USA v. Straub, No. 07-30182 (9th Cir. 2008)

Annotate this Case

Download PDF

FOR PUBLICATION UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT UNITED STATES OF AMERICA, Plaintiff-Appellee, v. DWAYNE EDWARD STRAUB, Defendant-Appellant. ï£¼ ï£½ ï£¾ No. 07-30182 D.C. No. CR-04-00020-6OMP OPINION Appeal from the United States District Court for the District of Oregon Owen M. Panner, District Judge, Presiding Argued and Submitted June 3, 2008 Pasadena, California Filed August 15, 2008 Before: Dorothy W. Nelson, Andrew J. Kleinfeld, and Jay S. Bybee, Circuit Judges. Opinion by Judge Bybee 10679 10682 UNITED STATES v. STRAUB COUNSEL Jeff S. Pitzer, Portland, Oregon, for the appellant. Kelly A. Zusman, Assistant United States Attorney, United States Attorney for the District of Oregon, Portland, Oregon, for the appellee. UNITED STATES v. STRAUB 10683 OPINION BYBEE, Circuit Judge: Dwayne Edward Straub challenges his conviction and sentence for narcotics crimes and the attempted robbery and shooting of Robert Garrett in Portland, Oregon. Straub claims that the district court s refusal to compel the prosecution to grant use immunity to defense witness Mike Baumann violated his due process rights under the Fifth Amendment. This case requires us to clarify the standard by which we determine when a district court must compel the prosecution to grant use immunity, as most recently stated in Williams v. Woodford, 384 F.3d 567 (9th Cir. 2004). We must address the question left open by Williams, whether a defendant requesting compelled use immunity on the ground that his witness has relevant testimony that directly contradicts that of an immunized prosecution witness must prove that the prosecution s purpose in denying use immunity to the defense witness was to distort the fact-finding process, or merely that the prosecution s selective denial of use immunity had the effect of distorting the fact-finding process. See id. at 600-01. We have jurisdiction under 28 U.S.C. § 1291, and for the reasons set forth below, we reverse the district court s refusal to compel use immunity and remand for further proceedings. I A Straub was arrested on February 6, 2003, following the execution of a search warrant at his residence in Oregon City, Oregon. Police found marijuana plants and packaged marijuana at Straub s home. Further investigation uncovered evidence that Straub was involved in a wide-ranging and longstanding conspiracy to manufacture and distribute methamphetamine. As charged in the indictment, Straub was involved 10684 UNITED STATES v. STRAUB in the sale and distribution of both marijuana and methamphetamine for a continuous period between 1998 and September 2004. Straub was part of a gang known as The White Neck Crew or just The Crew. Straub and his associates were in the business of unlawfully entering the residences of other drug dealers in order to steal cash and drugs for later distribution. In the many robberies they committed over a five-year period, Straub and others wore body armor and displayed firearms. The indictment alleged, inter alia, that on February 8, 2003, Straub and another person carried and used a firearm in connection with an attempt to rob Robert Garrett and take more than 100 marijuana plants. Straub allegedly discharged a firearm in connection with this robbery. On November 23, 2004, Straub was charged in a Second Superseding Indictment with conspiracy, possession with intent to distribute, and manufacture of methamphetamine and marijuana.1 Of particular relevance to this appeal, Counts 3 and 4 of the indictment related to the carrying, using, and discharging of a firearm in connection with the attempted robbery of Robert Garrett in his residence on February 8, 2003. At trial, the prosecution relied heavily on the testimony of David Adams. Adams was an associate of Straub s who, at 6 ² 5 ³ and 365 pounds, was known as Big Mix. According to the prosecution s own stipulation, Adams admitted to participating in: (1) the attempted robbery of the home of Robert 1 Straub was charged as follows: Count 1, conspiracy to manufacture and distribute marijuana and methamphetamine, 21 U.S.C. §§ 841(a)(1), 21 U.S.C. § 846(b)(1)(A)(iii), 21 U.S.C. 846(b)(1)(B)(ii); Count 2, manufacturing marijuana, 21 U.S.C. § 841(a)(1); Count 3, carrying, using and discharging a firearm in relation to a drug trafficking crime causing physical injury to another person, 18 U.S.C. § 924(c)(1)(A)(iii); Count 4, attempted possession with intent to distribute more than 100 marijuana plants, 21 U.S.C. § 841(a)(1), 841(b)(1)(B)(vii), 846; Count 5, possession with intent to distribute more than 350 grams of a mixture containing methamphetamine, 21 U.S.C. § 841(a)(1), 841(b)(1)(B)(viii); and Count 6 possession with intent to distribute more than 50 grams of actual methamphetamine, 21 U.S.C. § 841(a)(1), 841(b)(1)(B)(viii). UNITED STATES v. STRAUB 10685 Garrett, during which Garrett was shot in the chest; (2) an attempt to steal 50 or 60 pounds of marijuana from the home of Jacob Adams, during which a gun was discharged; and (3) a series of successful robberies of large quantities of cash, methamphetamines, and marijuana from homes of local drug dealers. In exchange for his testimony against Straub, the prosecution granted Adams use immunity2 for these crimes.3 Adams testified that he participated with Straub in the attempted robbery of Garrett, and that Straub was the person who shot Garrett. Adams was the only witness for the prosecution who could place Straub at the scene, and the only witness who could put the gun in Straub s hand. In a colloquy held outside of the presence of the jury, Straub s attorney explained to the district judge that he wanted to impeach Adams by introducing a prior inconsistent statement, but that the defense witness who could testify as to the inconsistent statement wanted to assert his Fifth Amendment privilege against self-incrimination. To demonstrate, Straub s attorney asked Adams the question he hoped to ask on the stand: Mr. Adams, did you have a conversation with Mike Bauman[n] at a bar in the winter of 2003 in which you admitted to him that you had just shot a man? Adams responded, No, I didn t. Straub s attorney then stated that Adams response was the statement he planned to impeach through the testimony of Mike Baumann. The following colloquy ensued: 2 The government s power to grant a witness use immunity is conferred by 18 U.S.C. §§ 6002 and 6003. Use immunity means that, while the government may prosecute the witness for an offense related to the subject matter of the witness s testimony, the testimony itself and any fruits thereof may not be used against the witness in any criminal case except a prosecution for perjury arising out of the testimony. United States v. Lord, 711 F.2d 887, 890 (9th Cir. 1983). 3 We discuss the details of the immunity agreement and other incentives to testify provided by the prosecution in greater detail below. 10686 UNITED STATES v. STRAUB [Defense]: I would call Mr. Bauman[n] . . . and it is my expectation he would testify that he saw Mr. Adams at one of the night clubs . . . and Mr. Adams was glum and sullen, and Mr. Bauman[n] asked him what s the matter, and Mr. Adams said, I just shot a man. The Court: Well, I don t know that that is relevant at this point. I mean, he could have shot somebody else. [Defense]: I would establish the timeframe, Your Honor, by the winter of 2003, which is consistent with the testimony we ve been hearing about the shooting of Robert Garrett on February 8, 2003. The Court: I m not suggesting that he shot anybody, but clearly it s possible that he could have shot several people about that time period. This is a wild story we re hearing all through the whole case, and I can t conclude from that that he s talking about this case. [Defense]: Well, the jurors could draw that inference, though, Your Honor, and I think it s a fair inference, and in defending my client I think I have the right to present that evidence. *** [Prosecution]: I believe Mr. Bauman[n] is a fellow gang member of the defendant Straub and was identified and with him as a gang member in a gang investigation in 92. The Court knows that gang members will take the stand and lie for one another and that that is a proper subject of testimony in the case. And if Mr. Bauman[n] would come to testify, and he certainly may, I intend to examine him: UNITED STATES v. STRAUB 10687 Aren t you a member of the gang with which the defendant belongs. And aren t you here lying for him as a gang member and dealing with it at that time. I think that s appropriate. [Defense]: And, Your Honor, I m not sure, I d let [Mr. Baumann s lawyer] address this. I m not sure that that would raise the Fifth Amendment concerns if that is as far as the cross-examination went. So *** [Mr. Bauman s Attorney]: My client does wish to assert a Fifth Amendment privilege. In this case the Government has alleged a far-reaching conspiracy and The Court: You don t have to go any further. Based on [the prosecutor s] statement about what he intends to cross-examine about, it s clear that your client could be subjected to all kinds of criminal proceedings including perjury if he doesn t testify accurately. So, I I m I guess the ruling properly is that he can come and testify, but if he does and takes the Fifth Amendment, I will allow him to remain silent. I don t think [Defense counsel] hasn t asked me and I doubt that I would be justified in granting him any kind of immunity to testify. [Defense]: That is my last resort, Your Honor, if you would grant immunity under 18 U.S.C. 6001, the Government has that opportunity and it s just inherently unfair for the defendant to have no access to the same power. The Court: Under certain circumstances I could grant immunity, but I don t think that I should based 10688 UNITED STATES v. STRAUB on what you ve said in your offer of proof. I don t I don t think that s appropriate. So that s my ruling. It s on the record. You can decide whether you re going to call him or not. *** [Defense]: I don t intend to call the witness who is going to take the Fifth Amendment on the stand in front of the jury, Your Honor. And I don t mean to waive my claim to present his evidence, but I do think it would be prejudicial to put on a witness and take the Fifth. Following a five-day trial, Straub was convicted on Counts 1, 3, 4, 5, and 6. On January 23, 2006, the district court sentenced Straub to 152 months on the drug charges (Counts 1, 4, 5, and 6) and an additional statutorily mandated consecutive term of 120 months on the firearms charge (Count 3), pursuant to 18 U.S.C. § 924(c)(1). B Straub timely appealed to this court for the first time on February 21, 2006. In that appeal, Straub challenged: (1) the district court s denial of a motion to sever Counts 3 and 4 of the indictment from the rest of the trial; and (2) the district court s denial of a request to compel the prosecution to grant use immunity to defense witness Baumann. In an unpublished memorandum, we held that the district court did not abuse its discretion in denying the motion to sever. See United States v. Straub, 224 Fed. App x 633, 2007 WL 731348 (9th Cir. March 12, 2007). Considering the district court s refusal to compel use immunity, we remanded for an evidentiary hearing because we were unable to determine from the record whether or not compelled use immunity was constitutionally required. As we characterized Straub s position, [Straub] does not allege prosecutorial misconduct, but he does main- UNITED STATES v. STRAUB 10689 tain that immunity was warranted because the government gave immunity or special considerations to many of its witnesses, while refusing immunity for Bauman[n]. Id. at 635. We stated that on remand, Straub would need to show, inter alia, that the government granted immunity to certain witnesses while denying immunity to a witness who would have directly contradicted testimony from the immunized government witness. Id. (citing Williams, 384 F.3d at 600). We suggested that the district court gather [i]nformation on why Bauman[n] required immunity, greater detail about his proposed testimony and the immunity agreements the government gave to its other witnesses. Id. 1 On April 24, 2007, on remand from this court, the district court held an evidentiary hearing. The court meticulously followed our instructions and gathered the information suggested by our order. First, the district court cleared the courtroom of everyone except Baumann and his counsel and took in camera testimony as to why Baumann required immunity. This portion of the record remains under seal. Although Baumann s counsel provided more detail in the sealed portion of the record, his statements only elaborated on the arguments made in the presence of all parties counsel when this issue was first discussed at Straub s trial in 2006. In short, the prosecution had stated that if Baumann took the stand, it intended to attempt to impeach him by soliciting testimony about Baumann s association with Straub and his membership in the gang. Given the wide-ranging nature of the conspiracy as alleged in the prosecution s indictment, Baumann considered it necessary to invoke the Fifth Amendment to prevent his testimony to his gang membership being used against him in a subsequent prosecution for acts of conspiracy related to those for which Straub was on trial. The district court also asked the prosecution what else it knew about Baumann that would have caused it to prosecute 10690 UNITED STATES v. STRAUB him. The prosecutor stated that Baumann was found with Straub on one occasion committing a criminal offense, and that another witness was willing to testify that Baumann had loaned Straub money. However, the prosecutor insisted: Baumann was not in any way a target of the criminal prosecution . . . . Baumann was nowhere on the radar screen. He was nowhere of any concern to the Government as a potential defendant. . . . Baumann was not a target. We weren t after Baumann. We had no interest in Baumann as a criminal defendant, and we had no evidence that Baumann was involved in this conspiracy at all. None whatsoever. The district responded, Why didn t you immunize him? To which the prosecutor responded, I wasn t asked. Next, the district court gathered evidence on the grants of immunity given by the prosecution to Adams and other government witnesses. This evidence was provided by an Amended Joint Stipulation of Facts entered into by the parties. According to the Stipulation, despite Adams having admitted to attempted armed robbery and numerous other thefts of cash and drugs, Adams was provided use immunity except with respect to violent crimes in exchange for his testimony. Although the formal agreement did not grant immunity for violent crimes, it appears Adams was given informal immunity, as he was never prosecuted for any of his violent crimes. Adams was charged with a single count of intimidation of a federal witness, in violation of 18 U.S.C. § 1512. For this one crime, Adams pleaded guilty, and in the plea agreement, the government agreed not to file or seek any new or additional charges against [Adams] . . . arising out of the investigation of this case. At sentencing for this conviction, the judge noted that Adams sentence represented a downward departure of three or four or five years because of [his] cooperation with the government. According to the Stipulation, Adams was released from prison on October 26, UNITED STATES v. STRAUB 10691 2006, while Straub s projected release date, stemming from the convictions which Adams testimony helped secure, was February 26, 2024. The Joint Stipulation also provided evidence as to the incentives use immunity, informal immunity, sentence reductions, and even cash offered to the prosecution s other witnesses for their testimony against Straub. The Stipulation is long, and we need not recount every detail. In summary, of the thirteen witnesses the government and Straub called or intended to call, eleven of the twelve government witnesses received some kind of immunity or other benefits. Most of these witnesses admitted to significant federal crimes. For example, Dustin Alsup testified that he: (1) partnered with another gang member in drug trafficking; (2) was arrested after a Tupperware container full of methamphetamines was found in his freezer; (3) grew a large amount of marijuana plants; (4) manufactured methamphetamines in a house in the area; (5) continued to make and distribute methamphetamines in another location; (6) sold a minimum of 40 pounds of methamphetamines for at least $200,000 and a minimum of 40 pounds of marijuana for at least $100,000; (7) bought drugs from Straub; and (8) manufactured pounds and pounds of methamphetamines over the years. Shortly after he was arrested, Alsup entered into a proffer agreement with the government which provided him with use immunity. In addition, Alsup agreed to plead guilty for possession with intent to distribute at least 15 grams of methamphetamines, which carried a 10-year minimum sentence. Despite his testimony that he had manufactured and sold great quantities of drugs, and despite his pleading to a crime with a 10-year minimum sentence, Alsup was eventually sentenced to only 30 months in prison, all in exchange for his cooperation with the government. As another example, Jacob McCord testified that he (1) was a methamphetamine cook for several years; (2) assisted Straub in growing marijuana at Straub s house; (3) stole 40 10692 UNITED STATES v. STRAUB marijuana plants from Straub; (4) used drugs; (5) sold 50 pounds of methamphetamine to Straub for $300,000; (6) purchased chemicals for use in manufacturing methamphetamines; (7) had 120 grams of methamphetamine seized from his house by police; and (8) obtained chemicals for further methamphetamine manufacturing while on pretrial release. McCord knew he faced a Guidelines-recommended sentence of 135 to 168 months for the 120 grams of methamphetamines alone, and that he faced a minimum of 10 years under the drug statute. In exchange for his testimony against Straub, McCord was permitted to plead guilty to possession with intent to distribute, and, following his testimony against Straub, his sentence was reduced to 63 months in prison. McCord was also never charged with possessing a loaded .38 caliber pistol found on the floor of one of his residences. As a final example, Jason Shaw testified to being a methemphetamine dealer for a few months, a big drug dealer for a couple of years, growing several marijuana plants, and using drugs himself. Shaw was paid a total of $8,300 for providing assistance to state and federal governments in drug investigations. He was also given informal immunity and expects not to be prosecuted for any crime. In addition to these three and David Adams, the government granted immunity or some other kind of benefit to Grant McFarland, Jeff Center, Anita Wealand, Lisa Clark, Cory Don Smith, Joseph Mockley, and Jacob Adams. The government did not grant immunity to one of its witnesses, Misty Dawn Day, who testified that she is a drug user. The one defense witness listed in the Stipulation, Mike Baumann, was denied use immunity. Finally, the district court moved on to the issue of whether Baumann s testimony would have directly contradicted Adams testimony. Adams was prepared to testify that he did not walk into a bar in the winter of 2003 and tell someone that he shot a man. Baumann was prepared to testify that on February 8, 2003, Adams walked into a club, acting very glum UNITED STATES v. STRAUB 10693 and quiet and morose, and told Baumann, I just shot a man. 4 Straub s counsel5 explained the importance of this contradiction: If that evidence had been heard, if that testimony had been heard by the jury, [the defense counsel] could have argued that based on that[,] Mr. Adams was himself perhaps the shooter of Robert Garrett. That perhaps his credibility was so suspect, if he s willing to lie about the key element of his testimony, the shooter who shot Robert Garrett, if he s lying about that, the jury could have discredited his testimony in its entirety. . . . [T]here were already a host of credibility problems with Mr. Adams, as the Government admits. He admitted to lying to the police, he admitted to lying to the Government when it helps him, especially when he could help himself. He admitted to threatening to beat up his own grandmother. He admitted to numerous drug and violent crimes. His credibility was already hanging in the balance, Judge. And if Michael Baumann could testify, it could have been the tipping point that the jury would have used to discredit Adams testimony in its entirety. And if they had done that, they likely would have acquit[t]ed on Counts 3 and 4[, ] having to do with the Robert Garrett shooting. And that resulted in Mr. 4 Baumann himself never testified, either at trial or in a colloquy outside of the presence of the jury. At different hearings, the parties attorneys described what Adams told Baumann variously as I just shot a man, I just shot a guy, and I ve shot somebody. These discrepancies are irrelevant to this appeal. 5 Straub was represented by new counsel beginning with the first appeal to our court. 10694 UNITED STATES v. STRAUB Straub getting 10 extra years in prison. Instead of 13 years he gets 23 years. So he has to sit in prison for ten extra years wondering every day what might have happened if Michael Baumann had testified. The prosecution made several arguments in rebuttal. It argued that Adams was not the star witness Straub made him out to be. Because the trial judge gave a Pinkerton instruction,6 the jury did not need to find that Straub held the gun in order to hold him liable. This, the prosecution argued, downplayed the importance of Adams being the only witness to place the gun in Straub s hand the jury could have convicted whether it was Straub or Adams who pulled the trigger. The prosecution also argued that the testimony of Misty Dawn Day supported its case on Counts 3 and 4. Day testified that, while high on marijuana, she overheard Straub and Adams planning the robbery through a closed door. The prosecutor conceded that the government had given numerous incentives to many witnesses to testify against Straub, but he represented that there was no intent to distort the fact-finding process. The prosecutor also reiterated that he was never formally asked to give immunity to Baumann. 2 On May 11, 2007, the district court entered an order again denying Straub s request to compel the government to offer use immunity to Baumann. The court recounted that the prosecution represented that it had no interest in Baumann as a criminal defendant . . . . None whatsoever. At the same time, the court noted, the prosecutor had asserted that Baumann was found with this defendant . . . committing a criminal offense . . . . And Baumann was in fact . . . a gang member, a friend of the defendant. The prosecutor had also made clear that his cross-examination of Baumann would begin by drawing out information about Baumann s gang membership and 6 Pinkerton v. United States, 328 U.S. 640 (1946). UNITED STATES v. STRAUB 10695 willingness to do anything for Straub. Under these circumstances, the district court re-affirmed its prior ruling that Baumann s assertion of privilege was proper, citing Kastigar v. United States, 406 U.S. 441, 445 (1972), and United States v. Hoffman, 341 U.S. 479, 486 (1951).7 Turning to the issue of whether the prosecution s conduct had distorted the fact-finding process, the district court expressed some hesitation as to the Ninth Circuit law on this issue. The court stated that it could not be sure from our precedents in Williams, Young, Baker, and Westerdahl whether, in order to compel use immunity, Straub had to prove that the denial of use immunity to Baumann implicated prosecutorial misconduct. See Williams, 384 F.3d at 600; United States v. Young, 86 F.3d 944, 949 (9th Cir. 1996); United States v. Baker, 10 F.3d 1374, 1415 (9th Cir. 1993), overruled on other grounds, United States v. Nordby, 225 F.3d 1053 (9th Cir. 2000); United States v. Westerdahl, 945 F.2d 1083, 1087 (9th Cir. 1991). The district court was unclear as to whether Straub must prove that the prosecution acted with the purpose of distorting the fact-finding process, or merely that the prosecution s conduct had the effect of distorting the fact-finding process. Therefore, the court made alternative rulings under each theory. The court ruled that if prosecutorial conduct is a required element that the prosecution acted with the purpose of distorting the fact-finding process then Straub s claim failed. As we had noted in our memorandum, Straub did not allege prosecutorial misconduct. See Straub, 224 Fed. App x at 635. Thus, he was not entitled to compel use immunity under a purpose theory. 7 We recount the district court s ruling on whether Baumann properly invoked the Fifth Amendment only to enrich the record. The propriety of the witness s invocation of the Fifth Amendment is not a prong of the test for compelled use immunity. See Williams, 384 F.3d at 600. Our opinion should not be construed as adding any such prong. 10696 UNITED STATES v. STRAUB The court then considered Straub s claim under the theory that a showing of the mere effect of distorting the fact-finding process was sufficient. The court found that the testimony of David Adams was crucial to the prosecution s case on Counts 3 and 4. Adams was the only witness who testified that Straub was armed on the night of the robbery. Adams also testified that, although he did not see Straub shoot Garrett, he heard the gunshot, and that he saw Straub fire other shots that night and during a previous robbery. The district court acknowledged the prosecution s argument that because of the Pinkerton instruction, the jury could have theoretically convicted Straub even if it did not believe he was the shooter. Nevertheless, the court ruled, the testimony of Adams was important to the prosecution s case. If a jury believed that Adams lied about this, and was shifting the blame to Straub for his own actions, it would seriously weaken the government s case on counts three and four . . . . The defense could have argued that Adams, not Straub, orchestrated the Garrett robbery and shooting, without Straub s involvement or knowledge. The district court also considered the prosecution s arguments about Misty Dawn Day, but found that [D]ay s testimony, by itself, would not be enough to sustain a conviction on counts three and four. The district court then ruled that Straub s claim failed because the testimony of Mike Baumann did not directly contradict that of Adams. See Williams, 384 F.3d at 600 (holding that the testimony must have directly contradicted testimony from the immunized government witness). The court noted that Adams was prepared to deny that on the night in question he told Baumann, I just shot a guy, and that Baumann was prepared to testify that Adams told him, I just shot a guy. The court ruled, however, that even if Adams did say, I just shot a guy, this did not prove he actually shot someone. Adams might have been joking, or trying to appear tough, or indicating he was merely present when Garrett was shot. Furthermore, the court ruled, even accepting the Baumann proffer literally, it does not prove Adams was referring UNITED STATES v. STRAUB 10697 to Garrett. The court concluded that Baumann s testimony would have given the defense some additional ammunition in attacking the credibility of Adams, but would not directly contradict the most important testimony by Adams, nor clearly exonerate Straub on these two counts. The court declined to compel use immunity. Straub appealed for a second time. II The question of whether a district court erred by refusing to compel use immunity is a mixed question of law and fact that we review de novo. United States v. Alvarez, 358 F.3d 1194, 1216 (9th Cir. 2004). Factual findings underlying the district court s ruling are reviewed for clear error. Id. [1] There are important separation of powers concerns present in a request to compel immunity: It has long been recognized that the Executive Branch of government has exclusive authority and absolute discretion to decide whether to prosecute a case . . . . To interpret the Fifth and Sixth Amendments as conferring on the defendant the power to demand immunity for co-defendants, potential codefendants, or others whom the government might in its discretion wish to prosecute would unacceptably alter the historic role of the Executive Branch in criminal prosecutions. Of course, whatever power the government possesses may not be exercised in a manner which denies the defendant the due process guaranteed by the Fifth Amendment. United States v. Alessio, 528 F.2d 1079, 1081-82 (9th Cir. 1976) (internal quotation marks and citations omitted). Our test for when due process requires that the district court impede on the discretion of the executive branch and compel 10698 UNITED STATES v. STRAUB use immunity has evolved over time. In Williams, we stated the test as follows: [T]he prosecution s refusal to grant use immunity to a defense witness denies the defendant a fair trial only when (1) the witness s testimony would have been relevant, and (2) the prosecution refused to grant the witness use immunity with the deliberate intention of distorting the fact-finding process. . . . To demonstrate the prosecutorial misconduct of the second prong, [the defendant] must show that the prosecution intentionally caused a defense witness to invoke the Fifth Amendment right against selfincrimination, or that the prosecution granted immunity to a government witness in order to obtain that witness s testimony, but denied immunity to a defense witness whose testimony would have directly contradicted that of the government witness. 384 F.3d at 600 (citations omitted and emphases added). As the district court observed, the facts of this case have isolated an ambiguity in our test as stated in Williams. The second prong of the test is that the prosecution refused to grant the witness use immunity with the deliberate intention of distorting the fact-finding process. Id. (emphasis added). Williams stated that prong may be satisfied in one of two ways. One would expect that both methods are designed to prove deliberate intention. However, the two alternative methods of showing misconduct differ in that only the first describes conduct with a particular intent: the prosecution intentionally caused a defense witness to invoke his Fifth Amendment rights. Id. The second merely requires showing that the prosecution granted immunity to a government witness in order to obtain that witness s testimony, but denied immunity to a defense witness whose testimony would have directly contradicted that of the government witness. Id. The Williams test uses the word or between the two methods for UNITED STATES v. STRAUB 10699 proving the second prong of the test, suggesting that the two methods are disjunctive; only one need be proven, not both. Both parties agreed to this at oral argument. Thus, our test as it has been refined over time suggests that one may prove that the prosecution acted with a certain purpose merely by demonstrating that the prosecution committed a set of acts (the selective denial of use immunity described) that had the effect of distorting the fact-finding process. Straub has conceded that the prosecution did not act with the purpose of distorting the fact-finding process. He argues that he is entitled to a grant of use immunity because the government s conduct otherwise met the Williams test and had the effect of distorting the fact-finding process. In order to determine whether Straub is correct, we first work through the first prong and first alternative method for proving the second prong as they apply to Straub. We then turn to the cases leading to Williams for guidance in interpreting Williams. Because, as discussed below, we find that our precedents lead to the conclusion that a showing of effects is sufficient, we then apply this method of proving the second prong of the Williams test to Straub s claim. A The first prong under the Williams test is that the witness s testimony would have been relevant. 384 F.3d at 600. We have described the relevance requirement for purposes of the Williams test as minimal. Id. The defendant need not show that the testimony sought was either clearly exculpatory or essential to the defense. Westerdahl, 945 F.2d at 1086 (internal quotation marks omitted). In our previous memorandum disposition, we found that Baumann s testimony was relevant because it raised credibility questions about a key prosecution witness. See 224 Fed. App x at 635. As the district court put it, Baumann s testimony would have given the defense some additional ammunition in attacking the credibility of Adams. If the jury credited Baumann, it could have found that Adams 10700 UNITED STATES v. STRAUB was a perjurer and might even be lying to shift the blame for his own conduct on to Straub. Straub satisfies the first prong. [2] The second prong under the Williams test is that the prosecution refused to grant the witness use immunity with the deliberate intention of distorting the fact-finding process. Williams, 384 F.3d at 600. As the first of two alternative methods, the defendant may satisfy this prong by showing that the prosecution intentionally caused a defense witness to invoke the Fifth Amendment right against self incrimination. Id. This method includes the word intentionally. In one sense, any time the prosecution causes a witness to invoke his Fifth Amendment right, this could be described as intentional. A decision to cross-examine a defense witness is a product of strategic analysis by the prosecution. The inclusion, then, of the word intentional in Williams suggests that the defendant seeking to compel immunity must demonstrate something more than the prosecution s mere intention to cross-examine the witness.8 Our cases have insisted that the government s actions need to amount to something akin to prosecutorial misconduct. In Williams, we stated that resolution of this claim turns on whether the prosecution took affirmative steps to prevent Williams s witnesses from testifying. 384 F.3d at 601 (emphasis added). We elaborated: Undue prosecutorial interference in a defense witness s decision to testify arises when the prosecution intimidates or harasses the witness to discourage the witness from testifying, for example, by threatening the witness with prosecution for perjury or other offenses. . . . The prosecution s conduct must 8 As Straub conceded at oral argument, there may be occasions where the defense witness decides on his own to invoke the Fifth Amendment, even in the absence of an indication from the prosecution that it intends to cross-examine the witness. Such a scenario would not satisfy the requirement that the prosecution intentionally caused a defense witness to take the Fifth. UNITED STATES v. STRAUB 10701 amount to a substantial interference with the defense witness s free and unhampered determination to testify before the conduct violates the defendant s right to due process. Id. at 601-02; see also United States v. Lord, 711 F.2d 887, 891 (9th Cir. 1983) (holding that the record supported finding that prosecutorial misconduct caused the defense witness to invoke his Fifth Amendment privilege because the prosecutor told the witness that whether he would be prosecuted depended on his testimony ); United States v. Paris, 827 F.2d 395, 401 (9th Cir. 1987); United States v. Touw, 769 F.2d 571, 573 (9th Cir. 1985). As Straub has not alleged prosecutorial misconduct, he cannot prevail under this theory. B [3] The alternative way that a defendant may satisfy the second prong of the Williams test is by showing that that the prosecution granted immunity to a government witness in order to obtain that witness s testimony, but denied immunity to a defense witness whose testimony would have directly contradicted that of the government witness. 384 F.3d at 600. As discussed above, ambiguity arises because demonstrating that the conduct described, which we shall refer to as the selective denial of immunity, had the effect of distorting the fact-finding process does not necessarily demonstrate that the prosecution had the deliberate intention of distorting the factfinding process. Id. A careful review of our cases leads us to the conclusion that, notwithstanding this ambiguity in phrasing, a showing that the selective denial of immunity had the effect of distorting the fact-finding process is sufficient. 1 Few of our cases considering compelled use immunity have had to confront the purpose/effects question, because there, since the defendant s witnesses frequently offered evidence 10702 UNITED STATES v. STRAUB that was irrelevant, we had no need to consider the purpose or effect of the government s denial of immunity. See, e.g., United States v. Alvarez, 358 F.3d 1194, 1216 (9th Cir. 2004) (the testimony sought did not directly contradict statements by the government s witnesses); United States v. Brutzman, 731 F.2d 1449, 1452 (9th Cir. 1984) (the testimony was cumulative or not exculpatory ); Alessio, 528 F.2d at 1082 ( The testimony sought by appellant was cumulative . . . . ). In Lord, our earliest case to develop a test for compelled use immunity, we adopted the Third Circuit s rule that [t]he defendant must be prepared to show that the government s decisions were made with the deliberate intention of distorting the judicial fact[-]finding process. 711 F.2d at 890 (quoting United States v. Herman, 589 F.2d 1191, 1204 (3d Cir. 1978)). Lord also adopted the Third Circuit s subsequent developments of this rule, in Virgin Islands v. Smith, 615 F.2d 964 (3d Cir. 1980). As Lord describes Smith, the refined test was: whether the federal prosecutor, in refusing to consent to extending immunity to the defense witness, acted with a deliberate intention of distorting the factfinding process. . . . If the district court found that prosecutorial misconduct had prevented the defense witness from giving relevant testimony, then the court was directed to acquit the defendant unless the prosecutor granted use immunity to the defense witness. 711 F.2d at 891 (citations omitted). Thus, Lord required prosecutorial misconduct as an element of the test. However, Lord did not consider selective denial of immunity that was admittedly not prosecutorial misconduct but that had the alleged effect of distorting the fact-finding process. Lord only considered a claim under the other method of showing prosecutorial misconduct intentionally causing a defendant to take the Fifth. The dispositive fact was that the defense wit- UNITED STATES v. STRAUB 10703 ness testified that before trial the prosecutor told him that whether he would be prosecuted depended on his testimony. Id. Lord remanded for further clarification of the prosecutor s pre-trial comments to the defense witness. Id. We expanded the test for compelled immunity in Westerdahl. We began with the acknowledgment that if the prosecution intentionally caused the witness to invoke the Fifth Amendment, the law of our circuit clearly compelled a grant of use immunity: In the past, our decisions on prosecutorial misconduct have focused on whether the government or its agents took affirmative actions to prevent defense witnesses from testifying. 945 F.2d at 1086. We then suggested that a selective denial of immunity test was also viable: [M]isconduct is not confined solely to situations in which the government affirmatively induces a witness not to testify in favor of a defendant. Id. at 1087. Although Lord only dealt with a claim that the prosecution intentionally caused a defense witness to invoke the Fifth Amendment, we reasoned that [f]or the government to grant immunity to a witness in order to obtain his testimony, while denying immunity to a defense witness whose testimony would directly contradict that of the government witness, is the type of fact-finding distortion we intended to prevent in Lord. Id. We remanded for an evidentiary hearing, saying that where two eyewitnesses tell conflicting stories, and only the witness testifying for the government is granted immunity, the defendant would be denied any semblance of a fair trial and thus compelled immunity would be the proper remedy. Id. (internal citation and quotation marks omitted). Our subsequent opinions have continued to state that there are two methods by which the defendant seeking to compel immunity may demonstrate a distortion of the fact-finding process, but none have been presented with the purpose/effect issue we confront here. In United States v. Baker, for example, we summarized the test for compelled immunity, acknowledging both the Lord test and the Westerdahl test. 10 10704 UNITED STATES v. STRAUB F.3d at 1414. In that case, the defendant s claim for compelled immunity was so weak on so many elements that we had no occasion to consider the purpose/effect question. The individual that the defense sought to exonerate with testimony from a witness who was denied use immunity was not even charged with the murder about which the non-immunized witness could testify. See id. 1414-15. Moreover, the proffered testimony was inadmissible hearsay. In United States v. Young, we found strong evidence that the defendant s witness did in fact have relevant testimony that would have impeached the government s star witness. See 86 F.3d at 948. Holding that the defendant likely had a strong claim for compelled immunity under Westerdahl, we remanded for an evidentiary hearing as to whether the government intentionally distorted the fact-finding process. Id. at 949. We had no occasion to consider whether the mere effect of distorting the fact-finding process might be sufficient. See also United States v. Duran, 189 F.3d 1071, 1087 (9th Cir. 1999) (holding that defendant had not demonstrated either the Lord or Westerdahl bases for compelling immunity). 2 [4] We conclude that our statement of the test for compelled immunity in Williams, to the extent it appears ambiguous, merely reflects the fact that while we have previously suggested that an effects test is viable, we have never yet been presented with such a case. It is equally clear that our jurisprudence has anticipated endorsing an effects test where the other elements of the test have been satisfied. The separation of powers concerns implicated by compelled immunity suggest that we should be hesitant in expanding Williams; nevertheless, the logic of Williams and the cases it drew from supports the adoption of an effects test. Even where the government has not denied a defense witness immunity for the very purpose of distorting the fact-finding process, the government may have stacked the deck against the defendant in a way that has severely distorted the fact-finding process at UNITED STATES v. STRAUB 10705 trial. See Westerdahl, 945 F.2d at 1087 ( Previously, we noted in dicta that where two eyewitnesses tell conflicting stories, and only the witness testifying for the government is granted immunity, the defendant would be denied any semblance of a fair trial. (quoting Brutzman, 731 F.2d at 1452). In those cases where the government has liberally used its discretion to grant immunity to numerous witnesses, and the defendant s witness could offer relevant testimony that would directly contradict that of an immunized government witness, the trial may become so fundamentally unfair that the defendant s due process rights are implicated. Our approach is consistent with Supreme Court authority on due process at trial. When dealing with due process rights outside the courtroom, the Court has been hesitant to find constitutional violations by law enforcement absent some kind of malicious intent. See, e.g., County of Sacramento v. Lewis, 523 U.S. 833, 842, 855 (1998) (noting that we have always been reluctant to expand the concept of substantive due process and holding that a high-speed police car chase does not violate substantive due process unless the officer acted with intent to harm (quotation marks and citation omitted)); Daniels v. Williams, 474 U.S. 327, 328, 331-32 (1986) (holding that a prisoner s due process rights were not violated when he slipped on a pillow negligently left on the stairs by a correctional deputy, because [h]istorically, [the] guarantee of due process has been applied to deliberate decisions of government officials to deprive a person of life, liberty, or property ); Bingue v. Prunchak, 512 F.3d 1169, 1177 (9th Cir. 2008) (approving the Eighth Circuit s observation that the Court s reluctance to expand the concept of substantive due process suggests that the intent to harm standard applies to both emergency and non-emergency high-speed police car chases). However, when dealing with due process violations in the context of the fundamental fairness of the trial, the Supreme Court has been more concerned with protecting the integrity of trial and the defendant s right to mount a defense, irrespective of any government intent to interfere with these 10706 UNITED STATES v. STRAUB rights. The Due Process Clause addresses the defendant s right to a fair trial, not just whether the government intended to deny the defendant his rights. See, e.g., Brady v. Maryland, 373 U.S. 83, 87 (1963) (holding that the suppression by the prosecution of evidence favorable to an accused upon request violates due process . . . irrespective of the good faith or bad faith of the prosecution (emphasis added)); cf. Estelle v. Williams, 425 U.S. 501, 503-05 (1976) (holding that compelling the accused to stand trial in jail garb violates due process because it inherently impairs the jury s basic presumption of the defendant s innocence). The Court has also been more willing to find violations of other constitutional rights that affect the fairness of trial even where there is no intent to harm the accused s defense. See, e.g., Doggett v. United States, 505 U.S. 647, 656-57 (1992) (holding that the government may violate the defendant s Sixth Amendment speedy trial right even if it had no intent to harm the accused s defense). Moreover, the Court has repeatedly held that the Sixth Amendment may require courts to hold unconstitutional (as applied) certain evidentiary rules and privileges that prevent a defendant from mounting his defense regardless of the purpose of the rule. See Rock v. Arkansas, 483 U.S. 44, 51-53, 62 (1987); Davis v. Alaska, 415 U.S. 308, 320 (1974); Chambers v. Mississippi, 410 U.S. 284, 294-95 (1973); Washington v. Texas, 388 U.S. 14, 19 (1967). The right to compel use immunity because of selective denial of immunity is a right to due process inside the courtroom, where the Constitution focuses our attention on the fundamental fairness of the trial more than on the intentions whether good or bad of the prosecution. Finally our conclusion that an effects test is viable is supported by considering the remedy the defendant seeks. The court s decision to compel use immunity is not a sanction for prosecutorial misconduct; it is a vindication of the defendant s Fifth Amendment due process right to a trial in which the fact-finding process has not been distorted. See Lord, 711 F.2d at 892 ( [T]he key issue in the analysis of defense use UNITED STATES v. STRAUB 10707 immunity is whether the defendant was denied a fair trial. ). Because compelling use immunity is not a sanction for prosecutorial misconduct, it follows that the defendant need not prove that the prosecution acted intentionally to distort the fact-finding process where the other elements of the Williams test are met. Moreover, the remedy granted is neither an automatic acquittal for the defendant, nor an automatic grant of use immunity for the defense witness. In the Lord scenario, in which the prosecution intentionally caused the defense witness to invoke the Fifth, [i]f the district court finds such prosecutorial misconduct by a preponderance of the evidence, it should enter a judgment of acquittal for [the defendant] unless the prosecution invokes 18 U.S.C. §§ 6002-6003 by asking the district court to extend use immunity to [the defense witness] at a new trial. Id. (footnote omitted). Where use immunity is required because of a selective denial of use immunity that had the effect of distorting the fact-finding process, the prosecution has a third option as well: it may, at a new trial, attempt to proceed without the witness whose testimony would have been contradicted by the defense witness. These three options give the prosecution several choices and provide some mitigation for the intrusion on prosecutorial discretion that compelled use immunity causes. A survey of our opinions suggests that in the majority of cases where a defendant seeks to compel immunity for a witness, that witness s testimony will not be directly contradictory to that of the prosecution s witness, or there will have been no distortion of the fact-finding process, and the district court may deny immunity on those bases. See Alvarez, 358 F.3d at 1216 (the testimony sought did not directly contradict statements by the government s witnesses); Duran, 189 F.3d at 1087 (neither the Lord test nor the Westerdahl test was even applicable); Baker, 10 F.3d at 1414-15 (defendant not even charged with the crime about which defense witness offered testimony); Brutzman, 731 F.2d at 1452 (evidence was cumulative or not exculpatory ); Alessio, 528 F.2d at 1082 ( The testimony sought by appellant was cumulative . 10708 UNITED STATES v. STRAUB . . . ). As we discuss below, this case appears to be the rare case in which the testimony was in fact directly contradictory. Furthermore, the prosecution granted immunity and other incentives to eleven of Straub s co-conspirators, while denying immunity to the one witness who had testimony that, if believed, would make the government s key witness both a perjurer and possibly the actual perpetrator of the crime. There is an unmistakable air of unfairness to a trial conducted under these circumstances, one that calls into question the fundamental fairness of Straub s trial and the meaningful protection of his due process rights. [5] In conclusion, we find that any ambiguity in Williams arose because Williams and its predecessors recognized the possibility of the case we have before us, but had not confronted such a case. We now hold that for a defendant to compel use immunity the defendant must show that: (1) the defense witness s testimony was relevant; and (2) either (a) the prosecution intentionally caused the defense witness to invoke the Fifth Amendment right against self-incrimination with the purpose of distorting the fact-finding process; or (b) the prosecution granted immunity to a government witness in order to obtain that witness s testimony, but denied immunity to a defense witness whose testimony would have directly contradicted that of the government witness, with the effect of so distorting the fact-finding process that the defendant was denied his due process right to a fundamentally fair trial. 3 [6] Having clarified the applicable law, we turn to the district court s ruling that, even under an effects test, Straub was not entitled to compel use immunity for Baumann. We have already stated that Baumann s testimony would have been relevant and that Straub concedes the prosecution did not intentionally cause Baumann to invoke the Fifth Amendment. Thus, we consider only whether the prosecution granted immunity to a government witness in order to obtain the wit- UNITED STATES v. STRAUB 10709 ness s testimony but denied immunity to a defense witness whose testimony would have directly contradicted that of the government witness, with the effect of so distorting the factfinding process that the defendant was denied his due process right to a fundamentally fair trial. [7] It is undisputed that the prosecution granted immunity to Adams in order to obtain Adams testimony. We must next determine whether Baumann s testimony would have directly contradicted, Williams, 384 F.3d at 600, that of Adams. Straub s attorney was prepared to ask Adams: Mr. Adams, did you have a conversation with Mike Bauman[n] at a bar in the winter of 2003 in which you admitted to him that you had just shot a man? Adams intended to reply: No, I didn t. Baumann would have testified that, some time during the winter of 2003, Adams walked into a bar looking glum and sullen, and said, I just shot a guy. If Baumann is prepared to testify that Adams stated, I just shot a guy, and Adams is prepared to deny having made the statement, we have no trouble concluding that Baumann s testimony would have directly contradicted that of Adams, and accordingly that Straub has satisfied this element of the test for compelled immunity. The district court gave several reasons why Baumann s testimony did not directly contradict that of Adams. First, the district court stated that, even if Adams made the statement, it does not prove that he actually shot someone. He could have been joking or bragging. The court also reasoned that even if Adams had meant that he shot someone, there was no proof the statement was an admission of shooting Garret. The district court erred because Baumann s statements, if believed by the jury, could also support a finding that Adams lied on the stand and possibly that Adams was the shooter. The Williams test requires only that the testimony directly contradict the government witness s testimony on a relevant issue, not that the testimony have compelled the jury to exonerate the defendant. Adams told the prosecution that he did 10710 UNITED STATES v. STRAUB not tell Mike Baumann that he shot someone. Mike Baumann said Adams did tell him that he shot someone. This is, simply put, a direct contradiction. Adams may have been joking, or he may have shot someone other than Garrett, but these are matters for counsel to probe and the jury to decide. Baumann s testimony satisfies the test for directly contradictory because the testimony, if believed by the jury, could have supported a finding that the testimony directly contradicted that of the government s witness. The jury might have thought that it was implausible that Adams was joking or bragging if he appeared morose, and that it was unlikely that Adams shot two men during the same time period and more likely that he shot only one, Garrett. The fact that the proffered testimony need only support (as opposed to compel) a finding by the jury that it was directly contradictory is implied by the rest of the Williams test. The first prong of the test is that the proffered testimony be relevant. Williams, 384 F.3d at 600. We have cautioned that the relevance requirement is minimal. See id. The defendant need not show that the testimony sought was either clearly exculpatory or essential to the defense. Westerdahl, 945 F.2d at 1086. It would narrow the due process right significantly if the test allows a low showing of relevance to pass the first prong, only to require that the same evidence must compel the jury to find that it is directly contradictory for the second prong. A brick, of course, is not a wall, and it would contravene the holdings of Williams, Westerdahl, and Lord to require that a defendant proffer a wall in order to compel use immunity for his brick. [8] Finally, we must consider whether the prosecution s refusal to grant use immunity to Baumann had the effect of so distorting the fact-finding process that the defendant was denied his due process right to a fundamentally fair trial. The district court found that Adams s testimony was crucial to the prosecution s case. Adams was the only witness who testified Straub was armed that evening. Adams was also the only UNITED STATES v. STRAUB 10711 witness who put Straub at the location of the crime. The court found that, if a jury believed that Adams lied about this, and was shifting the blame to Straub for his own actions, it would seriously weaken the government s case on counts three and four, even under a Pinkerton instruction. The court also found that [Misty Dawn] Day s testimony, by itself, would not be enough to sustain a conviction on counts three and four. These findings of fact suggest that for the government to grant use immunity for Adams while denying use immunity to the one witness who could possibly portray Adams as a perjurer and the true perpetrator of the crime had the effect of distorting the fact-finding process. In addition to considering the fact that Adams was granted use immunity, we may consider how the prosecution exercised its discretion to provide use immunity and other incentives to secure the testimony of its other witnesses. See Westerdahl, 945 F.2d at 1087 (observing that the prosecution had granted use immunity to two witnesses and dropped charges against a third witness, while denying immunity to a defense witness who could have contradicted one of the three prosecution witnesses). The Amended Joint Stipulation of Facts presents a remarkable picture. Of thirteen witnesses described, all but one of the prosecution s twelve witnesses received either formal immunity, informal immunity, or other substantial incentives such as cash compensation or a reduction of sentence in exchange for their testimony. Three government witnesses (David Adams, Alsup, and Smith) received formal grants of use immunity. Seven government witnesses (David Adams, Smith, McCord, Center, Weiland, Clark, Shaw, Jacob Adams, Alsup) appear to have been granted informal immunity, i.e., they were not charged with any of the crimes to which they testified at trial, including serious drug and weapons charges. Six government witnesses (David Adams, McCord, McFarland, Smith, Mockley, Shaw) received incentives for their testimony, including one who was paid $8,300 for providing assistance, and another whose sentence was reduced from 24 months in prison to three years 10712 UNITED STATES v. STRAUB probation. Only one government witness, Misty Dawn Day, appears to have received no incentives or immunity. All but one of the eleven government witnesses who received either immunity or another incentive testified to having committed crimes involving the sale or manufacture of drugs; several admitted to sales totaling hundreds of thousands of dollars. The only defense witness listed, Mike Baumann, was denied use immunity. [9] The prosecution granted Adams use immunity, and Adams testimony was crucial to the conviction for the attempted robbery with a firearm. The prosecution denied use immunity to the one defense witness who could have discredited Adams testimony, even though the prosecution claimed it had no interest in prosecuting that witness. Meanwhile, the prosecution granted use and informal immunity, gave cash, and issued sentence reductions to eleven other admitted drug offenders, all for the purpose of securing evidence against Straub. We conclude that this exercise of discretion by the prosecution impermissibly distorted the fact-finding process, in violation of the Due Process Clause. A trial under these circumstances is not fundamentally fair. The district court erred when it denied Straub s request to compel use immunity.9 9 The prosecution also argues that the district court properly refused to compel use immunity because Straub never formally requested use immunity from the prosecution. As the transcript of the colloquy quoted earlier demonstrates, Straub did make a formal request to the district judge in the presence of the prosecution. The prosecution observed the district court deny the request. It was also equally clear from the prosecution s support of the district court s denial of the request that the prosecution was prepared to deny the request. It would be unnecessarily formalistic for us to hold that this sequence of events was insufficient to substitute for a direct, formal request to the prosecutor. Also, both sides agree that the Baumann situation arose unexpectedly in the middle of trial. While we do not doubt the prosecution s claim that securing use immunity is a formal and timeconsuming process, the district court would have had discretion to grant a continuance. UNITED STATES v. STRAUB 10713 III The failure to compel use immunity is reviewed for harmless error. See Young, 86 F.3d at 949; Lord, 711 F.2d at 891. Straub argues that we must remand for trial on Counts 1, 3, 4, 5, and 6. Straub has not articulated any specific reasons why, even if the jury were to hear Baumann s testimony and discredit Adams entirely, the jury would not have convicted on Counts 1, 5, and 6. As for Counts 3 and 4, the only two counts relating to the armed robbery of Garrett, we find that the failure to grant use immunity was not harmless. The testimony of David Adams was crucial to the prosecution s case on Counts 3 and 4. Adams was the only witness who testified that Straub was armed on the night of the robbery. The district court also found that the testimony of the prosecution s other witness, Misty Dawn Day, by itself, would not [have been] enough to sustain a conviction on counts three and four. Baumann s testimony, if believed, would have revealed both that Adams was dishonest and that he may have actually confessed to the shooting himself. After hearing Baumann s testimony, it is possible that some jurors would have doubts that Straub was the shooter and decide to acquit on Counts 3 and 4. The government argues that the failure to grant use immunity was harmless as to Counts 3 and 4 because the court gave a Pinkerton instruction. Thus, the government argues, the jury could have convicted even if it believed that Adams was the shooter, as both would be liable as co-conspirators under Pinkerton. However, we agree with the district court s finding: In theory, the jury could have convicted Straub on a Pinkerton conspiracy theory even if the jurors believed Adams was the shooter. . . . Nevertheless, the testimony of Adams was important to the prosecution s case. If a jury believed that Adams lied about this, and was shifting the blame to Straub for his own actions, it would seriously weaken the gov- 10714 UNITED STATES v. STRAUB ernment s case on Counts 3 and 4. . . . The defense could have argued that Adams, not Straub, orchestrated the Garrett robbery and shooting, without Straub s knowledge. Moreover, as Straub argues, the way the court charged the jury meant that the case was not effectively tried on the Pinkerton theory. The district court instructed the jury on Count 4 before it instructed on Count 3, and the court did not instruct that Count 4 was a conspiracy count. The court stated that in order to convict, the government must prove that the defendant knowingly and willfully did something or caused another person to do something that was a substantial step toward the actual commission of the crime. In order to convict, the jury would need to believe Adams s story that Straub was on the scene and took action to further the crime. After the Count 4 instruction, the court then instructed the jury on Count 3. The first element the government had to prove in order to get a conviction was that the defendant is guilty of the crime charged in Count 4. The court did give a Pinkerton instruction that the jury could convict on a conspiracy theory on Count 3, if it found that Garrett was shot by a person with whom the defendant conspired to commit the crime described in Count 4. The problem lies with not instructing the jury that Count 4 was a conspiracy count; absent this instruction, it is possible that the jury would not have convicted on Count 4 unless they found that Straub was the person who shot Garrett. In order to reach that conclusion, they would need to believe Adams testimony. If the jury did not convict on Count 4, they could not convict on Count 3 because the first element of Count 3 is a guilty verdict on Count 4. The court mitigated some of this prejudice by stating, after he had instructed on Count 4 and then Count 3, that the same definitions [of the elements of a conspiracy] apply here. It is unclear what here means and whether the district court was UNITED STATES v. STRAUB 10715 implying that Counts 3 and 4 were both conspiracy counts. Convictions cannot rest on ambiguous jury instructions. See United States v. Washington, 819 F.2d 221, 226 (9th Cir. 1987) (finding that ambiguous and equivocal jury instructions on an important issue constitute reversible error). The ambiguity in the jury instructions meant that the jury was probably not able to convict on a Pinkerton theory. The jury had to find that Straub was a motivating force behind the shooting, and it probably would not have found this to be the case if it believed Adams was lying on the stand. IV The Fifth Amendment does not create a general right for a defendant to demand use immunity for a co-defendant, and the courts must be extremely hesitant to intrude on the Executive s discretion to decide whom to prosecute. Alessio, 528 F.2d at 1081-82. Nevertheless, in exceptional cases, the factfinding process may be so distorted through the prosecution s decisions to grant immunity to its own witness while denying immunity to a witness with directly contradictory testimony that the defendant s due process right to a fair trial is violated. At Straub s trial, eleven prosecution witnesses, many of them serious drug offenders, were granted substantial incentives or immunity to testify against him. The testimony of one such witness was crucial to the prosecution s case for armed robbery. The defense proffered one witness who could directly contradict a statement made by the prosecution s key witness, and if believed, would allow the jury to find that the prosecution s witness was a perjurer and possibly the actual perpetrator of the shooting for which the defendant was charged. That one witness was denied immunity, despite the prosecution s insistence that he was not worth prosecuting. Even absent evidence of prosecutorial intent to do so, this course of events denied Straub a fair trial. [10] We reverse the district court s denial of Straub s request to compel use immunity. We affirm the convictions as 10716 UNITED STATES v. STRAUB to Counts 1, 5, and 6. On remand, the district court should enter a judgment of acquittal on Counts 3 and 4 unless the prosecution invokes 18 U.S.C. §§ 6002-6003 to grant use immunity to Baumann at a new trial, or tries the case without Adams s testimony. AFFIRMED REMANDED. in part, REVERSED in part, and

Some case metadata and case summaries were written with the help of AI, which can produce inaccuracies. You should read the full case before relying on it for legal research purposes.

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.