FILED
                                                                      United States Court of Appeals
                      UNITED STATES COURT OF APPEALS                          Tenth Circuit

                             FOR THE TENTH CIRCUIT                        September 21, 2018
                         _________________________________
                                                                          Elisabeth A. Shumaker
                                                                              Clerk of Court
UNITED STATES OF AMERICA,

      Plaintiff - Appellant,

v.                                                          No. 17-2043
                                                  (D.C. No. 1:15-CR-01889-JCH-1)
RAYMOND MOYA,                                                 (D.N.M.)

      Defendant - Appellee.
                      _________________________________

                             ORDER AND JUDGMENT*
                         _________________________________

Before TYMKOVICH, Chief Judge, MORITZ and EID, Circuit Judges.
                 _________________________________

      In this interlocutory appeal, the government challenges the district court’s

pretrial orders excluding testimony from two expert witnesses.1 We conclude that the

district court didn’t abuse its discretion in excluding some of that testimony as a

sanction for the government’s failure to provide proper notice of its intent to present

that testimony at trial. But we agree with the government that the district court erred

in ruling that admitting the remaining testimony would violate the Confrontation

Clause. Accordingly, we affirm in part and reverse in part.

      *
         This order and judgment isn’t binding precedent, except under the doctrines
of law of the case, res judicata, and collateral estoppel. But it may be cited for its
persuasive value. See Fed. R. App. P. 32.1; 10th Cir. R. 32.1.
       1
         We have jurisdiction over the appeal under 18 U.S.C. § 3731. See § 3731
(providing such jurisdiction “if the United States attorney certifies to the district
court that the appeal is not taken for purpose of delay and that the evidence is a
substantial proof of a fact material in the proceeding”).
                                     Background

      Dawn Sherwood is a certifying scientist with National Medical Services

(NMS), a “full-service laboratory” that provides “forensic toxicological analysis of

body fluids and tissues for drugs and intoxicants.” App. vol. 1, 282. In 2011, the New

Mexico Office of the Medical Investigator (OMI) collected blood and urine samples

from a deceased individual (C.W.) and submitted those samples to NMS for testing.

Based on the results of these tests, Sherwood signed a toxicology report in which she

indicated that C.W.’s blood and urine tested positive for various illegal substances,

including heroin.

      Just over three years later, the government requested additional information

from NMS about the cause of C.W.’s death. In response, forensic toxicologist Laura

Labay reviewed Sherwood’s toxicology report, several law-enforcement investigation

reports, and some grand-jury testimony. Based on this review, Labay concluded that

C.W. died of a heroin overdose.

      Relying in part on that conclusion, the government charged Raymond Moya

with distributing the heroin that caused C.W.’s death. See 21 U.S.C. § 841(b)(1)(C)

(mandating longer sentence for distributing controlled substance “if death . . . results

from the use of such substance”). To prove this charge, the government planned to

introduce expert testimony about the cause of C.W.’s death.

      Under Federal Rule of Criminal Procedure 16, the government was required to

provide Moya with notice and a summary of its expert testimony by the district

court’s discovery deadline. And although it provided such notice for Labay, it failed

                                           2
to do so for Sherwood.2 Instead, two weeks before the scheduled trial date, the

government sought to have Sherwood testify by video. In response, Moya moved to

exclude Sherwood’s testimony because the government failed to timely disclose that

Sherwood would be one of its experts. See Fed. R. Crim. P. 16(d)(2)(C) (giving

district court discretion to exclude evidence if government fails to comply with Rule

16).

       The district court granted Moya’s motion to exclude. In doing so, it found that

(1) the government didn’t provide any legitimate reason for its failure to timely

disclose Sherwood’s expert testimony; (2) Moya would be prejudiced by having to

cross-examine a new expert witness on such short notice; and (3) a continuance

wasn’t feasible to cure this prejudice because of the parties’ and the court’s

scheduling concerns.

       Recognizing that the exclusion of Sherwood’s testimony and report could

impact the admissibility of some of Labay’s testimony, the government then filed a

motion in limine seeking to present that testimony. In its motion, the government

represented that Labay would testify, based in part on Sherwood’s report, that heroin

caused C.W.’s death. But Moya moved to exclude Labay’s testimony as well, arguing

that its admission would violate his rights under the Confrontation Clause. The

district court agreed. It concluded that in the absence of Sherwood’s testimony and


       2
         The government initially suggested that it complied with Rule 16 and
provided Moya with timely notice of its intent to present Sherwood’s testimony. But
on appeal, the government “does not challenge” the district court’s finding that it
failed to comply with Rule 16’s notice requirement. Aplt. Br. 18.
                                           3
report, Labay’s testimony would inappropriately parrot Sherwood’s excluded

testimony. Thus, the district court also excluded Labay’s testimony.

      The government appeals both of the district court’s orders.

                                       Analysis

I.    Sherwood’s Testimony and Rule 16

      The government first argues that the district court erred by excluding

Sherwood’s testimony as a sanction for the government’s failure to comply with Rule

16(a)(1)(G). We review the district court’s decision to exclude this evidence for an

abuse of discretion. United States v. Banks, 

, 1196 (10th Cir. 2014).

Under this deferential standard of review, “we will not disturb the ruling unless it is

arbitrary, capricious, whimsical[,] or manifestly unreasonable, or we are convinced

that the district court made a clear error of judgment or exceeded the bounds of

permissible choice in the circumstances.” 

(quoting United States v.

Nacchio, 

, 1241 (10th Cir. 2009) (en banc)).

      Rule 16 requires the government to disclose certain information to the

defendant, including, among other things, “a written summary of any [expert]

testimony that the government intends to use.” Fed. R. Crim. P. 16(a)(1)(G). When a

party fails to comply with this requirement, the court can impose any appropriate

sanction, including granting a continuance or excluding the undisclosed evidence.

Fed. R. Crim. P. 16(d)(2).

      In determining the appropriate sanction, a district court begins—as the district

court did here—by considering the three Wicker factors: “(1) the reason for the delay,

                                           4
including whether the non-compliant party acted in bad faith; (2) the extent of

prejudice to the party that sought the disclosure; and (3) ‘the feasibility of curing the

prejudice with a continuance.’” 

–99 (quoting United States v.

Wicker, 

, 1061 (10th Cir. 1988)). But these three factors aren’t

necessarily dispositive; although they should “guide the district court in its

consideration of sanctions,” they don’t “dictate the bounds of the court’s discretion.”

(quoting 

).

       A.     Reason for Delay

       In evaluating the government’s reasons for its delayed disclosure, the district

court first concluded that there was “no evidence of bad faith on the part of the

[g]overnment.” App. vol. 2, 311. Indeed, Moya conceded as much, telling the district

court, “I know that there is not bad faith here.” App. vol. 3, 402. No one disputes that

finding on appeal, and we’ve found nothing in the record to contradict it.

       But the district court also concluded that the government didn’t provide a

sufficient reason for its delayed disclosure. On appeal, the government challenges

that conclusion by insisting the district court never “suggest[ed] that the government

lacked a legitimate reason for the delay.” Aplt. Br. 19.

       We disagree. The only potential explanation the government offered below for

its failure to provide Moya with timely notice of Sherwood’s testimony was that “no

party is perfect.” App. vol. 3, 387. In response, the district court specifically stated

that “it [was] not convinced that the [g]overnment’s oversight [was] a sufficient

reason to justify the lengthy delay.” App. vol. 2, 312; see also 

1199. Thus, by finding the government’s only proffered reason insufficient, the

district court indicated that the government didn’t provide a legitimate reason for the

delay. And we find no abuse of discretion in that conclusion. See 

(holding that despite absence of bad faith, government’s “neglect[]” of its

disclosure duty justified district court’s conclusion that this factor weighed in favor

of defendant).

      B.     Prejudice

      Next, the government argues that the district court erred in concluding “that

requiring [Moya] to cross[-]examine a newly disclosed expert witness on such short

notice would unfairly prejudice [him].” App. vol. 2, 312. The government insists

Moya would not be prejudiced because he could not have been unduly surprised by

its decision to call Sherwood as an expert witness. See Fed. R. Crim. P. 16 advisory

committee’s note to 1993 amendment (noting that purpose of requiring expert-

testimony disclosure is “to minimize surprise that often results from unexpected

expert testimony”); United States v. Ivy, 

, 1281 (10th Cir. 1996)

(considering surprise as part of Wicker’s prejudice inquiry).3 Specifically, the



      3
         Ivy also notes that when a defendant doesn’t ask for a continuance, “a court
can often assume that counsel did not need more time to incorporate the information
into the defense’s game 

(quoting United States v. Sepulveda,

, 1178 (1st Cir. 1993)). Here, Moya didn’t request a continuance. But
the government doesn’t ask us to take that fact into account in evaluating the second
Wicker factor. Thus, we decline to do so. See Bronson v. Swensen, 

,
1104 (10th Cir. 2007) (noting that arguments not made in opening brief are waived).
And in any event, Ivy simply points out that a district court can make such an
assumption; it doesn’t indicate that a district court abuses its discretion if it chooses
                                            6
government argues that Moya wasn’t surprised because (1) he had Sherwood’s report

“well over a year before the scheduled trial date,” Aplt. Br. 20, and (2) he listed

Sherwood on the second of his three witness lists.

      The government’s factual assertions are accurate. But we disagree that those

facts mean Moya wasn’t surprised and thus wasn’t prejudiced.

      Merely having a copy of Sherwood’s report wouldn’t necessarily prompt

defense counsel to prepare to cross-examine Sherwood as an expert. Likewise, that

defense counsel once listed Sherwood as a potential defense witness would—at

most—suggest that defense counsel might have been prepared to conduct a direct

examination of Sherwood. But cross-examination and direct examination require

distinct preparation. And preparing to cross-examine an expert witness is particularly

tricky. See Smith v. Ford Motor Co., 

, 793 (10th Cir. 1980) (explaining

that “even with the help of his [or her] own experts,” an attorney “frequently cannot

anticipate the particular approach [an] adversary’s expert will take” (quoting Fed. R.

Civ. P. 26 advisory committee’s note to 1970 amendment)). Here, the government

indisputably failed to include Sherwood in its earlier expert-witness disclosure or its

earlier witness list. So the government’s belated decision to call Sherwood as an

expert witness likely surprised Moya. And that surprise operated to his detriment:

before the government’s late notice, Moya had no reason to plan or prepare to cross-




not to do so. Cf. 

(noting that Wicker factors don’t “dictate
the bounds of the court’s discretion” (quoting 

)).
                                            7
examine Sherwood. See United States v. Golyansky, 

, 1250 (10th Cir.

2002) (defining prejudice as impact on defendant’s ability to prepare).

      Citing United States v. Charley, 

(10th Cir. 1999), the

government disagrees. But Charley is distinguishable. True, we found no abuse of

discretion in the district court’s refusal to exclude expert testimony in that case.

. Yet unlike the government in this case, the government in

Charley timely notified the defense that it would call certain witnesses. It merely

failed to provide summaries of their planned testimony, as Rule 16 requires. 

Thus, the defendant in Charley knew that certain witnesses would testify on

behalf of the government. But here, Moya had no notice at all that Sherwood would

testify for the government: the government didn’t notify Moya that it would call

Sherwood as a witness until two weeks before trial, and the expert designation came

less than one week before trial. Thus, the district court didn’t abuse its discretion in

finding that Moya would be prejudiced by having to prepare to cross-examine an

expert witness on such short notice.

      C.     Feasibility of a Continuance

      In applying the third Wicker factor, a district court must consider whether a

continuance is a feasible way to cure the prejudice. See 

.

Here, the district court concluded that a continuance wasn’t feasible, in large part

because of its own significant “docket and scheduling limitations.” App. vol. 2, 312.

      At the outset, the government asserts that Moya’s opposition to a continuance

makes the third Wicker factor “essentially irrelevant.” Aplt. Br. 24 (quoting Ivy, 

); see also 

(characterizing “third Wicker factor” as

essentially irrelevant” because defendant “made it clear she did not want a

continuance”). But the government didn’t make this argument below. In fact, it

adopted precisely the opposite position: rather than characterizing this third factor as

“irrelevant,” Aplt. Br. 24, the government told the district court that—despite Moya’s

opposition to a continuance—the court nevertheless “need[ed] to consider” this factor

and couldn’t “skip over that step” in the Wicker analysis, App. vol. 3, 410–11. Thus,

we could apply the invited-error doctrine and hold that the government waived this

argument below. Cf. United States v. LaHue, 

, 1013 (10th Cir. 2001)

(noting that invited-error doctrine precludes review on appeal of argument when that

argument is “directly contradictory” to appellant’s position in district court).

Moreover, even if the government merely forfeited this argument below, it doesn’t

argue for plain error on appeal. Accordingly, we could decline to consider its

assertion that the third Wicker factor is irrelevant on that basis as well. See Richison

v. Ernest Grp., Inc., 

, 1131 (10th Cir. 2011) (noting that failure to

argue for plain error on appeal “surely marks the end of the road” for argument not

raised below).

      In any event, we reject this argument on the merits. As an initial matter, the

third Wicker factor simply requires the district court to consider the feasibility of a

continuance that could cure the prejudice. See 

. Wicker says

nothing about only considering the feasibility of a continuance if a party requests

one. Moreover, even assuming that Moya’s opposition to a continuance makes the

                                            9
third Wicker factor “essentially irrelevant,” that simply means the factor doesn’t

weigh in anyone’s favor; it doesn’t mean the factor weighs against Moya, as the

government contends. 

. And the other two Wicker factors weigh

in Moya’s favor, leaving the balance tilted toward him. Additionally, Moya’s

opposition to a continuance isn’t relevant to one of the district court’s primary

reasons for finding a continuance wasn’t feasible: its own docket and scheduling

limitations. In other words, even if Moya requested a continuance, it doesn’t appear

the district court would have found that option to be a feasible one. Thus, we reject

the government’s position that the third Wicker factor weighs against Moya because

Moya was opposed to a continuance.

      The government next argues that the district court’s analysis of the third

Wicker factor was insufficient. In particular, it complains that the district court didn’t

expressly consider the length of a continuance that might be necessary to cure the

prejudice. But the only authority the government cites to support this argument is

United States v. Yepa, 572 F. App’x 577, 586 (10th Cir. 2014) (unpublished) (noting

that findings about necessary length of continuance should be part of analysis). And

“[u]npublished decisions are not precedential.” 10th Cir. R. 32.1(A). Moreover,

Wicker doesn’t require that the district court consider any particular facts related to

the feasibility of a continuance; indeed, Wicker explicitly eschews such a proscriptive

approach. 

(noting that Wicker factors “are not intended to

dictate the bounds of the court’s discretion”). And in any event, we read the record as

suggesting the district court believed a continuance of any length would be

                                           10
problematic: it noted that multiple continuances had already occurred in this case and

emphasized the difficulty of rescheduling the trial both because of the court’s docket

and because of the “challenge to reschedule all the defense witnesses.” App. vol. 2,

312.

       The district court appropriately contemplated such considerations in

determining whether a continuance was feasible, even independent of the prejudice

analysis. As the Wicker court itself noted, “the district court may need to suppress

evidence that did not comply with discovery orders to maintain the integrity and

schedule of the court even though the defendant may not be 

. Here, the district court concluded that because of “the difficulties inherent in

rescheduling the case” and “the [c]ourt’s own docket and scheduling limitations,” it

wasn’t “clear that a continuance would be a feasible way to cure the prejudice.” App.

vol. 2, 312. The government hasn’t challenged that finding, and we find support for it

in the record.4 We thus hold the district court did not abuse its discretion in reaching

that conclusion.


       4
         Although the government never argues that the district court relied on an
erroneous factual finding when evaluating this factor, the dissent nevertheless asserts
that it did so. Specifically, the dissent contends that there’s no record support for the
district court’s finding that the government “did not want a continuance.” App. vol.
2, 312. We disagree. At the hearing, the government said, “[R]ight now we are not in
a position to ask for a continuance because we are ready to go.” App. vol. 3, 421–22.
Then, the district court said, “Nobody wants a continuance, including me.” 

And the government replied, “Right. We would certainly do it if that gets us to the
point where everybody gets what they want.” 

the government
ultimately acceded to a continuance in the unlikely event that such a continuance
would mean “everybody g[ot] what they want[ed],” it also unequivocally indicated it
didn’t want one. 

if the government had challenged this finding as clearly
                                           11
      D.     Balancing the Wicker Factors

      Finally, even assuming the district court didn’t abuse its discretion in

evaluating the individual Wicker factors, the government argues that we should

reverse because we have “held on numerous occasions that district courts abused

their discretion in excluding evidence even where the justification for exclusion was

considerably stronger than it was in this case.” Aplt. Br. 25. But the cases the

government cites don’t support this assertion.

      The government is correct that the facts in at least some of these cases are

arguably more egregious than the facts present here. See, e.g., United States v.

Gonzales, 

, 1291 (10th Cir. 1999) (“[T]he government knowingly and

intentionally violated discovery orders and misrepresented the witness’ status and

whereabouts to the court and defense counsel.”). But the government overlooks our

reasons for reversing in each of the cases it cites. It wasn’t because we found the

government’s behavior insufficiently egregious, as the government seems to suggest.

It was because the district courts’ ultimate decisions to exclude evidence were based

on intermediate findings that we determined were legally or factually flawed. See

(reversing because district court legally erred when it

relied on its finding that defendant would be prejudiced by “additional financial

erroneous, we would reject that challenge; this exchange supports the district court’s
finding that the government “did not want a continuance.” App. vol. 2, 312; see also
United States v. Sanchez-Urias, 

, 1071 (10th Cir. 2018) (explaining
that “factual finding is clearly erroneous” only “if it lacks evidentiary support or if a
review of the evidence leaves us with the definite and firm conviction that a mistake
has been made” (quoting United States v. Mirabal, 

, 1032 (10th Cir.
2017), cert. denied, 

(2018))).
                                           12
burdens should the case be continued”; holding that “[t]he concept of prejudice does

not encompass the expense of additional trial preparation”); 

(reversing because we found “no record support whatsoever” for district court’s

conclusion “that the prejudice to defendants” from government’s misconduct “was

irreparable”); United States v. Ivory, 131 F. App’x 628, 632 (10th Cir. 2005)

(unpublished) (reversing because district court legally erred in relying on

government’s discovery misconduct in separate cases without providing government

with advance notice of its intent to do so).

      But the government fails to identify any similar legal or factual errors here. As

discussed above, the district court found that the government didn’t sufficiently

justify its failure to provide notice of Sherwood’s expert testimony, that the

government’s failure prejudiced Moya, and that a continuance wasn’t feasible to cure

that prejudice. And the government has failed to demonstrate that these findings were

clearly erroneous or that the district court committed any legal error in taking these

findings into account. Accordingly, the cases the government relies on are inapposite

and do not require us to find an abuse of discretion here.

      Next, the government points out that we have previously affirmed the

exclusion of evidence when a party attempted to introduce it after trial had

commenced. See, e.g., 

(holding that district court didn’t

abuse its discretion in excluding certain testimony where defendant sought to “call

expert witnesses on the ninth day of trial”); United States v. Russell, 

,

1510, 1512 (10th Cir. 1997) (holding that district court didn’t abuse its discretion in

                                           13
excluding testimony of substitute witnesses when defense counsel gave government

notice of intent to call those witnesses on morning of fifth day of trial). But here, the

government’s late disclosure occurred before the trial was scheduled to begin, rather

than mid-trial. So the government argues that the prejudice in this case is less severe

than in Banks and Russell. But the fact that a district court doesn’t abuse its

discretion in excluding evidence that isn’t disclosed until after trial has commenced

doesn’t necessarily mean that a district court does abuse its discretion in excluding

evidence that is disclosed before that point. To hold otherwise would too narrowly

circumscribe the “broad discretion” that district courts possess to “impos[e] sanctions

on parties who violate discovery orders.” 

.

      In a related argument, the government insists that a continuance is the typical

remedy for a Rule 16 violation in the absence of bad faith or when trial isn’t already

in progress. See 

(“In the absence of a finding of bad

faith, the court should impose the least severe sanction that will accomplish prompt

and full compliance with the discovery order.”); 

(“A

continuance may normally be the most desirable remedy . . . .”). But “the

‘admonition that the trial court must impose the least severe sanction that will

accomplish . . . prompt and full compliance with the court’s discovery orders does

not mean that a continuance is necessary just because it will cure the prejudice.’”

(alteration in original) (quoting 

).

In fact, by indicating that a continuance “may normally be” the appropriate remedy,



                                           14
Wicker necessarily recognizes that other remedies—such as exclusion—will be

appropriate in some circumstances. 

(emphasis added).5

      Thus, for the reasons discussed above, we hold that the district court didn’t

abuse its discretion in excluding Sherwood’s expert testimony. See 

(noting that an abuse of discretion occurs when a decision “is arbitrary,

capricious, whimsical[,] or manifestly unreasonable,” or when “the district court

made a clear error of judgment or exceeded the bounds of permissible choice in the

circumstances” (quoting 

)). In reaching that conclusion, we

emphasize that the question before us isn’t whether exclusion is the sanction that we

would have chosen. See 

(McKay, J., concurring in part

and dissenting in part). The question is whether the district court abused its discretion

when it decided that this was one of the “rare case[s]” in which exclusion was

appropriate. 

. We conclude that it did not.

II.   Labay’s Testimony and the Confrontation Clause

      Next, the government argues that the district court erred when it found that

allowing the government to present Labay’s testimony—which would be based in

      5
         Relying on language in Golyansky, the dissent similarly states that a district
court is “oblige[d]” to impose “the least severe sanction that will accomplish prompt
and full compliance with the discovery order.” Dissent 6 (quoting 

). But the dissent reads the language in Golyansky too broadly. Contrary
to the dissent’s position, Golyansky doesn’t “limit[]” a district court’s discretion
when deciding on sanctions for Rule 16 violations. 

simply
notes that a district court “should impose the least severe sanction” and that “[t]he
preferred sanction is a 

(emphases added). Words like
“should” and “preferred” connote recommendations—not requirements. 

we disagree with the dissent’s assertion that the district court’s “discretion [was]
limited” or “cabined” by our caselaw. Dissent 1, 6.
                                           15
part on Sherwood’s toxicology report—would violate the Confrontation Clause. “We

review de novo a district court’s legal conclusions regarding the Confrontation

Clause.” United States v. Garcia, 

, 1211 (10th Cir. 2015).

      “The Sixth Amendment’s Confrontation Clause provides that, ‘[i]n all criminal

prosecutions, the accused shall enjoy the right . . . to be confronted with the

witnesses against him.’” Crawford v. Washington, 

, 42 (2004) (alteration

in original) (quoting U.S. Const. amend. VI). In practice, the Confrontation Clause

prohibits the admission of testimonial hearsay at trial unless the declarant is

unavailable to testify and the defendant had a prior opportunity to cross-examine the

declarant. See 

The key word here is “testimonial”: the Confrontation

Clause only applies to testimonial hearsay. See 

As such, we begin with the

government’s argument that Sherwood’s toxicology report isn’t testimonial, and thus

admitting Labay’s testimony based on that report wouldn’t violate the Confrontation

Clause.

      Generally speaking, a statement is testimonial if it’s “made with the primary

purpose of creating evidence for the prosecution.” United States v. Alcorta, 

, 1137 (10th Cir. 2017); see also Bullcoming v. New Mexico, 

, 659

n.6 (2011) (defining testimonial statement as one with “a ‘primary purpose’ of

‘establish[ing] or prov[ing] past events potentially relevant to later criminal

prosecution’” (alterations in original) (quoting Davis v. Washington, 

,

822 (2006))). Applying this primary-purpose test, see Ohio v. Clark, 

,

2179 (2015), we consider whether “a reasonable person in the position of the

                                           16
declarant would objectively foresee that the primary purpose of the statement was for

use in the investigation or prosecution of a crime.” United States v. Ibarra-Diaz, 

, 917 (10th Cir. 2015) (quoting United States v. Smalls, 

, 778

(10th Cir. 2010)).

      The government contends that Sherwood’s toxicology report doesn’t meet the

primary-purpose test and thus isn’t testimonial. Moya responds, although without any

elaboration or explanation, that the report is testimonial. Courts applying the

primary-purpose test to forensic reports tend to consider both (1) the general

relationship between law enforcement and the agency that completed the testing as

well as (2) the particular facts and circumstances of the case, such as the passage of

time between the creation of the report and the criminal charges. See, e.g., United

States v. James, 

, 97–99 (2d Cir. 2013); United States v. Ignasiak, 

, 1231–33 (11th Cir. 2012); United States v. Moore, 

, 73 (D.C.

Cir. 2011). For example, in James, the Second Circuit held that the declarant’s

primary purpose in preparing an autopsy report wasn’t to “creat[e] a record for use at

a later criminal trial” because (1) the New York City Office of the Chief Medical

Examiner was an agency that was independent from law enforcement and (2) that

office completed the autopsy report long before law enforcement began a criminal

investigation into the victim’s 

–99. Thus, the autopsy report

wasn’t testimonial. 

cf. 

–32 (holding that an

autopsy report was testimonial primarily because under statutory framework, Florida

Medical Examiner’s Office existed within Department of Law Enforcement).

                                           17
       Here, New Mexico requires medical examiners to conduct an autopsy if the

medical examiner suspects the death was caused by a criminal act or if “the cause of

death is obscure.” N.M. Stat. Ann. § 24-11-7. Medical examiners must also report

their findings directly to the district attorney in all cases they have investigated. 

And, as Moya asserts in a letter of supplemental authority, this statutory

framework could support the conclusion that some reports generated by medical

examiners are testimonial. See State v. Navarette, 

, 440–41 (N.M. 2013)

(finding autopsy report testimonial based in part on statutory connection between

OMI and law enforcement).

       But in this case, we aren’t concerned with an autopsy report generated by a

medical examiner. Rather, as the government points out, this case involves a

“toxicology report [that] was ordered not by law enforcement in connection with a

prosecution, but by the OMI in connection with an investigation into the cause of a

death.” Aplt. Br. 38–39. Thus, the statutory relationship between law enforcement

and OMI isn’t the only factor relevant to our inquiry about whether the toxicology

report created by NMS is testimonial. And when we focus on the other relevant

factors, we conclude that the toxicology report in this case isn’t testimonial. See

(noting that “[c]ourts must evaluate challenged statements

in context”).

       First, as the government points out, almost four years passed between August

2011, when C.W. died and Sherwood completed her report, and May 2015, when the

government charged Moya. The record doesn’t disclose exactly when, in this nearly-

                                            18
four-year period, law enforcement either (1) began investigating C.W.’s death or

(2) connected his death to Moya’s alleged heroin distribution. What we do know is

that Labay completed her first cause-of-death report in October 2014, over three

years after Sherwood certified the toxicology report. In addition, Labay’s report

identifies the investigation reports that she relied on, most of which are from 2013

and the earliest of which appears to be dated September 2012. Thus, it appears that

Sherwood certified her report at least one year before any criminal investigation into

C.W.’s death began, if not longer. This time gap suggests that the toxicology report

isn’t testimonial. See 

(finding autopsy report not testimonial in

part because it was completed “substantially before any criminal investigation into

[victim’s] death had begun”); 

(finding toxicology report not testimonial in

part because there was “no indication . . . that a criminal investigation was

contemplated during the inquiry”); cf. State v. Bass, 

, 1225 (N.J. 2016)

(finding autopsy report testimonial in part because autopsy took place during active

homicide investigation).

      Second, the government points out that law enforcement didn’t request the

toxicology report from NMS. Nor does it appear that law enforcement received a

copy of it directly from NMS. Instead, OMI requested the toxicology testing, and it

appears that Sherwood sent her report directly to OMI, not to law enforcement. And

although it’s not determinative, this further suggests that the toxicology report isn’t

testimonial. See 

n.3 (noting that involvement of law

enforcement in forensic investigation isn’t determinative); State v. Mattox, 890

                                           

, 267 (Wis. 2017) (“The toxicology report at issue in [this] case was not

prepared for or given to law enforcement, making it much less likely to be

testimonial.”); cf. 

(finding that statements made to someone

other than law enforcement are more likely to be nontestimonial).

       Third, the record doesn’t disclose any other facts that would suggest to a

reasonable person in Sherwood’s position that the primary purpose of her report

detailing the results of the blood and urine tests was for criminal investigation or

prosecution. See 

(noting that test is whether reasonable

person “would objectively foresee that the primary purpose of the statement was for

use in the investigation or prosecution of a crime” (quoting 

)).

True, she worked for a forensic lab and therefore conducted testing that could

theoretically be used in criminal investigation or prosecution. But as the government

points out, OMI routinely conducts autopsies that don’t ultimately lead to criminal

investigations or prosecutions. Cf. 

& n.10 (concluding, based

on statistics, that “there is reason to believe that [no criminal investigation] is

pursued in the case of most autopsies”). And nothing about the report Sherwood

signed in this case suggested any criminal investigation or prosecution would be

forthcoming. Her toxicology report is distinct, for example, from the testimonial

autopsy report in Bass. There, the “autopsy was conducted in the presence of two law

enforcement officers” and the autopsy report concluded that the manner of death was

a gunshot wound to the back of the victim’s torso—an obvious indication that the

wound wasn’t self-inflicted. 

–26; see also 

73 (finding autopsy reports testimonial when they concluded that manner of death

was “homicide caused by gunshot wounds” and homicide detectives attended several

autopsies and participated in creation of reports). Nor does the record disclose any

other interaction between NMS and law enforcement in connection with the creation

of the toxicology report. See 

n.13 (finding involvement of

law enforcement in transporting samples was routine and thus did “not indicate that a

criminal investigation was contemplated”). These facts thus further suggest the

nontestimonial status of this particular toxicology report.

      Additionally, we agree with the government’s assertion that Sherwood’s

toxicology report isn’t like the testimonial affidavits in Melendez-Diaz v.

Massachusetts, 

(2009), or the testimonial lab report in 

–65. In Melendez-Diaz, the Court held that affidavits certifying a seized

substance as cocaine were “functionally identical to live, in-court 

–11. But the toxicology report here differs from the affidavits in

Melendez-Diaz in two important ways. First, the testing in Melendez-Diaz was

conducted at law enforcement’s request. 

Here, as we’ve explained, law

enforcement had no involvement with the creation of the toxicology report. Second,

the affidavits in Melendez-Diaz stated their evidentiary purpose on their face. 

But the toxicology report contains no similar statement, and it served no obvious

evidentiary purpose at the time Sherwood signed it. In sum, the toxicology report is

not the same as affidavits sworn for the “sole purpose” of providing “‘prima facie



                                           21
evidence of the composition, quality, and the net weight’ of [an] analyzed

substance.” 

Gen. Laws ch. 111, § 13 (repealed 2012)).

      Nor is Sherwood’s toxicology report similar to the “certificate of analyst”

containing the results of a blood-alcohol test introduced without the testimony of the

analyst in 

(quoting App. 62). Like in Melendez-Diaz, the

blood alcohol testing in Bullcoming was administered after a law-enforcement officer

provided seized evidence to a state laboratory following a DUI arrest, and the lab

report itself included details about how it could be admitted as evidence in court. 

Neither of those facts exists here. True, the Bullcoming court said that “[a]n

analyst’s certification prepared in connection with a criminal investigation or

prosecution” is testimonial and can’t be introduced at trial absent testimony from the

certifying analyst. 

(emphasis added). But here, as we’ve already

outlined, Sherwood’s certification wasn’t “prepared in connection with a criminal

investigation or prosecution.” 

      For comparison’s sake, consider Labay’s reports. She completed those reports

at the request of law enforcement about six months before the government indicted

Moya. She relied on investigative reports, the autopsy report, the toxicology report,

and grand-jury testimony to conclude that C.W. died of a heroin overdose. And the

facts surrounding the creation of these reports—the closeness in time and law

enforcement’s involvement—indicate that Labay completed them “in connection with

a criminal investigation or prosecution.” 

reasonable forensic

toxicologist, when completing such a report, would foresee that “the primary

                                             22
purpose” was “the investigation or prosecution of a crime.” 

(quoting 

). On the other hand, there are no similar facts

indicating that Sherwood completed her report to establish or prove “past events

potentially relevant to later criminal prosecution.” 

n.6

(quoting 

). Instead, we conclude that the primary purpose of

Sherwood’s report was “to determine the cause of C.W.’s death.” Aplt. Br. 36; see

also 

(finding toxicology report nontestimonial “because

its primary purpose was to identify the concentration of the tested substances in

biological samples sent by the medical examiner as a part of her autopsy to determine

the cause of death”).

      As such, Sherwood’s toxicology report is not testimonial.6 And because the

Confrontation Clause only applies to testimonial hearsay, see 

, Labay’s reliance on Sherwood’s report doesn’t implicate that constitutional

provision.7 We therefore reverse the district court’s order excluding Labay’s

testimony.


      6
           The government further argues that the Supreme Court’s decision in Williams
v. Illinois, 

(2012), supports this conclusion. And to the extent that
Williams found a forensic test—in particular, a DNA report—to be nontestimonial,
the result in that case corresponds to the result we reach here. See 

–58 (plurality opinion). But as some of our sibling circuits have concluded,
Williams lacks a controlling rationale for that result. United States v. Duron-Caldera,

, 994 n.4 (5th Cir. 2013); 

. Thus, Williams offers
little, if any, actual guidance here.
         7
           Because we hold that Sherwood’s report isn’t testimonial under the primary-
purpose test, we don’t address the government’s argument that the toxicology report
isn’t testimonial because it contains only data generated by machines. But we do note
that this point is markedly similar to an argument that the Supreme Court rejected in
                                          23
                                     Conclusion

      Because the district court didn’t abuse its discretion when it examined and

balanced the Wicker factors and sanctioned the government for violating Rule 16, we

affirm the district court’s order excluding Sherwood’s testimony. But we also

conclude that Sherwood’s report isn’t testimonial. As a result, Labay may testify

based in part on that report without running afoul of the Confrontation Clause. So we

reverse the district court’s order excluding Labay’s testimony and remand for further

proceedings.


                                           Entered for the Court


                                           Nancy L. Moritz
                                           Circuit Judge




Bullcoming. 

–61 (rejecting New Mexico court’s conclusion that
analyst’s report and certification merely transcribed results produced by machines;
noting that report contained “more than a machine-generated number”).
                                         24
United States v. Moya, No. 17-2043
EID, J., dissenting.

       Our precedent states that the exclusion of a witness to remedy a discovery

violation should be a “rare” occurrence absent bad faith. United States v. Golyansky, 

, 1249 (10th Cir. 2002). In United States v. Gonzales, for example, we

concluded that the district court “abused its discretion in imposing what was obviously

the most severe available sanction, i.e., complete suppression of the witness’ statements

and trial testimony” because the sanction was “too severe” under the circumstances. 

, 1292 (10th Cir. 1999). In the case before us, however, the majority barely

acknowledges this aspect of our caselaw. Indeed, it never considers whether excluding

Sherwood was “too severe” a sanction or whether this is the “rare” case in which

exclusion is warranted. Instead, it relies upon the dissent in Gonzales that rejected such

an analysis and emphasized that “[t]he question before [us] . . . is not whether

[exclusion] . . . was the remedy we would have chosen, or even whether we think

[exclusion] . . . is unduly harsh.” 

(McKay, J., concurring in part and

dissenting in part) (disagreeing with majority’s conclusion that the district court abused

its discretion in excluding witness); see maj. op. at 15. But assessing whether a sanction

is unduly harsh is precisely the inquiry that our precedent requires. See 

(majority opinion). Because I would apply our governing precedent and

find that that this is not a “rare” case and that exclusion was “too severe” a sanction under

the circumstances, I respectfully dissent.
       As the majority points out, the district court’s decision to exclude Sherwood is

subject to abuse of discretion review. Maj. op. at 4. In the context of the exclusion of

witness testimony, however, we have significantly cabined that discretion. As we stated

in Golyansky,

       In the absence of a finding of bad faith, the court should impose the least
       severe sanction that will accomplish prompt and full compliance with the
       discovery order. The preferred sanction is a continuance. It would be a
       rare case where, absent bad faith, a district should exclude evidence rather
       than continue the 

(emphasis added) (citations omitted). As we have recognized,

exclusion of a witness is “almost never imposed in the absence of a constitutional

violation or statutory authority for such exclusion.” United States v. Charley, 

, 1262 (10th Cir. 1999) (quotations omitted). In deciding whether exclusion is “the

least severe sanction” sufficient to address the violation, 

, the

district court should consider three factors: (1) the reason for the delay; (2) the extent of

the prejudice to the party seeking disclosure; and (3) the feasibility of curing the

prejudice with a continuance, see maj. op. at 4–5 (citing United States v. Banks, 

, 1198–99 (10th Cir. 2014)); see also 

(considering these

factors and concluding that exclusion was “too severe” a sanction and remanding for

“consideration of less severe sanctions”).


       Here, it is undisputed that the failure to list Sherwood as a witness was not

motivated by bad faith. Maj. op. at 5. Therefore, the question is whether this was the

“rare case” in which exclusion was necessary. I believe the answer to that question is no.



                                              2
       First, the prejudice to Moya was relatively small. As the majority acknowledges,

Sherwood’s toxicology report was provided to Moya over a year before trial. Maj. op. at

7. Further, Moya retained two experts to challenge Sherwood’s findings. App. vol. 1,

190–99 (Dr. Steven Pike); App. vol. 2, 297–307 (Janine Arvizu). Both of these experts

challenged Sherwood’s findings in their own reports. App. vol. 1, 190–99; App. vol. 2,

297–307. Moreover, Moya listed Sherwood on his list of potential witnesses (later

deleting her once the government filed an emergency motion to add her to its list).

Compare App. vol. 1, 187–88 (amended witness list; including Sherwood), with 

(second amended witness list; omitting Sherwood). Moya was thus not only

familiar with Sherwood’s toxicology report, he had retained two experts to take issue

with it and had listed her as a potential witness. Under these circumstances, it is difficult

to say that the prejudice of nondisclosure was substantial or that Moya would have

needed significant time to prepare for cross-examining Sherwood. This case is thus

analogous to Charley where we concluded that, although some witnesses had been

mislabeled fact witnesses instead of expert witnesses, the defendant did not “suffer[] any

prejudice” because the “government turned over to the defense copies of all . . . records

from which the witnesses would be 

.1



1
  The majority distinguishes Charley on the ground that, “unlike the government in this
case, the government in Charley timely notified the defense that it would call certain
witnesses. It merely failed to provide summaries of their planned testimony . . . . Thus,
the defendant in Charley knew that certain witnesses would testify on behalf of the
government.” Maj. op. at 8. I disagree with the majority’s reading of Charley, as the
government in that case had not designated the witnesses as experts. 

(“Well before trial in this case, the government notified defense counsel that it

                                              3
       Second, the feasibility of a continuance was not properly considered by the district

court. The district court noted that the case had been continued “multiple times” already.

App. vol. 2, 312. Indeed, Moya received at least four continuances so that he could retain

experts to challenge Sherwood. See App. vol. 1, 121–59, 164–65, 205–23. However, the

district court did not explore the feasibility of a continuance. The court noted that that

the government “would acquiesce if [Moya] requested a continuance” so that Sherwood

could testify, but explained that “[Moya] expressed a clear desire to proceed with the trial

as scheduled because it would be quite a challenge to reschedule all the defense

witnesses.” App. vol. 2, 312. The court went on to find that it would not grant a

continuance because “both parties expressed that they did not want a continuance and

[because of] the difficulties inherent in rescheduling the case, along with the Court’s own

docket.” 

district court’s suggestion that the government “did not want a

continuance” is inaccurate.2 The government’s position was that it did not believe a



would call as witnesses certain health care professionals . . . . However, it did not
designate any of these witnesses as experts and, accordingly, did not provide defense
counsel with any summaries of expert witness testimony . . . .” (emphasis added)). The
case before us is actually a weaker case for exclusion: Not only was Moya aware of the
witness, as in Charley, but he also knew what the witness would say since he had had
Sherwood’s report for over a year.
2
  The majority suggests that the government “never argues that the district court relied on
an erroneous factual finding when [evaluating the feasibility of a continuance].” Maj. op.
at 11 n.4. The majority’s suggestion overlooks the government’s vigorous argument that
the district court’s “analysis [of this factor was] insufficient.” Aplt. Br. at 24 (calling to
the court’s attention the district court’s conclusion that neither side “want[ed] a
continuance” and contending that the district court’s analysis was “insufficient”).


                                              4
continuance was necessary given Moya’s possession of Sherwood’s report, his retention

of two witnesses to challenge that report, and the fact that Sherwood had appeared on his

witness list. See App. vol. 3 at 407–11; App. vol. 1, 187–88. But there is nothing in the

record to suggest that the government preferred exclusion over a continuance if faced

with the possibility that Sherwood would not be permitted to testify. Indeed, the

government was prepared to accept a continuance if one was necessary to allow Moya to

prepare for Sherwood’s testimony. Thus, the district court excluded Sherwood’s

testimony on the ground, not supported by the record, that the government opposed a

continuance. The absence of record support for the district court’s conclusion in this

regard constitutes an abuse of discretion. See 

(concluding

that the district court had abused its discretion in ordering exclusion where there was “no

record support whatsoever” for the court’s conclusion that a continuance could not cure

prejudice to defendants). Stated differently, the district court abused its discretion in this

case by not considering whether a continuance could cure any prejudice caused by the

government’s failure to timely disclose Sherwood. The most that can be said about the

district court’s assessment of this factor is that it would have been inconvenient to

reschedule the trial. Maj. op. at 8 (noting that the district court mentioned its own

“docket and scheduling limitations”).


       More fundamentally, there is nothing in the record to support a finding that this

was the “rare” case in which exclusion was a necessary remedy. I do not take issue with




                                              5
the majority’s conclusion that there was at least some prejudice to Moya due to the

government’s failure to timely list Sherwood, and at least some difficulty in rescheduling

the trial. Maj. op. at 8, 10, 14. But this would be true in virtually any case involving a

discovery violation. The question here is whether this is the “rare” case where a

continuance would not be the proper remedy. There is nothing in the record in my view

that sets this case apart so as to justify exclusion.


       The most troubling aspect of the majority’s opinion is its failure to recognize that

exclusion is to be the “rare” remedy. As we explained in Gonzales,

       we emphasize the Supreme Court has never approved exclusion of evidence
       as a sanction for government misconduct in the absence of a constitutional
       violation or statutory authority for such exclusion. Indeed, the Court has
       emphasized that penalties visited upon the Government, and in turn upon
       the public, because its officers have violated the law must bear some
       relation to the purposes which the law is to serve. Here, we are convinced
       the sanction of total exclusion is too severe and hinders, rather than
       forwards, the public interest in a full and truthful disclosure of critical 

(citations and quotations omitted). As noted above, the majority not

only overlooks this caselaw, but relies on the dissent in Gonzales that rejected such an

analysis and emphasized that “[t]he question before [us] . . . is not whether [exclusion] . .

. was the remedy we would have chosen, or even whether we think [exclusion] . . . is

unduly harsh.” 

(McKay, J., concurring in part and dissenting

in part) (disagreeing with majority’s conclusion that the district court abused its

discretion in excluding witness); see maj. op. at 15. But according to the Gonzales

majority opinion, we must assess whether the sanction is too 

–

93 (finding exclusion to be “too severe” of a sanction and remanding the case to the


                                                6
district court for “less severe sanctions”). Here, that inquiry into whether exclusion was

necessary is an easy one because there is nothing in the record to suggest that a brief

continuance would not have cured any prejudice to Moya, or even that a continuance was

necessary.


       In the end, the majority’s basic error, in my view, is to apply an abuse of

discretion standard without any recognition that the discretion has been limited. The

majority states that “[u]nder this deferential standard of review, ‘we will not disturb the

ruling unless it is arbitrary, capricious, whimsical[,] or manifestly unreasonable, or we are

convinced that the district court made a clear error of judgment or exceeded the bounds

of permissible choice in the circumstances.’” Maj. op. at 4 (second alteration in original)

(quoting United States v. Banks, 

, 1197 (10th Cir. 2014));3 see also maj.

op. at 15 (same). Here, I agree with the majority that the district court’s decision was not

arbitrary, capricious, whimsical, or manifestly unreasonable. It did, however, make a

decision that was outside the bounds of permissible choice because our caselaw obliges

the court to select “the least severe sanction that will accomplish prompt and full

compliance with the discovery order.” 

.



3
 To the extent that the majority relies on Banks, see maj. op. at 14–15, that reliance is
misplaced.
In Banks, the defendants attempted to call three undisclosed expert witnesses on the ninth
day of trial and after the close of the government’s 

–99. The
district court permitted one undisclosed expert to testify because the “exclusion of
evidence for violating discovery orders should not be done lightly,” but excluded the
other two undisclosed defense experts. 

(citation omitted). On appeal, we
upheld that exclusion order because Banks is indeed the rare case. 

I would reverse on the ground that the district court abused its discretion

in excluding Sherwood’s testimony, I do not need to reach the government’s second

ground for appeal, namely, whether Labay’s testimony would violate the Confrontation

Clause. I therefore express no view on that issue. For the reasons stated above, I

respectfully dissent.




                                             8