluded. In its response to the

defendant's claim in his post-trial briefing that Dr. Kraft's

sexual predation testimony should not have been excluded, the

government did not claim that the district judge never excluded

it. To the contrary, in its post-trial memorandum, the government

argued:

Accordingly, the Court properly determined that the proposed testimony was irrelevant and, even if relevant, it should be excluded because its probative value was far outweighed by the potential for confusing and misleading the jury, and unnecessarily protracting the trial.

In contrast, the trial judge's post-trial ruling does

not appear to be based on the parties' mutual misunderstanding of

the basis for exclusion at trial:

[T]he exclusion of part of Dr. Kraft's expert testimony was warranted. It is undisputed that he did not evaluate, interview, meet nor

- 44 - treat defendant. He himself admitted this at trial upon questioning by the Court. Thus, Dr. Kraft could not testify as to defendant's state of mind.

In my view, the parties' post-trial arguments and the

district judge's post-trial ruling pass like ships in the night.

The parties, including the government, claim that the trial court

excluded Dr. Kraft's proposed testimony as irrelevant, and the

trial judge reiterates that he excluded only Dr. Kraft's testimony

about the defendant's state of mind, which is consistent with Rule

704(b).

The situation evidenced by the record is thus unusual,

if not unique. Fairly read, I cannot find a defense objection

consistent with the preservation requirements of Federal Rule of

Evidence 103, but the record strongly suggests that both the

government and the defense believed that the district judge

excluded the sexual predation part of the proffered expert

testimony. If only the defendant had misunderstood the trial

judge, it would be one thing, but here, as evidenced by its post-

trial filing, the government was also under the misimpression that

the trial judge had excluded this part of Dr. Kraft's testimony as

irrelevant. From the government's post-trial filing, I agree that

even if the government's position does not quite tie the bow on

the question of preservation, both parties, for whatever reason,

treated the district judge's rulings and reminders as prohibiting

- 45 - the sexual predation testimony of Dr. Kraft rather than as

reminders of the boundaries of Rule 704(b). In my view, although

the issue is a close one, there is enough play within Federal Rule

of Evidence 103(e) to allow us to reach the substantive question,

even absent proper preservation. Fed. R. Evid. 103(e) ("A court

may take notice of a plain error affecting a substantial right,

even if the claim of error was not properly preserved.").

Given this conclusion, and for the reasons well stated

by the majority, I join in the majority's conclusion that the jury

should have been presented with Dr. Kraft's proposed testimony

about sexual predation and the internet, including the

characteristics and behavioral patterns of a typical online

predator. I am also influenced by the majority's view that Dr.

Kraft's expert opinion on the hallmarks of online sexual predation

and the significance of the online chats in this case could have

made a difference in the jury verdict and therefore, I concur with

the majority's resolution of this appeal.

- 46 -