United States v. Baptiste

596 F.3d 214 (2010)

UNITED STATES of America, Plaintiff-Appellee,
v.
Nigel Humphrey John BAPTISTE, Defendant-Appellant.

No. 07-4493.

United States Court of Appeals, Fourth Circuit.

Argued: December 2, 2009. Decided: February 26, 2010.

ARGUED: Gary Proctor, Law Offices of Gary E. Proctor, LLC, Baltimore, Maryland, for Appellant. John Francis Purcell, Jr., Office of the United States Attorney, Baltimore, Maryland, for Appellee. ON BRIEF: Rod J. Rosenstein, United States Attorney, Baltimore, Maryland, for Appellee.

Before WILKINSON, GREGORY, and DUNCAN, Circuit Judges.

Affirmed by published opinion. Judge DUNCAN wrote the opinion, in which Judge WILKINSON and Judge GREGORY joined.

OPINION

DUNCAN, Circuit Judge:

This is an appeal from a conviction and sentence on one count of conspiracy to distribute and possess with intent to distribute controlled substances in violation of 21 U.S.C. § 846, and four counts of possession with intent to distribute cocaine and cocaine base in violation of 21 U.S.C. § 841(a)(1). Appellant Nigel Baptiste asserts that the district court committed several errors during his jury trial. Constrained by our review for plain error, we affirm.^[1]

I.

In the spring of 2005, a drug task force organized by the Sheriff's Department of Cecil County, Maryland, began a large-scale investigation of a cocaine distribution ring centered in the small community of Winding Brook. The investigating officers conducted twelve state-court authorized wiretaps, targeting first the lower-level dealers and then proceeding up the distribution channel to several higher-level traffickers. Through the wiretaps and related surveillance, the officers identified an individual named Larry Brown as one of the ring's mid-level cocaine suppliers. The investigation then focused on ascertaining Brown's source of supply. In November 2005, the officers discovered that Brown's supplier was Baptiste. After monitoring Baptiste's actions and telephone calls, the officers determined that Brown and Baptiste were regularly conducting kilogram-quantity cocaine transactions together.

On January 2, 2006, following a meeting during which the officers believed that Baptiste delivered one kilogram of cocaine to Brown, the officers arrested Baptiste. At the time of his arrest, Baptiste possessed six cellular telephones, two of which he had been using to communicate with Brown. Baptiste also possessed a driver's license under a fictitious name.

On May 11, 2006, Baptiste was indicted by a federal grand jury on one count of conspiracy to distribute and possess with intent to distribute controlled substances in violation of 21 U.S.C. § 846, and four counts of possession with intent to distribute cocaine and cocaine base in violation of 21 U.S.C. § 841(a)(1). Several other defendants were also indicted on charges stemming from the Winding Brook investigation. Only Baptiste and one codefendant, Lamar Ricks, elected to proceed to trial; the rest pleaded guilty.

After jury selection but before the start of the trial, the judge learned that some of the jurors had "expressed some feeling of intimidation with respect to people in the hallway outside of [the] courtroom doing what is called staring them down." J.A. 28. Once the court had reconvened after lunch, the judge issued the following warning to the courtroom audience and the parties:

[The alleged staring] will absolutely not be tolerated. No attendant at this trial will be permitted to stand in the hallway here on the fourth floor. You will either sit in the courtroom or you will go down to the first floor and sit in the lobby area, in the couch area. Is that understood by everybody in attendance here? And if anybody disobeys that order by me, they will be taken into custody by the U.S. Marshal and find themselves on the sixth floor lockup. So that's the way it will be. You're in this courtroom or you're downstairs in the first floor lobby. And the marshals will make sure that there's absolutely no effort by anybody to try to intimidate a juror in this case.

J.A. 28-29. Significantly, the court asked both parties if they had any questions. Neither did. The court then addressed defense counsel:

And out of respect to defense counsel, I'm [not] holding that against the defendants in any way, shape or form, but this is not the wild west. You're in the United States District Court. You're in federal court and that kind of conduct will not be tolerated by anybody who's a witness at this case.

J.A. 29. The following day, Ricks entered a guilty plea, and the courtroom became noticeably less crowded for the remainder of the five-day trial.

The evidence presented at trial included testimony from one of the police officers who arrested Baptiste. According to the officer, at the time of his arrest Baptiste told him that "he was glad that he was finally arrested and that he could finally stop looking over his shoulders." J.A. 423. Baptiste also showed the officers a secret compartment in his vehicle which he stated he had used in the past to smuggle cocaine. Another officer testified that, after he informed Baptiste of his Miranda rights, Baptiste volunteered that "he was supposed to be involved in a 150 to 200-kilo deal the following day." J.A. 435. The evidence also included testimony from some of Baptiste's coconspirators, one of whom testified that Baptiste had been a source of cocaine for Brown for many years.

The bulk of the remaining evidence consisted of testimony regarding the investigation of the conspiracy, including descriptions of the surveillance conducted and analyses of numerous wiretapped telephone calls among the coconspirators. To elicit the purpose of the wiretaps and the context of the calls themselves, the government proffered Detective Kenneth Russell, who had been the lead investigator in the case, as an expert witness.

When Russell first took the stand, the government sought to lay the foundation for his expertise regarding "the manner and means of drug trafficking." J.A. 63. In response to the government's foundation-laying questions, Russell indicated that he had worked as a narcotics officer for approximately nine years, that his job required him to work in close proximity with drug users on a daily basis, and that he had previously testified as an expert regarding the manner and means of drug trafficking. Russell also testified that, through his experience, he was familiar with the street jargon associated with cocaine trafficking.

After the government requested that Russell be qualified as an expert, the court asked defense counsel if he wanted to voir dire Russell regarding his qualifications to testify as an expert. Counsel for the defense made no such indication. The court then proceeded to qualify Russell as an expert, but reminded defense counsel of his continuing ability to challenge Russell's testimony on cross-examination. The court also gave the following cautionary instruction to the jury regarding Russell's testimony:

Ladies and gentlemen, under the case law of the United States Court of Appeals for the Fourth Circuit which is the court directly above this court, it has been established that there may be expert testimony with respect to the method and means of drug pack aging and drug distribution as well as the use of slang terms in terms of drug explanations. Having said that, that just means that this witness has been qualified as an expert whereas most times witnesses are not permitted to give their opinions. One exception to that is in the area of expert opinion where a witness is permitted to give his expert opinion. But it's for you to accept, reject or whatever in terms of whether you accept that testimony or not and certainly, [defense counsel] can challenge certain opinions in his cross-examination. But Mr. Russell has been qualified, Detective Russell has been qualified as an expert and is accepted as such by the Court and so accordingly, he is permitted then to give his opinion as an expert.

J.A. 63-64. This was the district court's only jury instruction as to Russell's role in the proceedings.

Russell's testimony throughout the trial followed a regular pattern. The government would ask Russell if he "ha[d] an opinion" about whether a particular phone call "relate[d] to drug trafficking activity" and Russell would respond in the affirmative. See, e.g., J.A. 101. He would then describe the call's content and import. At that point, the government would play for the jury the recording of the call in question. This pattern was repeated for numerous phone calls. Russell's testimony was interrupted by occasional testimony from other officers who participated in surveillance stemming from a particular phone call. In these instances, Russell would leave the witness stand and then return once the other officer finished testifying. A slightly longer interruption occurred near the end of the trial when two of Baptiste's coconspirators offered testimony. Otherwise, the pattern of Russell's testimony remained largely the same throughout the trial.

Although the pattern of Russell's testimony remained the same, its content with respect to particular calls varied slightly. Sometimes Russell merely described the call's content, sometimes he opined about whether it concerned drug distribution, and sometimes he translated code words. For instance, the government at one point asked Russell to interpret the code "2-3-5," which he interpreted to mean "$23,500." J.A. 377-78. On some occasions, Russell would also testify as to the surveillance activities conducted in connection with the intercepted calls. Throughout Russell's testimony, the court's involvement was limited to warning Russell sua sponte at one point against improperly opining about a coconspirator's intent, and to sustaining several of defense counsel's objections unrelated to Russell's dual-role as an expert and fact witness. Defense counsel never requested additional jury instructions from the court regarding Russell's dual-role testimony.

In its closing argument, the government sought to articulate for the jury whether it would be foreseeable from Baptiste's perspective that the cocaine he sold to his underlings would be ultimately turned into crack. To illustrate the concept of foreseeability, the government provided the following analogy:

[L]et me give you an example of something that might not be foreseeable. . . .

Now say we were charging here a case where Mr. Baptiste was [sic], I'm charging him with terrorism. Why? Because—we're not going to charge him with terrorism. But because, for instance, maybe, maybe because the terrorism is a conspiracy to blow up the courthouse or something. He's a cocaine dealer and he sold cocaine to [a coconspirator] who's a terrorist and he used that money before he became a terrorist and he used that money to fund a terrorist organization and go blow up the courthouse. And we saw that the evidence showed that the money came [from] that, all came from cocaine that he bought from Mr. Baptiste. So we're charging Mr. Baptiste, not for being a drug dealer, but being part of a conspiracy to blow up a courthouse. Now that might not be foreseeable. And if I were arguing to you at that point, well, it should be foreseeable that a cocaine dealer should know that people he sells crack to may use the profits to fund a terrorist organization, you would be going I'm not so sure, I'm not so sure. That's all you have? That's not what we're saying. All we're saying is that you can find from the evidence that it's foreseeable that a person who sells powder [cocaine] should expect that powder is going to be turned into crack. That's a decision you have to make.

J.A. 447-48. Neither defense counsel nor the court raised an issue regarding the government's analogy, and the government proceeded with its closing argument.

The jury convicted Baptiste on all counts. Baptiste received concurrent sentences of 300 months' imprisonment on each of the five counts. This appeal followed.

II.

Baptiste raises four distinct arguments on appeal. First, he argues that the district court erred when it failed to voir dire jurors following the alleged incident of jury intimidation. Second, he asserts that the court erred in allowing Detective Russell to testify as an expert witness because Russell did not have a reliable methodology supporting his expert testimony. Third, he argues that the court erred in allowing Russell to testify as both an expert and a fact witness without establishing any safeguards to prevent jury confusion about Russell's dual role. Finally, Baptiste asserts that the court erred in permitting the prosecutor to include in his closing statement an improper metaphor that involved Baptiste unwittingly funding a terrorist act. We will consider each argument in turn.

Because Baptiste did not raise any of these issues at trial, we review for plain error. See Fed.R.Crim.P. 52(b); United States v. Olano, 507 U.S. 725, 731-32, 113 S.Ct. 1770, 123 L.Ed.2d 508 (1993). On plain-error review, "`[i]t is the defendant rather than the Government who bears the burden of persuasion.'" United States v. Hastings, 134 F.3d 235, 240 (4th Cir.1998) (quoting Olano, 507 U.S. at 734, 113 S.Ct. 1770) (alterations in original). Accordingly, to succeed on any one of his arguments, Baptiste must persuade us that there was "an `error' that [was] `plain' and that `affect[ed] substantial rights.'" Olano, 507 U.S. at 732, 113 S.Ct. 1770 (quoting Fed.R.Crim.P. 52(b)). With regard to the third element of that standard, Baptiste must show that the alleged error actually "affected the outcome of the district court proceedings." Id. at 734, 113 S.Ct. 1770. However, "[e]ven if the error is plain and affects substantial rights, . . . we [do] not exercise our discretion to correct the error `unless a miscarriage of justice would result or the error seriously affect[s] the fairness, integrity or public reputation of judicial proceedings independent of the defendant's innocence.'" United States v. Johnson, 219 F.3d 349, 353 (4th Cir.2000) (quoting United States v. Cedelle, 89 F.3d 181, 184 (4th Cir.1996)) (last alteration in original).

Using this framework, we turn to consideration of Baptiste's contentions.

A.

Baptiste first argues that the court erred in failing to adequately address the issue of juror intimidation. He asserts that, upon learning that jurors felt intimidated by stare-downs from members of the crowd, the court was obligated under Remmer v. United States, 347 U.S. 227, 74 S.Ct. 450, 98 L.Ed. 654 (1954), to conduct a voir dire examination of the jurors. Remmer instructs: "In a criminal case, any private communication, contact, or tampering directly or indirectly, with a juror during a trial about the matter pending before the jury is, for obvious reasons, deemed presumptively prejudicial . . ." Id. at 229, 74 S.Ct. 450. We ordinarily review a district court's decision not to hold voir dire following a jury intimidation allegation under an abuse of discretion standard. United States v. Basham, 561 F.3d 302, 320 (4th Cir.2009). However, because Baptiste did not raise this issue at trial, we review only for plain error.

As we have explained, "while a presumption of prejudice attaches to an impermissible communication, the presumption is not one to be casually invoked." Stockton v. Virginia, 852 F.2d 740, 745 (4th Cir.1988). For that reason, we have held that to trigger a Remmer presumption, the defendant bears the initial burden of "`establish[ing] both that an unauthorized contact was made and that it was of such a character as to reasonably draw into question the integrity of the verdict.'" Fullwood v. Lee, 290 F.3d 663, 678 (4th Cir.2002) (quoting Stockton, 852 F.2d at 743.)

We first note that Baptiste made no effort in this case to meet this threshold burden.^[2] But, even assuming without deciding that the alleged stares were sufficient to constitute unauthorized contact,^[3] there is no evidence on the record that the alleged contact with the jury was such that it was likely to influence the verdict. This alone would be sufficient to reject Baptiste's contention on this point. However, we note that there is no evidence on the record that trial counsel ever alleged that the communication would affect the verdict or that voir dire was required. Therefore, because Baptiste did not even attempt to trigger the Remmer prejudice presumption, we cannot say that the district court erred, much less plainly erred, in not responding to the situation by holding a special voir dire.^[4]

Furthermore, Baptiste has failed to show that the alleged error affected his substantial rights as required by the third Olano prong. To do so, Baptiste would have to show that the court's failure to conduct a special voir dire actually affected the outcome of the trial. Olano, 507 U.S. at 734, 113 S.Ct. 1770. Baptiste's assumption, however, that the stare-downs were somehow attributed to him, and therefore turned the jury against him, is based on pure speculation. There is no evidence in the record as to who was responsible for the conduct in question. There is also no indication that the people responsible were linked to Baptiste. In fact, Baptiste alleges that those involved were Ricks's family members, who stopped attending the trial as soon as Ricks pleaded guilty the day after the court issued its warning. Baptiste does not explain why the jury would impute the actions of Ricks's family members to Baptiste. Therefore, nothing on the record supports a conclusion that the intimidation was somehow attributed by the jury to Baptiste to his detriment. Without a showing that the alleged intimidation prejudiced Baptiste, we cannot find that the court's failure to conduct voir dire following the allegations of intimidation affected his substantial rights. Accordingly, Baptiste has failed to establish prejudice as required by Olano.

B.

Baptiste next argues that the court erred in allowing Russell to testify as an expert witness. The admissibility of expert testimony is governed by Federal Rule of Evidence 702. We ordinarily "review for abuse of discretion the district court's decision to admit expert testimony under Federal Rule of Evidence 702." United States v. Wilson, 484 F.3d 267, 273 (4th Cir.2007). However, because Baptiste failed to object at trial to the court's certification of Russell as an expert, our review is for plain error. We therefore must determine whether the district court plainly erred in finding that Russell's testimony complied with the requirements of Federal Rule of Evidence 702.

Federal Rule of Evidence 702 requires, among other things, "that [expert] testimony . . . be the product of reliable principles and methods." Fed.R.Evid. 702 advisory committee's note 2000. As we have noted, the district court has "`broad latitude'" in determining whether the expert's methodology is reliable. Wilson, 484 F.3d at 274 (quoting Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 142, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999)). Baptiste argues that the district court erred in finding Russell's methodology sufficiently reliable.^[5] He asserts that Russell's "purported methodology plainly does not pass muster under Federal Rule of Evidence 702" because it consisted of interpreting coded conversations based on three "shaky bases . . . [:](1) the defendants' identities; (2) their behavior; and (3) the repetition of certain words." Appellant's Br. at 25.^[6]

What Baptiste refers to as "shaky" methodology is very similar to what we found to be sufficient methodology in Wilson, 484 F.3d at 275. There, a narcotics officer was allowed to testify as an expert interpreting drug-related code language. The officer explained his methodology as follows:

It all depends on the situation. I mean, there's, obviously, there's a lot of words to mean one thing. So like I say, you take it into the context of what you're talking about. That's how you determine.. . . It all depends on the context of the call. You know, drug dealers use coded language. And the reason that they do that is because they don't want police involvement or police to know what they're talking about. . . . I take the person who's talking, the conversation. I take what has this person, what's the routine pattern of this person before and the pattern after. And that's how I make my determination. . . . [W]hen you hear [a] word time and time again . . . then there's a pattern that develops. And when that pattern develops, that ultimately shows you what they're talking about.

Id. at 275 (alterations in original). The Wilson court found the officer's methods to be sufficient to meet the requirements of Rule 702. Id. at 276. The description in Wilson of acceptable methodology is comparable to the methodology employed by Russell as described by Baptiste, that is, using context and repetition to determine the meaning of certain words. This type of methodology is exactly what is contemplated by the commentary to Rule 702 for a narcotics officer interpreting coded language. As the advisory committee notes on Rule 702 explain:

[T]he principle used by the agent is that participants in such transactions regularly use code words to conceal the nature of their activities. The method used by the agent is the application of extensive experience to analyze the meaning of the conversations.

Fed.R.Evid. 702 advisory committee's note 2000 (quoted in Wilson, 484 F.3d at 274-75). Because Russell's approach to decoding language is so similar to that which we found acceptable in Wilson, and to that which is contemplated by the commentary to Rule 702, we cannot say that the district court plainly erred in finding that Russell's methodology was sufficiently reliable to allow him to testify as an expert.^[7]

C.

Baptiste next argues that the court erred by failing to employ safeguards to prevent jury confusion regarding Russell's dual role as an expert and fact witness. As a result, Baptiste argues, the jury gave undue weight to Russell's factual testimony and was confused as to when his testimony was based on his expert opinion and when it was based on his factual recollections of the investigation.^[8] While we usually review a court's decisions regarding an expert's testimony for abuse of discretion, Wilson, 484 F.3d at 273, here we review only for plain error because Baptiste failed to raise this issue at trial.

We have previously recognized that while "such dual witnesses could confuse the jury," dual-role testimony is acceptable where "the district court took adequate steps . . . to make certain that [the witness's] dual role did not prejudice or confuse the jury." Id. at 278 n. 5. In Wilson, we found safeguards sufficient where the court had the witness "testify first as a fact witness" and also "issu[ed] a cautionary instruction to the jury." Id. Unlike in Wilson, the court here did not have Russell take two separate trips to the stand, nor did it issue a cautionary instruction specific to Russell's dual role. We have yet to consider the precise circumstance presented here, namely, in which lay and expert witness testimony is presented simultaneously.

Those circuits that have considered simultaneous dual-role testimony have generally found it to be properly admitted so long as the court implements adequate safeguards to prevent juror confusion or jurors giving undue weight to the lay testimony. The Seventh Circuit case of United States v. Farmer, 543 F.3d 363 (7th Cir. 2008) and the Sixth Circuit case of United States v. Lopez-Medina, 461 F.3d 724 (6th Cir.2006) are illustrative of the lines that courts have drawn regarding adequate safeguards.

The defendant in Farmer was charged with drug offenses. At trial, a federal agent testified both to the alleged conduct and to the meaning of code words in drug distribution. He gave both lay and expert testimony during the same trip to the witness stand, following a pattern much like that followed at Baptiste's trial.

The Farmer court noted four safeguards that the district court implemented with regard to the agent's dual-role testimony. First, the district court gave a cautionary instruction to the jury, reminding the jury that it could "give the testimony whatever weight you think it deserves." Farmer, 543 F.3d at 371 (quotation marks and alterations omitted). Second, defense counsel cross-examined the agent about his expert opinion, "which further clarified the testimonial capacities for the jury." Id. Third, the district court required the government to establish a proper foundation for the witness's expertise. Id. at 370-71. Finally, "[t]he government . . . prefaced [the agent's] expert testimony by asking him to interpret the coded language's meaning `based on [his] expertise.'" Id. at 371 (last alteration in original). Without noting which safeguard was most important, the Seventh Circuit concluded that the district court had "adequately alleviated" any potential concerns regarding juror confusion. Id.

By contrast, the Sixth Circuit found safeguards to be inadequate in Lopez-Medina. The only safeguard provided at trial in Lopez-Medina was a cautionary instruction to the jury indicating that the government agent's testimony was not entitled to any greater weight because of the agent's dual role and that defense counsel could attack the agent's credibility. 461 F.3d at 744. On review, the Sixth Circuit also noted that the agent's testimony "lacked any clear demarcation between expert and fact witness roles." Id.

The safeguards implemented at Baptiste's trial fall somewhere between Farmer and Lopez-Medina. On one hand, the district court ensured that the government laid the foundation for Russell's expert testimony. The court also instructed the jury regarding Russell's testimony, saying that "it's for you to accept, reject or whatever in terms of whether you accept that testimony or not," and noted that defense counsel could challenge Russell's opinions. J.A. 64. However, the government's questioning failed to demarcate between lay and expert testimony, a consideration that both Farmer and Lopez-Medina weighed heavily. See Farmer, 543 F.3d at 371; Lopez-Medina, 461 F.3d at 744. Furthermore, defense counsel's cross-examination of Russell did little to contribute to the distinction between lay and expert testimony. Thus, we believe that the district court could have done more to ensure that Russell's lay and expert testimony were demarcated more clearly in order to prevent juror confusion and to prevent jurors from giving undue weight to Russell's lay testimony.

To call this error "plain" under Olano, however, requires a finding that the error is "obvious" or "clear under current law." 507 U.S. at 734, 113 S.Ct. 1770. Given that our circuit has not spoken directly on this issue and that the facts in the present case place it in a gray area of the law as applied by other circuits, we cannot say that the error is obvious or clear at this time.^[9]

Furthermore, Baptiste has failed to show that the alleged jury confusion affected his substantial rights as required by the third Olano prong. Baptiste's brief cites only two examples of situations where the government failed to "delineate when it was eliciting fact testimony and when it was eliciting expert opinions." Appellant's Br. at 29. We find both examples unavailing.

The first example is a situation where, in the middle of Russell's expert testimony regarding a call, the government attorney asked him, "By the way, what was the weather like that night?," to which Russell responded "It was freezing rain." J.A. 378. We fail to see, and Baptiste does not explain, how this statement about the weather would confuse the jury in such a way as to affect Baptiste's substantial rights.

The second example is one in which Russell agreed with counsel that a call that had just been played to the jury was a "sort of [']where are you['] call," J.A. 268, and then proceeded to testify about the surveillance related to that call. In that call, Larry Brown, one of Baptiste's coconspirators, asked Baptiste to pick him up, Baptiste asked Brown to confirm where he was, and Brown responded that he was at the Shop Rite store. Because of the call's straightforward meaning, Russell's agreement that it was indeed a "where are you call" is hardly a translation of the call for which his expertise would be required. In fact, it seems to be simply factual back ground preceding his description about the surveillance surrounding the call, which confirmed that Baptiste picked Brown up at a Shop Rite store. Therefore, this is not a clear example of Russell "seamlessly transitioning from a translation of calls to an account of specific surveillance in the case", Appellant's Br. at 30, in a way that would confuse the jury and prejudice Baptiste.

Furthermore, Baptiste does not assert that Russell's factual testimony in either of these examples was incorrect or inaccurate. Therefore, even if the jury had assigned greater credibility to Russell's factual testimony due to his dual role, it would not have been misled as to the facts.^[10]

Accordingly, because Baptiste has failed to show that the alleged lack of clear distinction between Russell's role as a fact witness and his role as an expert witness actually affected the outcome of the trial, he has not shown the prejudice necessary to support a finding of plain error.

D.

Baptiste's final allegation focuses on the portion of the government's closing argument in which counsel used a metaphor involving Baptiste's hypothetical role in unknowingly helping to fund a terrorist act. Baptiste argues that the remarks confused the jury into thinking that Baptiste was a dangerous man linked to terrorism and therefore deprived Baptiste of his "right to a fair, unbiased jury." Appellant's Br. at 20. Baptiste therefore asserts that the court erred in allowing such remarks. Generally, "[t]he district court is afforded broad discretion in controlling closing arguments and is only to be reversed when there is a clear abuse of its discretion." United States v. Rhynes, 196 F.3d 207, 236 (4th Cir.1999), vacated in part on other grounds, 218 F.3d 310 (4th Cir.2000) (en banc). However, because Baptiste did not object to the closing argument at trial, we review for plain error.

We have found that, "[w]ith respect to claims of prosecutorial misconduct, an appellant must show that the remarks were improper and that they `prejudicially affected the defendant's substantial rights so as to deprive the defendant of a fair trial.'" United States v. Adam, 70 F.3d 776, 780 (4th Cir.1995) (quoting United States v. Mitchell, 1 F.3d 235, 240 (4th Cir.1993)). In determining whether improper remarks require reversal we consider:

(1) the degree to which the prosecutor's remarks have a tendency to mislead the jury and to prejudice the accused; (2) whether the remarks were isolated or extensive; (3) absent the remarks, the strength of competent proof introduced to establish the guilt of the accused; (4) whether the comments were deliberately placed before the jury to divert attention to extraneous matters [;] . . . (5) whether the prosecutor's remarks were invited by improper conduct of defense counsel [;] . . . and (6) whether curative instructions were given to the jury.

United States v. Wilson, 135 F.3d 291, 299 (4th Cir.1998) (internal citations omitted). The ultimate question "is whether the prosecutors' comments `so infected the trial with unfairness as to make the resulting conviction a denial of due process.'" Darden v. Wainwright, 477 U.S. 168, 181, 106 S.Ct. 2464, 91 L.Ed.2d 144 (1986) (quoting Donnelly v. DeChristoforo, 416 U.S. 637, 643, 94 S.Ct. 1868, 40 L.Ed.2d 431 (1974)).

As a threshold matter, we are unconvinced that the prosecutor's statements were improper. The government's apparent purpose in posing the hypothetical was to provide the jurors with an example of unforeseeable criminal activity so as to give them a baseline for determining whether the actual criminal activity being charged—distribution of crack by Baptiste's underlings—was, in fact, foreseeable to Baptiste. The remarks made clear that the government was not actually alleging that Baptiste was funding terrorism; in fact, it was asserting the contrary. At the outset of the metaphor the government stated: "[L]et me give you an example of something that might not be foreseeable.. . . Now say we were charging here a case where Mr. Baptiste was [sic], I'm charging him with terrorism. Why? Because —we're not going to charge him with terrorism." J.A. 447. The government then proceeded to pose a hypothetical situation in which Baptiste sold cocaine to Larry Brown and Brown then used the cocaine proceeds to fund a terrorist organization, unbeknownst to Baptiste. The government gave that as an example of something that Baptiste could not have foreseen. It then contrasted it with something that it alleged Baptiste could have foreseen—that someone would use powder cocaine to make crack.

Nothing in the government's remarks hints at all at the fact that Baptiste was involved in terrorism. The government did not allege, as Baptiste claims, that Baptiste was "involved in a terrorist plot to blow up a courthouse." Appellant's Br. at 18 (emphasis in original). In fact, even as part of the metaphor the government held Baptiste harmless for the unforeseeable terrorist acts of Larry Brown. Therefore the metaphor was actually an example in which Baptiste could not be held responsible for someone else's dangerous acts in which he played no role. Even though we agree that the specter of terrorism should not be invoked lightly, we do not see how the statement could be interpreted as attributing acts of terrorism to Baptiste. We therefore cannot agree with Baptiste that the metaphor was "undoubtedly improper." Id.

Furthermore, even if the prosecutor's remarks in this case had been improper, we find that under the relevant Wilson factors discussed above, Baptiste has failed to show that the comments prejudicially affected his substantial rights so as to deprive him of a fair trial. First, as we explained above, we do not think that the metaphor was such that it would mislead the jury into thinking that Baptiste was somehow dangerous or involved in terrorism. Second, the comment constituted only a very minor part of the government's otherwise unobjectionable closing argument —fourteen lines out of a thirty-six page transcript. Third, the record contains ample evidence against Baptiste absent the remarks. This evidence includes a multitude of coded, drug-related telephone calls between Baptiste and his coconspirators, surveillance corroborating the meanings of those phone calls, Baptiste's statements to officers regarding his involvement in cocaine trafficking, and information from a cooperating coconspirator confirming that Baptiste was Larry Brown's cocaine source for many years. Fourth, there is no indication that the government's use of the metaphor was in any way intended to divert the attention of the jury or prejudice Baptiste. Instead, it is clear that the metaphor was intended to illustrate the concept of foreseeability, which was part of the government's case. Finally, although the comments were not invited by the defense, and the court did not give a curative instruction, we find these factors of minimal relevance in light of the other considerations discussed here. Weighing all of the relevant factors, we find that the remarks did not affect Baptiste's substantial rights.

Accordingly, because we find that the remarks were not plainly inappropriate and did not, in any event, affect Baptiste's substantial rights, we hold that the district court did not plainly err in allowing the prosecutor's closing argument.

III.

For the reasons stated above, we reject Baptiste's allegations of trial mismanagement on the part of the district court, and his conviction is therefore

AFFIRMED.

NOTES

[1] Although represented by counsel on appeal, Baptiste filed a pro se supplemental brief alleging ineffective assistance of counsel at trial. Claims of ineffective assistance of counsel may be raised on direct appeal only where the record conclusively establishes ineffective assistance. See United States v. King, 119 F.3d 290, 295 (4th Cir.1997). Otherwise, the proper avenue for such claims is a 28 U.S.C. § 2255 motion filed with the district court. See Massaro v. United States, 538 U.S. 500, 504-06, 123 S.Ct. 1690, 155 L.Ed.2d 714 (2003) (recognizing that "in most cases a motion brought under § 2255 is preferable to direct appeal for deciding claims of ineffective assistance" because the trial record is "often incomplete or inadequate for [addressing such claims on direct review,]" thereby risking the failure of "[e]ven meritorious claims"); see also United States v. Richardson, 195 F.3d 192, 198 (4th Cir.1999).

Baptiste alleges, among other things, that trial counsel failed to effectively represent him due to a conflict of interest resulting from counsel's prior representation in an unrelated matter of one of Baptiste's co-defendants, who pleaded guilty before trial. Our review of the record shows that it does not conclusively establish any of the alleged grounds for Baptiste's ineffective assistance claim. Therefore, because the claim is not properly before us, we do not address it as part of this appeal.

[2] We find unavailing Baptiste's position that, although he failed to meet his burden of establishing the Remmer presumption, the court should nonetheless have sua sponte held voir dire. Baptiste cites no case law suggesting that the district court has such a duty. As we have made clear, the existence of such a presumption did not absolve Baptiste of his duty to raise the matter. Further, even if such a sua sponte requirement could theoretically be triggered, the record before us does not contain facts that unequivocally show that the contacts were of the sort that would automatically require voir dire.

[3] We note that our sister circuits have disagreed on the issue of whether stares alone can be considered "unauthorized contacts" for purposes of Remmer. Compare United States v. Brown, 923 F.2d 109, 112 (8th Cir.1991) (holding that "physical closeness, stares, and rebuffed efforts at conversation . . . are neither unique nor uncommon to public trials and do not of themselves trigger the Remmer presumption), with United States v. Rutherford, 371 F.3d 634, 643 (9th Cir.2004) (finding stares by government witnesses constitute unauthorized contacts). However, because we find that Baptiste failed to show that any contact that occurred was such that it would affect jury impartiality, we need not resolve this question here.

[4] Baptiste relies on Rutherford for the proposition that "stares" alone can give rise to the Remmer presumption. 371 F.3d at 643. The contrast between Rutherford and this case, however, provides a great example of why Baptiste's allegations are insufficient. In Rutherford, jurors complained that government agents glared at them throughout the trial. The agents in question were sitting behind the government attorneys during the trial, consulted often with those attorneys in view of the jury, and served as key witnesses for the government. The Rutherford court found particularly significant the fact that the persons who had allegedly intimidated the jurors were "government agents intimately associated with the prosecution." Id. at 643. By contrast, Baptiste has provided no evidence showing that the people who allegedly intimidated the jury in this case were linked in any way to him or to the government. It would therefore be mere speculation for us to determine that the conduct prejudiced the jury against Baptiste.

[5] Baptiste does not challenge whether Russell was sufficiently qualified to testify as an expert witness. Indeed, we have previously held that experienced narcotics officers are qualified to testify as expert witnesses regarding drug trafficking, code interpretation, and similar matters. See, e.g., Wilson, 484 F.3d at 275-76.

[6] Although Baptiste argues that Russell offered "no particular methodology" for his determination, he asserts that Russell's testimony "suggests" that he based his testimony on the three grounds specified above. Indeed, although Russell did not specifically mention the word "methodology" in his testimony, he did state that, based on his experience, he was familiar with the street-level jargon associated with drug trafficking and that he decoded conversations by examining them in context of other calls placed between the suspects.

[7] Further supporting our conclusion regarding the absence of prejudice, we also note that many of Russell's interpretations of coded conversations were corroborated at trial by the testimony of actual participants to those conversations.

[8] Baptiste also argues in passing that allowing Russell's dual-role testimony "resulted in his testifying to matters that were not appropriate as expert or lay testimony." Appellant's Br. at 30 (emphasis in original). Whether or not this occurred, we fail to see how it could be a result of his dual role as a lay and expert witness. Both lay and expert witnesses occasionally venture into impermissible areas of testimony. We are unconvinced that Russell's dual role made him somehow especially likely to give such testimony.

[9] Although we find that the error was not plain under current law, we note that district courts should take steps to ensure that there is a clear demarcation in the jury's mind between a witness's lay and expert roles. This may be accomplished, for example, by cautionary warnings or instructions, by requiring the witness to take separate trips to the stand in each capacity, or by ensuring that counsel makes clear when he is eliciting lay versus expert testimony. While the means of ensuring the demarcation between the lay and expert roles of the witness lie within the discretion of the district court, jurors should be made to understand that they may not give the witness's lay testimony additional weight simply because of his dual-role as an expert.

[10] We also note that Russell's factual testimony in the second example related only to surveillance of Baptiste and Brown conducted by Sergeant Gerald Widdoes and Sergeant Timothy McDonald. Both Widdoes and McDonald were witnesses in the trial, and Widdoes testified as to the surveillance on that day. Therefore, had there been any inaccuracies in Russell's testimony, Baptiste's counsel could have easily resolved that confusion by cross-examining Widdoes and McDonald about the surveillance. This would mitigate any potential prejudice.