Before Boudin, Chief Judge, Torruella and Lynch, Circuit Judges.
William Tejeda was convicted after a twelve-day jury trial of conspiracy to possess with intent to distribute cocaine base in the amount of 50 grams or more. Tejeda was a New York supplier of drugs for a Cape Cod drug ring. This criminal appeal attacks both Tejeda's conviction and the sentence imposed. The attack on the conviction is based largely on an incident in the courtroom in which a spectator made a throat-slitting gesture.
Four of Tejeda's co-conspirators pled guilty, and two, Amanda Eldridge and Desiree Alves, testified against him. Tejeda and a co- conspirator, Carmen Figueroa, went to trial; she testified, he did not. Both were convicted. Tejeda was sentenced to twenty years' imprisonment, five years of supervised release, and a special assessment of $100.
As to his conviction, Tejeda does not attack the sufficiency of the evidence. The evidence against him was extremely strong, a point discussed later. He does argue that the trial court committed two errors that deprived him of a fair trial and require his conviction be set aside. The first claim of error is that the court should have granted a mistrial when an older man sitting in the gallery of the courtroom on the first day of trial (and later possibly connected to defendant by the evidence) made a throat-slitting gesture. That gesture was seen by two jurors and discussed by them with other jurors. The second trial error alleged was that the court erred in not severing Tejeda's trial from co-defendant Figueroa's or bifurcating the trial so as to isolate the case against him.
Tejeda also attacks his sentence on the grounds that the district court erroneously relied on the Guidelines applicable to crack cocaine offenses and that the sentence was unreasonable.
Acknowledging the able advocacy on both sides, we affirm the conviction and sentence.
I. The Facts in Evidence at the Trial Fairly Establish the Following
From January 2003 through March 2004, Tejeda supplied crack cocaine to a drug distribution ring operating in Cape Cod, Massachusetts. Tejeda was based in New York; his address, ironically, was 1234 Boston Road, in the Bronx. The Cape Cod ring was headed by Manuel Mendes, a co-defendant, who operated the ring while imprisoned at the Plymouth County House of Correction. That imprisonment limited Mendes in his mobility and in his command operations, but it did not stop him. Mendes operated the drug ring by giving instructions to co- defendant Figueroa, and sometimes to co-defendant Alves, both of whom he was permitted to telephone. Figueroa, in turn, contacted Tejeda and the other conspirators.
On Mendes' instructions, testifying co-conspirator Amanda Eldridge drove to New York in January 2003, along with Carmen Figueroa and Desiree Alves, to pick up crack from Tejeda. Eldridge already knew Tejeda from prior drug transactions. She identified Tejeda in court and had done so in an earlier photo array. Eldridge testified that in New York she met with Tejeda alone in a red van, picked up the drugs, and returned to Cape Cod. Eldridge's testimony was corroborated by Alves, who testified that these drug trips to New York took place about every two weeks.
Eldridge testified that, under Mendes' direction, she made the trip to New York to pick up drugs from Tejeda every week or so for a four-month period. Eldridge then drove back to Cape Cod and gave the drugs to either Alves or Figueroa. Each time, Tejeda gave Eldridge disk-shaped packages containing crack cocaine. Eldridge also identified Tejeda's voice in a call recorded by the government. The recording captured Tejeda and Figueroa arranging a drug transaction for February 17, 2004 in New York.
The February 17, 2004 transaction went off as planned, as was established by the testimony of Alves and two drug enforcement officers who witnessed the event. Co-defendant Christopher Custer was the courier this time. Alves did not go on the trip, but she did give Custer the money to purchase the drugs the night before, and she received the drugs at her home in Yarmouth from Custer when he returned from his successful trip to New York.
The DEA had established surveillance over Tejeda's home in the Bronx and observed Custer's arrival and actions on February 17. Tejeda left his home and entered the red van. A short while later, Custer arrived, parked behind the van, and entered the van. Photographs of Tejeda at the scene of the February 17 deal were admitted into evidence at trial. Within approximately five minutes, Custer got back into his car and left. He was tracked by police as he drove back to Cape Cod and was seen entering Alves' home with a black bag and emerging without it.
The police made arrests immediately after observing another drug deal, this one on March 16, 2004. Custer, after again obtaining money on Cape Cod to pay for the crack, went to New York. Agents saw Tejeda leave the Boston Road address in the Bronx. Tejeda then got in a car, and police tracked the car to Manhattan. In Manhattan, an agent observed Custer on the same street as the car carrying Tejeda. A short time later, after losing sight of Custer, the agent saw Tejeda get out of the front passenger seat of the car, and Custer get out of the back passenger seat and into the front passenger seat of the car. The police thereafter stopped the car; in it were two disk-like objects, which later tested positive for cocaine base. That evening Tejeda was arrested.
The government's case rested primarily on this evidence, and not on the testimony of the non-pleading co-defendant, Figueroa.
II. Trial Error Claims
A. The Throat-Slitting Gesture and Fair Trial Rights
1. Facts And Procedural History
On the morning of the third day of trial, Friday, May 6, 2005, the government's witness was Lt. Balcom. During questioning, Lt. Balcom indicated that on one occasion he had observed some of the individuals in the courtroom in the red van. At the prosecutor's request, Lt. Balcom pointed out these people in the courtroom.
During the morning break, a juror reported privately to the court that she and another juror had seen one of the individuals identified by Lt. Balcom make a throat-slitting gesture on the second day of trial, Wednesday, May 4. After the break and outside of the presence of the jury, the court inquired and was told by defense counsel that the man in question was Tejeda's grandfather. The jury was never informed of this fact.
The court reported to counsel, outside the presence of the jury, that a juror had stated that she had observed Tejeda's grandfather making a throat-slitting gesture. The court then ordered the gesturer and his wife to leave the courthouse and barred them from the courthouse and the surrounding vicinity. At least some members of the jury were later informed of the court's order. On the record, the court described the man barred from the courtroom as an "obviously frail appearing, old man."
Tejeda and Figueroa both moved for mistrial. When the court discussed holding voir dire of the two jurors who reportedly had seen the gesture, Tejeda's counsel expressed concern that the proposed voir dire would only heighten the issue for the jury.
Before starting the trial again that Friday, the court said to the jurors:
A matter has been conveyed to . . . me and I have taken care of it completely. Put it out of your mind. It has nothing to do with this case. Have in mind, while the report is fully appropriate, indeed we want such a report, do not discuss the case among yourselves, the case now, in the jury room.
Tejeda did not object to the instruction, and the trial continued for the day. Later, several of the jurors, on query, said that the jury had complied with this instruction. Assuming that to be true, any discussion amongst the jurors occurred before the court dealt with the report of the gesture.
On Sunday, May 8, Tejeda submitted a written mistrial motion. On Monday, May 9, the court denied the motion but told the parties it would grant their request to voir dire the juror who had reported the gesture, Juror 11.
The court then individually questioned Juror 11. She said that she was concerned about the gesture but did not report it on the day she saw it. It was only after she heard the testimony of Lt. Balcom identifying the spectator as having been at the red van that she started to feel uncomfortable about what she had seen the spectator do two days earlier. After hearing Lt. Balcom's testimony on Friday morning, she reported the gesture to several of the other jurors, some of whom told her she should disclose it to the court, and so she did. The court had earlier instructed the jurors not to discuss the case.
When asked whether she had linked the individual making the gesture to any particular party in the case, Juror 11 answered that she didn't know whether the older gentleman and the lady who had been sitting with him had anything to do with a particular person or side. The juror stated that all she knew was that Lt. Balcom had testified to seeing one of them in the red van. While she recognized the public nature of the courtroom setting, she was concerned that people in the courtroom could know the names of the jurors. The court asked whether she could put the incident out of her mind when evaluating the case and be fair to all parties. She answered affirmatively, and the court, observing her, found her to be forthcoming. Tejeda again moved for mistrial.
The court then questioned Juror 6, who also had seen the gesture. Juror 6 described the gesture as odd and a strange thing to do in the middle of a court proceeding but did not think there was anything to it. She said that it was unclear to her toward whom the gesture was directed. Juror 6, like Juror 11, answered in the affirmative as to her ability to be fair and impartial. The court then denied the motion for mistrial.
At the request of Tejeda's counsel, on the fifth day of trial, Tuesday, May 10, 2005, the court conducted a voir dire of the remaining jurors. None of the other jurors had seen the incident; all but one said that they had heard of it from Jurors 11 and/or 6. The jurors reported that they did not discuss the incident other than on the Friday morning before it was reported.
Each juror was asked whether he or she could be fair and impartial, and every one responded in the affirmative. While several of the jurors indicated that they thought the gesture was inappropriate, no juror expressed fear for his or her safety. Two jurors reported that the gesture had been unsettling or disconcerting for the jurors who had seen it. One juror stated that the reported gesture put pressure on the jury, but that the remedial action taken by the court had been appropriate. Another juror stated that there could be safety concerns, but after hearing of the court's order barring the gesturer from the courthouse, she stated that she was satisfied with her own safety. One juror reported that he knew nothing of the gesture.
Following the voir dire of the entire jury, Tejeda renewed his motion for a mistrial and also requested that the jury foreperson, Juror 11, be deemed an alternate so she would not participate in the jury deliberation. These motions were ultimately denied.
The jury returned a verdict of conviction. Thereafter, Tejeda filed a motion for new trial based on, inter alia, the throat-slitting incident. At that time, Tejeda did not argue that the purported error was a structural one, or that the court was required to apply a presumption of prejudice to the gesture. The court denied the new trial motion.
2. Appropriate Analytical Framework
a. Structural Error
Tejeda argues that it is clear that the jury here was not impartial, despite the trial judge's finding to the contrary and the jurors' individual declarations that the incident did not render any of them unable to remain impartial in reaching a verdict.
Tejeda leaps from this purported factual certainty to an argument, presented for the first time on appeal, that this issue must be analyzed as structural error. Tejeda would like this to be so because in structural error cases the defendant is entitled to automatic reversal. We reject Tejeda's argument.
Structural error analysis has been constricted in its use to a limited category of claimed errors, none of which fits this case. The Supreme Court has held that it is structural error for a criminal defendant to be tried before a judge who has a financial interest in convicting him. Tumey v. Ohio, 273 U.S. 510, 523, 535 (1927). Tejeda infers from Tumey that his claim that a juror is biased must also be analyzed as a structural error. This is not a situation in which one or more jurors has a financial interest in convicting the defendant. We reject Tejeda's argument. The law is clear both that this is not the stuff of structural error and that Tejeda's jury bias claim is subject to a different standard of review. On review, the appellate court asks whether the trial court abused its discretion in denying a mistrial. That standard gives deference to the trial judge, who is in the best position to evaluate the matter.
Mere error in the trial process itself is not structural error. United States v. Gonzalez-Lopez, 126 S.Ct. 2557, 2563-64 (2006). Structural errors are reserved for cases where criminal defendants are denied basic protections which "necessarily render a trial fundamentally unfair" such that "no criminal punishment may be regarded as fundamentally fair." Neder v. United States, 527 U.S. 1, 8- 9 (1999) (quoting Rose v. Clark, 478 U.S. 570, 577-78 (1986)) (internal quotation marks omitted). Tejeda's claim of juror bias is not within this very limited class of cases. See United States v. Mackey, 114 F.3d 470, 474 (4th Cir. 1997) (juror bias and misconduct claims are not structural errors).1
b. The Remmer Presumption of Prejudice
On appeal, Tejeda also argues for the first time2 that he is entitled to a presumption of prejudice under Remmer v. United States, 347 U.S. 227 (1954). We hold that the Remmer presumption does not apply here.
In Remmer, the defendant learned after his conviction that a third party had attempted to bribe a juror to get a verdict favorable to the defendant, and he moved for a new trial. Id. at 228. The court had not informed defense counsel of the bribe, but had initiated an investigation of the bribe and had held an ex parte meeting with the prosecution. Id. The defendant's motion for new trial was denied. Id. at 229. The Supreme Court held that Remmer was not entitled to an automatic reversal, but rather to a hearing before the trial court, and that the jury tampering would be presumed to be prejudicial. Id. at 229-30.
There is an ongoing debate in the circuits about the limits on and the ongoing vitality of the presumption of prejudice rule announced in Remmer. Compare United States v. Pennell, 737 F.2d 521, 532-33 (6th Cir. 1984) (presumption of prejudice no longer exists), with United States v. Sylvester, 143 F.3d 923, 934 (5th Cir. 1998) (proper inquiry is whether likelihood of prejudice is high enough to assign to the government the burden of proving harmlessness), United States v. Williams-Davis, 90 F.3d 490, 497 (D.C. Cir. 1996) (same), United States v. Lloyd, 269 F.3d 228, 238 (3d Cir. 2001) (presumption applies "when the extraneous information is of a considerably serious nature," such as "when a juror is directly contacted by third-parties"), United States v. Dutkel, 192 F.3d 893, 897 (9th Cir. 1999) ("[A] presumption of prejudice arises if a juror was subjected to coercion or bribery, and if this intrusion may have affected the juror in the exercise of his judgment."), United States v. Scull, 321 F.3d 1270, 1280 (10th Cir. 2003) (prejudice presumed "[w]hen members of a jury are exposed to extraneous information about a matter pending before [them]"), and United States v. Greer, 285 F.3d 158, 173 (2d Cir. 2000) (similar). As we discussed in United States v. Bradshaw, 281 F.3d 278, 287 (1st Cir. 2002), two later Supreme Court cases, United States v. Olano, 507 U.S. 725, 737-39 (1993), and Smith v. Phillips, 455 U.S. 209, 215-17 (1982), narrowed the broad language in Remmer.
This court has already rejected defendant's argument that the Remmer presumption applies to all claims of juror bias resulting from extraneous contacts. See, e.g., Bradshaw, 281 F.3d at 288-89; United States v. Gomes, 177 F.3d 76, 82-83 (1st Cir. 1999); United States v. Boylan, 898 F.2d 230, 260-62 (1st Cir. 1990).
We employed the presumption of prejudice in United States v. Gaston-Brito, 64 F.3d 11 (1st Cir. 1995), where a trial court summarily denied, without conducting any inquiry, a motion for mistrial made before the verdict. Id. at 13. We did so in the context of holding that a trial court is obligated to "conduct a sufficient inquiry to determine whether the communication was harmless." Id. (quoting United States v. O'Brien, 972 F.2d 12, 14 (1st Cir. 1992)) (internal quotation marks omitted). Tejeda relies on Gaston-Brito to argue that given "the jurors' expressions of fear and concern . . . , it [was] error to find that . . . Tejeda was not prejudiced by the throat-slitting gesture."
Gaston-Brito is inapposite. In Gaston-Brito, a cooperating witness was asked to identify the person who had threatened to kill his daughter if his wife did not turn over drug proceeds. A case agent seated at the prosecution table pointed to the defense table, possibly indicating to the jury that the defendants had made the threat, although no evidence established that. Id. at 12. The gesture at issue here is of a completely different nature.
There are other distinctions. As Gaston-Brito noted, the government created the problem there when a case agent for the prosecution made an inappropriate gesture conveying substantive evidence. Id. at 13. Since the jury could well think that the agent had inside information, there was a risk the jurors would consider the information during deliberations. Id. The court in Gaston-Brito said it was applying a heightened standard when the prosecution was responsible for improper ex parte conduct.3 Id. Here, the gesture did not come from the prosecution and was not an effort to put evidence in front of the jury. We add that there are different considerations at play when a defendant attempts to vacate a conviction, in the face of overwhelming evidence of guilt, on the basis that someone associated with the defense made an improper gesture to the jury. For example, we would not want to create an incentive for such gesturing by individuals associated with defendants.
3. Merits Analysis
Our usual standard of review once the trial judge has made an appropriate inquiry, and the one that we utilize here, is an abuse of discretion standard, which recognizes that the district court "has wide discretion in deciding how to handle and how to respond to allegations of juror bias and misconduct that arise during a trial." United States v. Rodr¡guez-Ortiz, 455 F.3d 18, 23 (1st Cir. 2006); see also United States v. Mikutowicz, 365 F.3d 65, 74 (1st Cir. 2004); Bradshaw, 281 F.3d at 286-87. We review the district court's factual findings for clear error. Bradshaw, 281 F.3d at 291.
Where a colorable claim of jury taint surfaces before jury deliberations occur, our law describes the sequence of steps a trial judge should take. See id. at 289. The judge should investigate the allegation promptly, addressing whether the taint-producing event occurred, and if so, assessing the magnitude and extent of any prejudice caused. Id. The trial court has wide discretion in how it goes about this inquiry. Id. at 290. The district court, if faced with the issue initially post-verdict, may convene an evidentiary hearing, but it is not obligated to do so. Boylan, 898 F.2d at 258.
If the court determines that there is a taint-producing event and a significant potential for prejudice, the trial court should then examine whether prophylactic measures will alleviate the prejudice (and if so, take them), or whether the threat can otherwise be dispelled or disproved. Bradshaw, 281 F.3d at 289. The court may determine that no curative measures will suffice and grant a timely motion for a mistrial. Id.
Here, the court followed every step in the procedure. There is no realistic objection to the process it used. Rather, Tejeda's attack is on the court's conclusion that any potential prejudice had been adequately addressed. The court did not apply an incorrect legal standard, so we review its conclusion for abuse of discretion.
This case involves (1) a risk of a perception by a juror of an implicit threat from someone who might, in the juror's view, be associated with the defendant; and (2) the risk that this "threat" might influence the juror's ability to impartially evaluate the evidence. These risks are weighed against the individual jurors' own statements that they were not so influenced and the trial judge's findings of fact that the jurors could fairly and impartially reach a verdict.
The risks in this case are considerably weaker than those posed by the facts in Rodr¡guez-Ortiz, where we upheld a district court's denial of mistrial. 455 F.3d at 23-24. There, a juror reported he had received an explicit death threat related to the case. Id. at 24. On inquiry, the juror said he could remain impartial, and the court concluded he could do so. Id. Other circuits have likewise affirmed the denial of a mistrial when a juror was threatened but assured the court that he could remain impartial in deciding the case. See, e.g., United States v. Simmons, No. 99-50381, 2000 WL 429704, at *1-2 (9th Cir. Apr. 20, 2000) (unpublished table decision) (mem.) (affirming denial of mistrial when spectator made gestures during closing arguments that made some jurors feel threatened); Leisher v. Conrad, 41 F.3d 753, 754-55 (D.C. Cir. 1994) (affirming denial of mistrial when two jurors misinterpreted as menacing a gesture made by defendant outside of courtroom); United States v. Garner, No. 90-5613, 1991 WL 150788, at *2, 3 (6th Cir. Aug. 6, 1991) (unpublished table decision) (per curiam) (affirming denial of mistrial when juror was threatened by men presumed to be affiliated with defendant who made noise like a gunshot outside the courtroom); United States v. Zelinka, 862 F.2d 92, 93, 94-96 (6th Cir. 1988) (affirming denial of mistrial when spectator who appeared to be associated with defendant made statement that it would be too bad if the elevator the jurors were boarding should crash, causing fear on the part of some of the jurors).
Tejeda argues that more than a single juror is involved, which increases the risk of taint. This, he says, necessarily requires a different result from that reached in Rodr¡guez-Ortiz. It does not. The implicit threat, while serious, was of a different nature than the direct threat in Rodr¡guez-Ortiz. Further, it is unclear here to whom the gesture was intended, and here there was a possible, but not direct, association between the threat and the case.
Importantly, the district court did not ignore the risk that one or more of the jurors could perceive this gesture as a threat or that a threat might impair impartiality. First, the court took immediate remedial action. The court had the spectator who was the source of the "threat" removed from the courtroom and informed the jurors not to be concerned. The elderly spectator was no longer present when the evidence connecting the red van to the defendant became stronger at trial, and at least some of the jurors were informed that the spectator was not permitted to be in the vicinity of the courthouse. Second, the court queried each juror about whether the throat-slitting gesture would render him or her unable to impartially evaluate the evidence. The court observed the demeanor of each juror and concluded that each could be impartial.
Nor did the court leave the incident for further conversation and musings among the jurors. The court instructed them not to discuss it, and, those who were later questioned said the jurors had complied with that instruction. Normally, if jurors say they have followed instructions, their statements are credited. Cf. Penry v. Johnson, 532 U.S. 782, 799 (2001); Boylan, 898 F.2d at 263.4
There are, of course, extreme cases in which jurors' responses will not be credited. See Bruton v. United States, 391 U.S. 123, 135-36 (1968) ("[T]here are some contexts in which the risk that the jury will not, or cannot, follow instructions is so great, and the consequences of failure so vital to the defendant, that the practical and human limitations of the jury system cannot be ignored. Such a context is presented here, where the powerfully incriminating extra-judicial statements of a co-defendant, who stands accused side-by-side with the defendant, are deliberately spread before the jury in a joint trial." (citations omitted)). This is the argument that Tejeda makes -- that the incident was so strongly and indelibly harmful that no instruction and no voir dire could cure it. Tejeda adds that the jurors' voir dire responses demonstrated that they would not be able to disregard the gesture in reaching a decision in the case. We disagree for the reasons stated above.
The district court did not abuse its discretion in its handling of the throat-slitting gesture. Rather, it responded sensitively and correctly.
B. Severance and Bifurcation
Appellate review of trial court decisions to sever trials of criminal co-defendants is for manifest abuse of discretion. United States v. DeLeon, 187 F.3d 60, 63 (1st Cir. 1999); Boylan, 898 F.2d at 246. Ordinarily, criminal co-defendants are to be tried together. United States v. Houle, 237 F.3d 71, 75-76 (1st Cir. 2001). This rule has particular resonance in drug conspiracy cases, where multiple defendants often share a single indictment. See United States v. Soto- Ben¡quez, 356 F.3d 1, 29 (1st Cir. 2003). Tejeda bears the burden of making "a strong showing of prejudice" in order to gain a new trial. Boylan, 898 F.2d at 246 (quoting United States v. Porter, 764 F.2d 1, 12 (1st Cir. 1985)) (internal quotation marks omitted).
Tejeda twice moved to sever his trial from that of his co- defendants. He did so as to all co-defendants by motion before trial. Later, on the first day of trial, he orally moved to sever his trial from that of co-defendant Figueroa, after he learned that Figueroa would testify and present a battered woman defense as to her relationship with Mendes.5 Her testimony would be that Mendes regularly beat her. Her proffer indicated that she would not implicate Tejeda in the abuse she suffered.
On appeal, the only issue has to do with the motions to sever the case against Figueroa. Tejeda's severance motion argued that it was likely that Figueroa would take the stand and admit her involvement in the conspiracy, but present a case of duress because she had been under the influence of Mendes, who had pled guilty. The court denied the motion for severance saying Tejeda could cross-examine Figueroa, and that no inconsistent defenses were involved.
On May 12, 2005, Tejeda proposed to the trial court that it bifurcate the trial, differentiating the evidence applicable to his case and to Figueroa's. Specifically, Tejeda proposed that he put on his case, the prosecution put on its rebuttal, then the jury decide on Tejeda's guilt or innocence alone. Thereafter, Figueroa would present her defense, the prosecution its rebuttal, and the jury would then render its verdict as to Figueroa. The court denied Tejeda's motion. The motion was renewed at the close of the prosecution's case-in-chief and again denied.
The prosecution questions whether the severance motion on the day of trial and the bifurcation motion6 were timely. See Fed. R. Crim. P. 12(b)(3)(D) (severance motions must be made before trial). We bypass the question and rule on the merits.
Severance should be granted where "defenses are so irreconcilable as to involve fundamental disagreement over core and basic facts." United States v. Peña-Lora, 225 F.3d 17, 34 (1st Cir. 2000) (quoting United States v. Paradis, 802 F.2d 553, 561 (1st Cir. 1986) (emphases added)) (internal quotation marks omitted). But where there is merely some dissonance, where the defenses are just "somewhat antagonistic," we will usually not reverse a trial court's denial of severance. United States v. Serafino, 281 F.3d 327, 329 (1st Cir. 2002). Tejeda relies on United States v. Johnson, 478 F.2d 1129 (5th Cir. 1973), to argue that severance was necessary in this case. But in Johnson, the co-defendants' theories of defense were truly antagonistic. While Johnson defended on the basis that he was not present when the charged crime was committed, his co-defendant made a confession explicitly and necessarily incriminating Johnson and defended only on the basis that he lacked the necessary mens rea. Id. at 1131-32.
Here, by contrast, there was no true antagonism of defenses. Tejeda's defense explicitly acknowledged there was a drug conspiracy; he simply argued the prosecution could not prove beyond a reasonable doubt that he was the New York source of drugs for the conspiracy. Thus, Figueroa's defense that there was a drug conspiracy, but that she had acted under duress, was not an antagonistic defense at all. Her defense did not in any way hinge on Tejeda's participation in the conspiracy. Indeed, even if the jury accepted her duress defense it could either accept or reject Tejeda's defense. Severance is not required every time a duress defense is asserted by one defendant. Peña-Lora, 225 F.3d at 34; United States v. Arias-Villaneuva, 998 F.2d 1491, 1507 (9th Cir. 1993), overruled on other grounds by United States v. Jimenez-Ortega, 472 F.3d 1102, 1102-04 (9th Cir. 2007).
Even if Tejeda had shown that severance from Figueroa was appropriate in this drug conspiracy case, it was his burden to show prejudice. Boylan, 898 F.2d at 246; Porter, 764 F.2d at 12. "[P]rejudice means more than just a better chance of acquittal at a separate trial." Boylan, 898 F.2d at 246 (quoting United States v. Martinez, 479 F.2d 824, 828 (1st Cir. 1973)) (internal quotation marks omitted). "Garden variety" prejudice, which always exists when more than one defendant or offense are tried together, does not warrant a new trial. Id. Tejeda has not met his burden here.
Tejeda's real argument is not one of antagonism. It is that Figueroa's testimony about being regularly beaten by her boyfriend Mendes, the head of the conspiracy, would spill over to Tejeda, who was accused of conspiring with Mendes in drug distribution. This is a sort of argument that "you are known by the friends you keep," and bad friends will taint you in the jury's eyes. It is almost inherent in drug conspiracy cases that a co-conspirator may have engaged in other types of blameworthy conduct. That is not enough to warrant severance of co-defendants' trials. While Mendes' treatment of Figueroa may have been more shocking than run-of-the-mill blameworthy conduct from drug co-conspirators, Tejeda's role in the conspiracy -- as a geographically- removed supplier several states distant -- lessened the risk that the jurors might ascribe Mendes' conduct to him. Moreover, the trial judge instructed the jury that evidence of Mendes' mistreatment of Figueroa was not to be considered in the case against Tejeda.
Figueroa did identify Tejeda in court as the drug ring's New York source of supply. She also testified about numerous trips to New York to pick up drugs. But all of this testimony, including the identification, was cumulative of Eldridge's testimony. There was more than sufficient evidence against Tejeda without Figueroa's identification. In any event, Figueroa's testimony regarding Tejeda was admissible against Tejeda regardless of whether the trials were severed. Further, Tejeda was permitted to cross-examine Figueroa at trial, minimizing any thought of misconduct by association.
The district court did not abuse its discretion in denying the severance motion. Nor was there any abuse of discretion in the denial of the bifurcation procedure. Such a procedure risked confusion of the jury and unwarranted singling out of the case against Tejeda.
Given that we have found no error in the district court's handling of the throat-slitting gesture or the severance and bifurcation motions, Tejeda's cumulative error claim is without merit.
III. Sentencing Claims
A. Application of Crack Cocaine Guidelines
We review de novo sentencing issues involving questions of law. United States v. McCarthy, 77 F.3d 522, 535 (1st Cir. 1996). The government argues that Tejeda has forfeited his claim that the district court erred in applying the crack cocaine Sentencing Guidelines, and that review therefore should be only for plain error. See Fed. R. Crim. P. 52(b); see also Olano, 507 U.S. at 732-35. The government notes that Tejeda failed to object to the district court's limiting of the verdict slip to refer only to "cocaine base" and not "crack." Nor did he object to the district court's statement at the charge conference that it would decide the crack issue in a jury-waived proceeding, or the district court's instruction to the jury, which equated crack with cocaine base. The prosecution argues that Tejeda's presentencing motion to preclude application of the crack Guidelines was insufficient to preserve the claim because it was at that time too late to present the issue to the jury. See United States v. Pacheco, 434 F.3d 106, 115 (1st Cir. 2006).
We bypass the question of forfeiture because even if Tejeda preserved his claim, it fails on the merits. As Tejeda conceded before the district court, our decision in United States v. Medina, 427 F.3d 88 (1st Cir. 2005), forecloses any argument that as a general matter the government must prove to the jury beyond a reasonable doubt that the cocaine base at issue is crack cocaine. In Medina, the defendant was convicted of, inter alia, possession with intent to distribute over fifty grams of cocaine base. Id. at 90. On appeal, Medina argued that the trial court had erred in not instructing the jury that the government was required to prove that the cocaine base he possessed was crack cocaine.7 Id. at 92. This court held that "the government is not required to prove that the substance involved in a given case is crack in order to secure a conviction under" 21 U.S.C. § 841, the statute specifying the substantive offense that was the object of the conspiracy for which Tejeda was convicted. Id. The fact that the cocaine base was crack was "only relevant to the court at sentencing." Id. at 92 n.3. Moreover, in United States v. O'Brien, 435 F.3d 36 (1st Cir. 2006), we held that, as long as the statutory maximum is not affected, a sentencing judge is permitted to determine by a preponderance of the evidence the factual basis for a sentencing enhancement. Id. at 41. Such was the case here. There was no error in the district court's determining that the cocaine base at issue in this case was crack.8
Tejeda argues that the sentencing court was not permitted to make the crack determination in his case because it was the law of the case that all sentencing enhancements be proved to the jury beyond a reasonable doubt. This argument is without merit. At times the court did state that it would require the government to prove sentencing enhancements to the jury beyond a reasonable doubt; but it also determined that whether the cocaine base at issue was crack cocaine would be decided in a jury-waived proceeding, and it further decided at sentencing that it was permitted to make the determination that the cocaine base was crack. The court was not prohibited from revising its position on proof of sentencing enhancements to the jury. See Harlow v. Children's Hosp., 432 F.3d 50, 55 (1st Cir. 2005) ("[L]aw of the case permits a lower court to review prior interlocutory orders as long as that review is not an abuse of discretion."); Fiori v. Truck Drivers, Local 170, 354 F.3d 84, 90 (1st Cir. 2004) (stating that law of the case doctrine does not bar the district court from revising a prior ruling, as long as prejudice does not result).
Tejeda argues that he was prejudiced by the district court's change of position because (1) the court never conducted a jury-waived proceeding to determine whether the cocaine base at issue was crack cocaine and (2) the court had promised this jury-waived proceeding, and as a result, the trial record, on which the district court relied in determining that the cocaine base was crack, was underdeveloped with respect to the crack issue.
Tejeda is correct that the district court repeatedly stated that the government would not be able to seek a sentencing enhancement for crack unless it had pled and proved the issue to the jury. It is possible that the court's about-face at the sentencing disposition9 might have constituted "unfair surprise." See, e.g., United States v. Moody, 903 F.2d 321, 331 (5th Cir. 1990) ("There is also an element of unfair surprise in the court's belated reversal concerning Bauman's testimony. . . . By reversing its earlier ruling, . . . the district court was obliged to provide Moody an opportunity for surrebuttal, assuming of course that the defense could proffer evidence to parry the expanded government case.") It would have been better for the court here to have explicitly informed the parties of its ruling that it could determine the crack enhancement in the absence of a jury verdict and allowed the parties to submit factual evidence with regards to the enhancement.
The evidence that Tejeda states he would have offered, however, would not have precluded the court from finding that the crack Guidelines were applicable. Prior to sentencing, Tejeda submitted a brief to the district court in which he claimed -- in a footnote -- as he does now, that the DEA's test results showed that the drug sample linked to Tejeda had not tested positive for sodium bicarbonate, a "signature ingredient" of crack. Tejeda did not purport to have performed any independent tests of the drug sample connected to him, but rather relied solely on the DEA's analysis. Tejeda attached to his brief the DEA's work papers and test results. As a result, at the point it made its determination that the drugs at issue in the case were crack, the district court had before it the very results on which Tejeda's expert would have offered testimony. Moreover, even if Tejeda is correct that the particular sample tested by the DEA did not contain sodium bicarbonate, there nonetheless was ample evidence at trial for the district court to conclude that the drugs involved in the conspiracy were crack. Three law enforcement officers testified, based on visual observation, that the drugs seized from the car on March 16 were crack, and Eldridge and Figueroa both testified that the drugs regularly purchased from Tejeda were crack. See United States v. Walters, 904 F.2d 765, 770 (1st Cir. 1990) ("Proof based on scientific analysis or expert testimony is not required to prove the illicit nature of a substance, and identification of a substance as a drug may be based on the opinion of a knowledgeable lay person."). The identity of the drugs that Tejeda dealt does not hinge on this one sample.
As to Tejeda's second argument, it is unpersuasive for Tejeda to suggest that he would have presented an expert at trial had he known what would be the nature of the jury-waived proceeding. Tejeda did attempt to have his expert testify at trial. In his proffer Tejeda indicated that the expert would testify about the difference between cocaine base and crack cocaine and the difficulty of telling the two apart visually. The district court, however, refused to allow the expert to testify because the request was not timely,10 and because, the court concluded, the testimony would confuse the jury. The court did say at the time it precluded the expert from testifying that it would hear the expert at the time of sentencing. But this was not the reason that the expert was precluded from testifying at trial.11
The jury was not required to find that the cocaine base possessed by Tejeda was crack. The district court did not abuse its discretion when it revised its position on proof of sentencing enhancements. Tejeda was not prejudiced by the district court's revision of its prior ruling.
Finally, Tejeda argues that the district court's deciding that the crack cocaine Guidelines applied violated his protection against double jeopardy and his due process rights. Tejeda essentially argues that the government failed to prove at trial that the drugs were crack, and that he should not have been forced at sentencing to once again mount a defense to such a charge. There was no error in the district court's application of the crack cocaine Guidelines, much less a due process violation. Likewise, double jeopardy principles are inapplicable here. Double jeopardy concerns arise only when a jury verdict or a trial court's ruling, "whatever its label, actually represents a resolution, correct or not, of some or all of the factual elements of the offense charged." Pacheco, 434 F.3d at 112 (quoting United States v. Martin Linen Supply Co., 430 U.S. 564, 571 (1977)) (internal quotation marks omitted). There was no such trial court ruling here. Whether or not the cocaine base at issue was crack is not a factual element of the offense charged. Medina, 427 F.3d at 92 & n.3.
B. Reasonableness of Tejeda's Sentence
Tejeda argues that his sentence is unreasonable as a matter of law. See United States v. Jiménez-Beltre, 440 F.3d 514, 519 (1st Cir. 2006) (en banc). He argues that the nature and circumstances of his offense, his personal history and characteristics, his potential for rehabilitation, sentences imposed by other federal courts, and the sentences imposed on his co-defendants all weigh in favor of his receiving the mandatory minimum sentence of 120 months. Tejeda's counsel made all of these arguments to the district court at his sentencing hearing.
The district court explained its reasons for imposing a twenty-year sentence. It discussed the fact that Tejeda was "the supplier in a massive [crack] distribution ring" and stated that "crack cocaine is one of the most addictive, dangerous substances known to our society." The court stated that its sentence took rehabilitation into account, although the court did not think rehabilitation particularly likely,12 and it referred to Tejeda's family circumstances, concluding that while they were tragic, they had been caused by Tejeda. The court further stated that comparison to the average sentence imposed by other courts for the offense in question was not necessarily instructive since that average did not account for factors such as drug quantity. Finally, we note that "[a]lthough a district court may consider disparities among co-defendants in determining a sentence, [a defendant's] sentence [is not] unreasonable simply because his co-defendants agreed to help the government in exchange for reduced sentences." United States v. Vázquez-Rivera, 470 F.3d 443, 449 (1st Cir. 2006). Tejeda's sentence is not unreasonable.
Tejeda's conviction and sentence are affirmed.
1. As the Ninth Circuit has noted, it would be inconsistent with the Supreme Court's decision in Remmer v. United States, 347 U.S. 227 (1954), discussed below, to apply structural error review to these sorts of jury bias issues. See United States v. Dutkel, 192 F.3d 893, 899 n.4 (9th Cir. 1999).
2. Tejeda claims that he made this argument below, but we are unable to find support for that assertion in the record.
3. Similarly, in O'Brien, we applied a presumption of prejudice when a police officer who had been summoned to testify on behalf of the government (but who had not yet testified) engaged in conversation with three jurors in the hallway outside the courtroom during a recess. 972 F.2d at 13.
4. Similarly, where the claim is that jurors have been tainted by exposure to improperly admitted inflammatory evidence, our rule is that jurors are presumed to properly follow curative instructions. See United States v. Sepulveda, 15 F.3d 1161, 1185 (1st Cir. 1993). That presumption is overcome only by a showing that it is probable that (1) responsible jurors will be unable to disregard the evidence, and (2) the evidence likely will have a seriously prejudicial effect on the defendant. Bradshaw, 281 F.3d at 285.
5. Figueroa had informed the court and counsel of this defense when she moved on April 27, 2005 to continue the trial so she could prepare such a duress defense. The court denied the motion for a continuance.
6. Although unusual, bifurcation has been used. See United States v. Zarnes, 33 F.3d 1454, 1471-72 (7th Cir. 1994); United States v. Joshi, 896 F.2d 1303, 1306-09 (11th Cir. 1990).
7. The issue in Medina was different. Medina argued that the jury instruction was error because the statute at issue, 21 U.S.C. § 841, regulated only possession of crack, not cocaine base more generally. 427 F.3d at 92.
8. Tejeda does not dispute that there was sufficient evidence presented at trial to conclude that the drugs at issue were crack.
9. The district court did in fact conduct a jury-waived proceeding on the crack issue on September 22, 2005. At that proceeding, Tejeda offered an expert to testify to the difference between crack and other forms of cocaine base. However, the parties all agreed that crack was merely one form of cocaine base, and there was no factual dispute as to whether crack and cocaine base were one and the same. Tejeda also offered his expert to cast doubt that the drugs seized on March 16, 2004 from the car in which Tejeda had been riding were crack. He proffered that his expert would testify that while the DEA had concluded that a different drug sample -- seized from co-conspirator Custer's home -- was crack, the DEA had not concluded that the drugs linked to Tejeda were crack. This was an inaccurate characterization of the DEA's conclusions, as the prosecution pointed out. The parties had stipulated that the DEA had concluded that both tested samples contained cocaine base. The DEA had made no finding as to whether either sample contained crack cocaine. As a result, Tejeda's expert's testimony would not have been informative on any factual dispute. The district court solicited additional briefs from the parties and stated that if necessary, it would convene an evidentiary hearing at some point in the future.
10. Tejeda conceded this in the district court.
11. Tejeda also refers to thwarted attempts to cross-examine the government's witnesses about the differences between crack and cocaine base. Tejeda was precluded from questioning government witnesses about the chemical differences between crack and other types of cocaine base because the witnesses in question were not chemists and were not qualified to testify on the subject.
12. We disagree with Tejeda's assertion that the district court did not consider his particular potential for rehabilitation. The court stated that the sentence did not "emphasize rehabilitation due to the nature of the offense." We understand this to mean that given the particulars of Tejeda's offense, not drug offenses in general, the court did not think rehabilitation particularly likely.