Opinion
No. Crim. 97-70-P-H.
August 7, 1998.
Jonathan A. Toff, AUSA, Office of the U.S. Attorney, Portland, ME, for U.S.
Bruce M. Merrill, Portland, ME, for Defendant.
ORDER ON MOTION FOR NEW TRIAL
This new trial motion raises two issues: (1) Is a new trial required because of a Brady/Giglio violation? (2) Does newly discovered evidence of perjury by two government witnesses require a new trial? I find no "reasonable likelihood" or "reasonable probability" that the verdict would or could have been affected by either the alleged Brady/Giglio violation or the admitted perjury. Accordingly, I DENY the defendant's motion for a new trial on both grounds.
1. FACTS
The defendant Mark Huddleston was tried and convicted of attempting to possess cocaine with the intent to distribute it on November 6, 1997. The jury heard taped telephone conversations in which Huddleston spoke on November 5 and 6, 1997, with a cooperating witness whom I shall call by his nickname "Gonzo." (Huddleston's phone number was actually recorded on Gonzo's electronic watch.) In these telephone conversations Huddleston and Gonzo discussed setting up a drug transaction (without ever using the word "cocaine"), in which Huddleston referred to an ounce or 28 (grams), to the existence of customers ready for it, to a debt of "600 bucks" and to having money. A meeting place was set up at the Burger King on the Maine Turnpike ostensibly for Gonzo to deliver the cocaine to Huddleston. Because Gonzo had previously been arrested and was cooperating, Huddleston did not obtain the cocaine at Burger King, but was arrested there instead; hence, the charge and the conviction for attempt. The jury also heard several incriminating remarks Huddleston made after his arrest while two law enforcement officers drove him back to his home to complete a consensual search. Specifically, Huddleston said things such as "hey, man, you got to eat"; "it's only been about a year"; "those guys made it easy for me"; "you guys (referring to the law enforcement officers) are doing the right thing, you had to come and stop it some time"; "650 dollars, two or three ounces, it's not a lot"; and "there's three guys I deal to, that's all, man." Gonzo and another cooperating witness, whom I will call by his nickname "Chago," both testified that they had supplied Huddleston cocaine for distribution in the past and Gonzo testified about his phone conversations with Huddleston that were taped.
Huddleston testified himself and claimed, as part of his defense, that although he once had a cocaine habit (several of Huddleston's own witnesses testified to the presence and joint use of cocaine at Huddleston's residence), he had successfully kicked it. He also maintained that he had lent Gonzo and Chago various items of personal property such as a VCR and certain videogames. According to Huddleston, since Gonzo and Chago had disappeared for a time in the fall of 1997, he was willing to say anything to Gonzo on the phone in order to see him again in person so that he could get back his property. Huddleston also asserted that Gonzo and Chago were both originally from the Dominican Republic and that his previous dealings with them were for the purpose of setting up a business to export used automobile or motorcycle parts to the Dominican Republic. Several of Huddleston's witnesses testified that Huddleston had talked about an export business to the Dominican Republic. With respect to the taped telephone conversations, Huddleston testified that he did not understand several of Gonzo's questions, and was unable to explain some of his own answers.
On direct examination, the government did not elicit the national origin of Chago and Gonzo. On cross-examination by defense counsel, however, when the subject was explored they both denied coming from the Dominican Republic, although Gonzo admitted that his wife was Dominican and that his mother-in-law lived there. Chago asked and was permitted to consult his own lawyer during this line of questioning in front of the jury. The DEA agent involved in the investigation testified at trial that he thought they might be Dominicans. In one taped telephone conversation that he recorded in August, 1997 (played for the jury), he indicated that they were in fact Dominicans.
After the Huddleston trial, in the course of preparing presentence reports on Chago and Gonzo (who had earlier pleaded guilty), the Probation Office discovered that the Puerto Rican identities and social security numbers Chago and Gonzo had furnished at the time of their arrests were false. As a result, Huddleston filed his motion for new trial. Later, at their respective sentencings, both defendants/witnesses admitted that they were, in fact, from the Dominican Republic and that they had given false names and identities.
I conducted an evidentiary hearing on Huddleston's motion for new trial on July 17, 1998. The focus of the evidentiary hearing was on what the prosecuting and defense lawyers knew and when. In that respect, I make the following findings of fact. The defendant's lawyer believed all along that the two government witnesses were from the Dominican Republic. He did not know, until he cross-examined the two witnesses, that they might deny that. I find that there was no reason for the defendant's lawyer to have known that these two government witnesses claimed to be from Puerto Rico and that he did not fail to make the appropriate inquiries. (Defense counsel did make a timely Brady/Giglio motion.) All of the government's discovery materials furnished to the defendant's lawyer indicated a connection to the Dominican Republic. The defendant's lawyer was not privy to the prosecutions against the two witnesses in which they claimed at their arraignments, during pretrial investigations and thereafter that they were from Puerto Rico.
I also find that there was no intentional failure to disclose by the prosecutor. As is customary in this District, the government's file on the case against Huddleston was open to the defendant's lawyer, but it did not refer to Gonzo's and Chago's claims in their own cases to be from Puerto Rico. The Assistant United States Attorney testified that Gonzo's and Chago's nationality was not important to his case and that it was credible that they might be Puerto Rican even though they had claimed during their drug dealings to be from the Dominican Republic (because of the "panache" that the Dominican association would yield). At the time of Huddleston's trial, the prosecutor did not know their true nationalities. He believed that he had mentioned their claims of Puerto Rican background to the defendant's lawyer, but could not be sure. The DEA agent testified that he was sure that the Puerto Rican claim was mentioned at a meeting he attended with the defense counsel and the prosecutor, but I find his recollection to be mistaken. I observed defense counsel's surprise at trial when the witnesses denied being from the Dominican Republic. Moreover, there is no reason to believe that, had defense counsel been aware that the government's two most important witnesses might be claiming false identities, he would have failed to take steps to explore their identities such as requesting a continuance. At the very least, given the defendant's claim of setting up an export business, he would have avoided giving the witnesses the opportunity to deny they were from the Dominican Republic, since the DEA agent had already testified that he thought they were Dominicans. I find it far more plausible that the DEA agent and the prosecutor told the defendant's lawyer that they believed the two witnesses were "probably" from the Dominican Republic and "probably" were Dominicans, but unintentionally failed to explain that their reservations derived from the fact that Gonzo and Chago were claiming to be from Puerto Rico in their own cases.
At a conference before the evidentiary hearing on the motion for new trial, the parties stipulated that Gonzo and Chago had testified falsely in this case about both their identity and their national origin. The government also agreed that their perjury about their national origin was material, but was unwilling to stipulate that the perjury about their identity was material. I agree — and would find independently — that the perjury about their national origin was material given one element of the defense (export business to the Dominican Republic). I also find that the perjury about their identity (in this case their actual names) was material. See United States v. Ojeda, 23 F.3d 1473, 1477 (8th Cir. 1994) ("If a man would lie about his name, a jury may reasonably infer that he would lie about other matters, even on the witness stand.").
In sum, I make the following factual findings relevant to the motion for new trial.
1. Each of the government's two most important witnesses deliberately perjured himself on a fact material to the Huddleston trial.
2. The defendant's lawyer did not know and had no reason to know that these witnesses would deny that they were from the Dominican Republic.
3. The defendant's lawyer was surprised at the denial, and there was nothing that he could do, beyond the line of cross-examination he pursued at trial, to deal with the surprise.
4. The government's lawyer did not knowingly present perjured testimony.
5. The government's lawyer failed to disclose in advance his knowledge of the witnesses' claims in their own cases that they were from Puerto Rico, but the failure was unintentional.
2. ANALYSIS A. Brady and Giglio Issues
In Brady v. Maryland, 373 U.S. 83, 83 S.Ct. 1194, 10 L.Ed.2d 215 (1963), the United States Supreme Court held that a prosecutor must turn over to the defendant's lawyer exculpatory evidence possessed by the prosecution. Id. at 87, 83 S.Ct. 1194. In Giglio v. United States, 405 U.S. 150, 92 S.Ct. 763, 31 L.Ed.2d 104 (1972), the Supreme Court made clear that this constitutional obligation includes impeachment evidence. Id. at 154, 92 S.Ct. 763.
Here, the prosecutor disclosed what would classically be called "exculpatory" evidence — specifically, the two witnesses' Dominican connection. (This was arguably exculpatory because Huddleston claimed a legitimate reason, a potential export business to the Dominican Republic, for his association with Gonzo and Chago.) What was not disclosed — their claims of Puerto Rican backgrounds in their own cases — was impeachment evidence because it was inconsistent with their claims to be Dominicans during their drug dealings. See Giglio v. United States, 405 U.S. at 154, 92 S.Ct. 763 ("When the `reliability of a given witness may well be determinative of guilt or innocence,' nondisclosure of evidence affecting credibility falls within" the Brady rule.) (citation omitted). Although the Puerto Rican claims turned out to be false, at the time of trial no one knew they were false and there was, therefore, a government obligation to inform the defendant of these claims as impeachment of these two witnesses' credibility. Theoretically, the defendant's lawyer could have impeached Gonzo and Chago at trial with these claims given the other evidence about their Dominican backgrounds. This would have been a problematic line for defense counsel to pursue, however, because it would have directly contradicted the defendant's claim that Gonzo and Chago were in fact Dominicans engaged with the defendant in a legitimate business venture. Moreover, the defendant's lawyer did elicit at trial Gonzo's and Chago's denial of the Dominican connection. He could have used these denials to impeach their testimony inasmuch as he also had the DEA agent's testimony and other testimony about their Dominican connections. Although I find that the defendant's lawyer was surprised at Gonzo's and Chago's denials of Dominican origin, he did not request a continuance to explore the matter further. More important, I conclude that using the actual claims of Puerto Rican backgrounds rather than the mere denials of Dominican origin in cross-examination would not create any "reasonable probability" of a different verdict, the standard for a new trial in the event of a Brady/Giglio violation. See Kyles v. Whitley, 514 U.S. 419, 433-34, 115 S.Ct. 1555, 131 L.Ed.2d 490 (1995) (stating that the "touchstone of materiality is a `reasonable probability' of a different result, and the adjective is important"; the inquiry is whether the verdict was "worthy of confidence") (construing United States v. Bagley, 473 U.S. 667, 682, 105 S.Ct. 3375, 87 L.Ed.2d 481 (1985)).
The defendant makes an additional argument in claiming a Brady/Giglio violation. Specifically, he argues that with proper disclosure of the witnesses' claims to be Puerto Rican his lawyer might have been able to obtain more information to support an outright attack on the identity of the government witnesses. At this point, the defendant's claim is 20/20 hindsight inasmuch as the court's probation officer learned later of the witnesses' false identities by pursuing information provided by the Social Security Administration and from his colleagues in Puerto Rico when he prepared Presentence Reports for Gonzo and Chago. There is no assurance that defense counsel would have or could have obtained the same information before trial if he had known of the witnesses' claims of Puerto Rican backgrounds. More important under the Brady/Giglio analysis, this later-developed information was not information in the government's possession at the time of trial.
The defendant also argues that Gonzo had a criminal history in Massachusetts and that the government failed to inform him of such information. There are two problems with this argument. One, as the defendant conceded at oral argument, the criminal history would have been inadmissible to impeach Gonzo under the federal rules. Two, the government prosecutor did not have the information; it was developed after trial when a fingerprint inquiry was made of Massachusetts. The fingerprint inquiry did elicit additional aliases that might have helped marginally in cross-examining Gonzo. But, since the government simply did not have this information, there was no obligation to turn it over before trial. Also, such evidence would not create any reasonable probability of a different verdict if it had been disclosed to the defendant. See Kyles v. Whitley, 514 U.S. at 433-34, 115 S.Ct. 1555.
B. Newly Discovered Evidence of Perjury
I turn now to the consequences of the post-trial discovery that Gonzo and Chago committed perjury at Huddleston's trial. The First Circuit standard for new trials based upon newly discovered evidence is clear. A defendant must establish four things:
1. The evidence in question was "unknown or unavailable at the time of trial";
2. The defendant and his lawyer exercised due diligence;
3. The evidence is material (not merely cumulative or impeaching); and
4. The evidence is "likely to result in an acquittal upon retrial."United States v. Tibolt, 72 F.3d 965, 971 (1st Cir. 1995); see also United States v. Natanel, 938 F.2d 302, 313 (1st Cir. 1991). (The court has also said that the third and fourth factors are "less stringent" if the government had control of the evidence and withheld it. United States v. Tibolt, 72 F.3d at 971.) If all Huddleston had in this case was a motion on the ground of newly discovered evidence, I would deny his motion for a new trial. Although Huddleston clearly satisfies the first two factors, and assuming the evidence satisfies the third factor (that it is material, not merely cumulative or impeaching), I would find that he cannot meet the final factor.
But Huddleston has more: his newly discovered evidence is false testimony by two government witnesses on a material issue — perjury. The Seventh Circuit confronted such a situation in Larrison v. United States, 24 F.2d 82 (7th Cir. 1928), and announced that where a government witness committed deliberate perjury, the defendant need only establish that the perjury "might" have affected the result. Id. at 87. Larrison was decided in 1928. Over the succeeding years, the Larrison test has been roundly criticized both inside and outside the Seventh Circuit. See, e.g., United States v. Mazzanti, 925 F.2d 1026, 1029 nn. 2-3 (7th Cir. 1991); United States v. Stofsky, 527 F.2d 237, 245-46 (2d Cir. 1975). Among other things, both the Seventh Circuit and other circuits have recognized that the verbal formulation is exceedingly vague and that many of the results in cases supposedly applying Larrison could just as easily be explained with the general formulation for newly discovered evidence — "likely" acquittal. In fact, in instances of perjured testimony where the government did not knowingly use false testimony, the Second Circuit has rejected Larrison and has adopted a "`probably would have acquitted'" standard. United States v. Torres, 128 F.3d 38, 49 (2d Cir. 1997) (citation omitted). The Ninth Circuit agrees with the Second Circuit. See United States v. Krasny, 607 F.2d 840, 844 (9th Cir. 1979). Although in cases where the prosecutor knowingly uses false testimony the Second Circuit has sometimes applied the Larrison "might" standard, see United States v. Sanchez, 969 F.2d 1409, 1414 (2d Cir. 1992); Sanders v. Sullivan, 863 F.2d 218, 225 (2d Cir. 1988); see also United States v. Sinclair, 109 F.3d 1527, 1532 (10th Cir. 1997), it has also adopted from United States v. Agurs, 427 U.S. 97, 96 S.Ct. 2392, 49 L.Ed.2d 342 (1976), the standard that the verdict must be set aside only if there is "any reasonable likelihood that the false testimony could have affected the judgment of the jury," id. at 103-04, 96 S.Ct. 2392. See United States v. Torres, 128 F.3d 38, 49 (2d Cir. 1997) (citation omitted); see also United States v. Tierney, 947 F.2d 854, 860-61 (8th Cir. 1991) (citation omitted).
In using this other standard, the Second Circuit does not appear to have explicitly rejected the Larrison standard; in at least two instances the court has cited cases which applied the Larrison "might" standard when describing the standard as "reasonable likelihood." See, e.g., United States v. Wong, 78 F.3d 73, 81 (2d Cir. 1996); United States v. Wallach, 935 F.2d 445, 456 (2d Cir. 1991).
In the midst of this debate, the First Circuit has assiduously avoided committing itself. In 1971, it referred to the Larrison test but found it "inoperative" because the testimony in question was not false. United States v. Strauss, 443 F.2d 986, 989-90 (1st Cir. 1971). In 1977, it twice referred to Larrison and each time found no need to resolve the issue. See United States v. Street, 570 F.2d 1, 4 (1st Cir. 1977); In re United States, 565 F.2d 173, 177 n. 3 (1st Cir. 1977). In 1979, it referred to Larrison, but found it "not applicable." Pelegrina v. United States, 601 F.2d 18, 21 (1st Cir. 1979). In 1980, it referred to Larrison as "arguably applicable," noting that some of its previous dictum had said that the Circuit had not decided whether Larrison applied, but that other cases "have suggested" its application. United States v. Wright, 625 F.2d 1017, 1020 (1st Cir. 1980). Once again, however, the First Circuit found no need to resolve the issue. Id. at 1020-21. In 1989, the First Circuit referred to the Larrison test as "arguably" the correct approach. United States v. Carbone, 880 F.2d 1500, 1502 (1st Cir. 1989). In 1991, the court stated that "we have observed" what Larrison requires, but once again found it unnecessary to resolve the issue. United States v. Natanel, 938 F.2d 302, 313 (1st Cir. 1991). Most recently, in 1993, the First Circuit avoided the issue by only "assum[ing]" that Larrison applied and finding that the standard had not been met. United States v. Sepulveda, 15 F.3d 1216, 1221 (1st Cir. 1993).
Given the diffident appellate references to Larrison in this Circuit, a trial judge is well advised to approach the inquiry under all available standards. If Larrison applies literally, I hold that Huddleston is entitled to a new trial, for I cannot foreclose the conclusion that, had the jury been informed that two important government witnesses had in fact perjured themselves about their identities and their connection with the Dominican Republic, the jury "might" have acquitted, believing Huddleston's tortured explanation of both the recorded conversations and his later incriminating remarks. That is the problem with Larrison's "might" standard. It is simply too broad, because any time there is perjured testimony on a material item one can hardly deny that a verdict "might" have been affected. It would be better simply to say that perjury on a material item leads to a new trial in all cases.
The caselaw is in general agreement that the reviewing court must plumb the effect on the jury of both the incorrect evidence (here, the denial of Dominican Republic origin) and the fact of perjury. See, e.g., United States v. Mazzanti, 925 F.2d 1026, 1030 n. 6 (7th Cir. 1991); United States v. Stofsky, 527 F.2d 237, 246 (2d Cir. 1975).
If the Supreme Court's Kyles v. Whitley standard for situations when the prosecution knowingly presents perjured testimony applies (and I have found that the prosecutor here did not know or have a duty to know), Huddleston loses. There is no "reasonable likelihood" that the verdict "could" have been different with the newly discovered evidence of perjury. Kyles v. Whitley, 514 U.S. at 433 n. 7, 115 S.Ct. 1555 (quoting United States v. Agurs, 427 U.S. 97, 103, 96 S.Ct. 2392, 49 L.Ed.2d 342 (1976)). Huddleston's incriminating comments to law enforcement officers after his arrest, the taped recorded telephone conversations prior to arrest, his trip to meet Gonzo following those conversations and his admitted past dealings with cocaine are overwhelming. Huddleston also loses if I apply the less lenient traditional standard for newly discovered evidence (set forth in United States v. Tibolt, 72 F.3d at 971), or for a Brady/Giglio violation (set forth in Kyles v. Whitley, 514 U.S. at 433-34, 115 S.Ct. 1555). Specifically, for the reasons I have just given, I cannot conclude that the disclosure of the perjury to the jury would "likely" result in or create a "reasonable probability" of acquittal. (I observe that the government did not possess the evidence of perjury such that the "less stringent" analysis of Tibolt applies, id. at 971, but even if it did, I would reach the same result.)
Finally, I conclude that, when forced to decide, the Court of Appeals for the First Circuit will not apply the Larrison "might" standard in cases of newly discovered evidence of perjury, in light of the United States Supreme Court's recent approval in Kyles v. Whitley, 514 U.S. 419, 115 S.Ct. 1555, 131 L.Ed.2d 490 (1995), of the "reasonable likelihood" test for even knowing presentation of perjured testimony. Because Huddleston loses under all other standards, I Deny his motion for new trial.