UNITED STATES of America, Appellee, v. Jesus M. ACOSTA, Defendant, Appellant.
Jesus Acosta was indicted on two counts of possession of a firearm by a convicted felon. 18 U.S.C. § 922(g). A jury convicted Acosta on one count and acquitted on the other, and Acosta was then given a mandatory minimum sentence of 15 years' imprisonment under the Armed Career Criminal Act, 18 U.S.C. § 924(e)(1). He now appeals, raising as his main issue a claim of entrapment. The evidence at trial, taken in the light most favorable to the jury's verdict against Acosta, United States v. Tuesta-Toro, 29 F.3d 771, 773 (1st Cir.1994), cert. denied, 513 U.S. 1132, 115 S.Ct. 947, 130 L.Ed.2d 890 (1995), revealed the following.
Acosta is a 42-year-old man, married, with a prior record of drug offenses but no prior weapons convictions. Sometime in mid-1993-probably in early July-Acosta met Neal San Souci at a pawn shop in Pawtucket, Rhode Island, and the two men engaged in small talk regarding gold jewelry. Acosta and San Souci had apparently met once many years before. Unknown to Acosta, San Souci either was then or soon thereafter became a government informant.
A few days after the pawn shop meeting, Acosta and his brother-in-law stopped by San Souci's apartment to inspect some gold jewelry that San Souci had offered to sell Acosta. Instead of providing the jewelry, San Souci asked Acosta and his brother-in-law whether they could furnish San Souci with cocaine. When they declined, San Souci asked whether the two men could provide a gun. According to San Souci, Acosta said, “he'd check into it ․ he didn't know of anybody or anything at that moment.”
By his own testimony at trial, San Souci was a former drug addict and present alcoholic. Around the first week in July 1993 he began to work as an informant for Special Agent Stephen Woods of the Bureau of Alcohol, Tobacco and Firearms (“ATF”). Prior to working for ATF, San Souci had been jailed for failing to pay child support; for his assistance on this case and other matters, ATF paid San Souci approximately $4,000. San Souci conceded that he thought that he would be paid only if he succeeded in persuading Acosta to sell him the firearm.
San Souci testified that following the visit to his apartment he called Acosta on a number of occasions, usually leaving messages with Acosta's wife or Acosta's answering machine. Acosta's wife also testified that the telephone calls were very frequent, sometimes more than once a day, and that Acosta himself appeared uninterested in the messages and often made dismissive gestures. The purpose of San Souci's efforts to reach Acosta was to obtain firearms for San Souci to purchase.
On July 21, 1993, San Souci spoke to Acosta by telephone, again asking to purchase a firearm. Acosta told San Souci that he would “check into it” and advise San Souci. On July 23 Acosta told San Souci that he had a .25-caliber automatic pistol for sale. Later that day, after some bargaining, San Souci gave Acosta $125 in exchange for the weapon, which was fully loaded and had an obliterated serial number. San Souci asked for more guns, and Acosta said that he would “get back” to San Souci. The conversation was taped but the tape was inaudible.
At agent Woods' direction, San Souci did not call Acosta for a couple of weeks because telephone records were being secured. On August 6, 1993, San Souci called Acosta to ask for weapons and in the conversation-which was taped and played at trial-Acosta said that he was going to get them but needed more time. A similar conversation occurred on September 15, 1993, and on the following day, Acosta telephoned San Souci to tell him that he had a .32-caliber revolver for sale. The same day San Souci purchased the gun from Acosta for $130, again after bargaining about price.
During this sale, San Souci asked Acosta if he could get more guns. Acosta replied, “I'm going to get .38 specials” and “Maybe I can come up with an Uzi for $200.” This conversation was recorded, and the recording played at trial. There is no indication that any other transactions were attempted or accomplished. In January 1994, Acosta was arrested for the two weapons sales and charged in two counts under the felon-in-possession statute.
At his trial, Acosta testified, admitting the transactions and his status as a prior felon. He relied primarily on the entrapment defense. The judge gave an entrapment charge, whose correctness is not challenged. The jury acquitted Acosta on the count relating to the July 23 transaction and convicted him for the September 16 transaction.
On this appeal, Acosta's main argument is that a verdict of acquittal should have been directed on count II on grounds of entrapment or, alternatively, that a new trial should have been offered. It appears that the motion for a new trial was made out of time, see Fed.R.Crim.P. 33; but in any case the district court's denial of such a motion is reviewed only for abuse of discretion. We think, therefore, that the central issue of this appeal is whether the evidence was sufficient to permit a reasonable jury to reject the entrapment defense.
The legal tests for entrapment are well established. What is required is (1) that the government induce the offense and (2) that the defendant not be predisposed to commit it. See Jacobson v. United States, 503 U.S. 540, 112 S.Ct. 1535, 118 L.Ed.2d 174 (1992). The bare terms-inducement and predisposition-do little to disclose the encrusting precedent. For our purpose, the most useful discussion is the decision of then Chief Judge (now Justice) Breyer in United States v. Gendron, 18 F.3d 955 (1st Cir.), cert. denied, 513 U.S. 1051, 115 S.Ct. 654, 130 L.Ed.2d 558 (1994). That decision, which is post-Jacobson, not only illuminates the entrapment concept but remains the governing law in this circuit.
Gendron makes clear that despite some general strictures against the government's “manufacturing” of crimes, inducement requires something more than that a government agent or informant suggested the crime and provided the occasion for it. Rather, inducement “consists of [providing] an ‘opportunity’ plus something else-typically, excessive pressure by the government ․ or the government's taking advantage” of the defendant in an improper way. 18 F.3d at 961 (emphasis added). There is no better means of getting a sense of what the courts have regarded as “improper” inducement than the list of cases and parentheticals set forth in the Gendron opinion. Id. at 961-62.
Although the entrapment doctrine is primarily concerned with curbing such improper pressure by the government, a competing policy has led to the second requirement, namely, that the defendant also not be predisposed to commit the crime. The notion is that a defendant predisposed to commit the crime should not get off merely because the government gave the defendant too forceful a shove along a path that the defendant would readily have taken anyway. Gendron suggests that one might ask whether defendant would have been likely to commit the same crime without the undue pressure actually exerted. 18 F.3d at 962.
Entrapment is called a defense, but it is settled that once the defendant has made a threshold showing, the burden shifts to the government to prove beyond a reasonable doubt either that there was no undue government pressure or trickery or that the defendant was predisposed. See United States v. Rodriguez, 858 F.2d 809, 815 (1st Cir.1988). In this case, the facts were largely although not entirely undisputed. Thus, the problem for the jury was primarily that of applying a vague general standard-actually two such standards: inducement and predisposition-to a unique pattern of facts.
Because the facts were largely undisputed, one might think that on review this court necessarily decides as an issue of law whether the facts do or do not make out entrapment. Yet, even where there are no credibility issues or tensions in the evidence-and some do exist here-entrapment is treated as a issue of fact for a jury. That does not mean complete freedom for the jury, see Jacobson; it does mean that where a rational jury could decide either way, its verdict will not be disturbed. United States v. Gifford, 17 F.3d 462, 467 (1st Cir.1994).
Starting with inducement, the problem (as it is so often) is one of degree. On the one hand, the government does not disclaim responsibility for San Souci's conduct even if it occurred before he was hired, and here that conduct went some distance beyond “simply offer[ing] [defendant] the opportunity” to commit the crime. Jacobson, 503 U.S. at 550, 112 S.Ct. at 1541. True, Acosta did not close the door in the first discussion (Acosta, according to San Souci, said “he'd check into it”). But it took a campaign of persistent calls by San Souci before Acosta responded, apparently several weeks later.
On the other hand, there is no evidence that San Souci threatened Acosta or even rebuffed an explicit request by Acosta to be let alone. Nor does this case involve improper appeals to sympathy, cf. Sherman v. United States, 356 U.S. 369, 78 S.Ct. 819, 2 L.Ed.2d 848 (1958), promises of extravagant reward, or the kind of relentless and extreme trickery engaged in by postal and customs agents in Jacobson. See 503 U.S. at 543-47, 112 S.Ct. at 1537-40. In other words, the facts fall somewhere in a middle ground between what is plainly proper and what is plainly improper.
If the district court had refused to submit the entrapment issue to a jury for lack of threshold evidence of inducement, we would have reversed. But we do not think that the evidence was so overwhelming as to establish improper conduct by the government as “a matter of law.” San Souci's efforts, although far from pristine, were dubious rather than flagrant, or at least a rational factfinder could so determine. To assume that we are dealing with a sharp boundary rather than a spectrum is an illusion.
By tradition issues associated with guilt or innocence (duress, insanity, entrapment) are submitted to the jury. United States v. Gaudin, 515 U.S. 506, ----, 115 S.Ct. 2310, 2313-14, 132 L.Ed.2d 444 (1995). Other issues, perhaps similar in kind but related to collateral matters, are determined by the court (e.g., the reasonableness of a search and, in many jurisdictions, the voluntariness of a confession). In the former category of merits-related issues, the jury in close cases effectively decides not only what happened but also whether what happened deserves the legal label described in the jury instruction.
If the jury gets to make these middle-ground decisions on inducement, deference is even easier to understand on the issue of predisposition. In large part, predisposition turns on making a judgment as to how a defendant of a given character, background, and behavior would have acted in somewhat different circumstances. On questions of this kind, the joint common sense of a jury is hard to best. At least as a composite body, the jury probably knows quite as much as the judge, or more, about how human beings behave outside of court.
Except where a jury acquits in a criminal case, judges remain as a check on juries in the extreme case-one where the judge thinks that a rational jury could reach only one result. Acosta invokes this exception, arguing that he had no known prior record of dealing in guns, did not hurry to accept San Souci's initial overtures, and showed little enthusiasm throughout the venture. We agree that in this case the trial judge could not have withdrawn the predisposition issue from the jury by refusing to instruct on entrapment, and nor would we have been surprised if the jury had chosen to acquit.
But there is another side to the coin. Acosta had a criminal record in drug dealing, properly made known to the jury, and drug dealing is often associated with access to weapons. He did not flatly rebuff the initial overture by San Souci, and it is uncertain whether Acosta's delay in supplying a weapon reflected inertia, suspicion, or a genuine reluctance to commit a criminal act. The second gun was provided with less prompting than the first, and the prospect of a third sale, possibly of an Uzi, was suggested at the end.
Thus, a rational jury could conclude that Acosta was predisposed to commit the offense. His prior record with drugs would not be enough by itself; but, with one qualification, the government was entitled to rely as evidence of predisposition on Acosta's own behavior after he was approached by San Souci. This included an initial willingness at least to consider supplying arms, the later provision of two weapons with the suggestion that more would be available, and a certain measure of professional finesse in making the transfers.
It is quite true (and this is the qualification just mentioned) that under Jacobson predisposition does not count if it is itself the product of improper government conduct. That could reasonably have been said in Jacobson. There, the government, through its own mailings to the defendant, purporting to come from others, encouraged the defendant to believe that procuring child pornography was a blow against censorship and in favor of the First Amendment. If there were predisposition, said the Court, the government instilled it. 503 U.S. at 553, 112 S.Ct. at 1542-43.
In our case, the facts are more equivocal. San Souci was less persistent and less deceptive than the agents in Jacobson. Further, San Souci did nothing to encourage the defendant to alter his views of right and wrong; he just offered money in exchange for weapons. On the present facts, the jury could have concluded that Acosta was predisposed or that he was not, but it could not easily have concluded that San Souci created in Acosta a predisposition toward crime. Certainly it was not obliged to reach this conclusion.
As authority for a judgment of acquittal, Acosta points us to United States v. Beal, 961 F.2d 1512 (10th Cir.1992). There, a government informant, to secure relief from criminal charges against him, pestered Beal with telephone calls until drugs were supplied; two sales were made by Beal within a twenty-four hour period; and presented with an entrapment defense, the jury acquitted Beal as to the first sale and convicted on the second. The district judge in Beal then ordered an acquittal on the count of conviction. By a two-to-one vote, the Tenth Circuit affirmed, saying: “Because the two counts were founded upon one continuous course of conduct, it follows that the original inducement which ‘beguiled’ Mr. Beal carried over to the second charge.” 961 F.2d at 1517.
The government says that unlike Beal the two sales in this case took place two months apart and were not part of “the same course of conduct.” Why this should matter, either in Beal or in this case, is unclear. A jury in a criminal case is not obliged to be consistent in its verdicts; on virtually the same evidence the jury may acquit on one count and convict on the other. United States v. Powell, 469 U.S. 57, 65, 105 S.Ct. 471, 476-77, 83 L.Ed.2d 461 (1984). The only question for the judges-district and appellate-is whether the evidence on the count of conviction compelled an acquittal.
It is not clear to us whether Beal was rightly decided or whether there are nuances in the evidence there that made it a stronger case for entrapment than our own. What we do know-for reasons explained above-is that as to the second count in our case, the jury was entitled either to find that Acosta was entrapped or to reject the defense. Since the evidence permitted a conviction on count II, it is irrelevant whether the conviction is logically consistent with Acosta's acquittal on count I.
As it happens, the two verdicts in our case are not inherently inconsistent. Merely as an example, the jury could rationally have concluded that Acosta was not predisposed as to either sale; that the degree of badgering in connection with the first sale did constitute inducement; that the lack of further badgering distinguished the second sale; and that therefore entrapment was established as to the first sale but not as to the second. Whatever reasoning the jury adopted, the evidence as to count II permitted a conviction.
Acosta's remaining claims of error on the appeal are much weaker. First, we reject the suggestion that the government did not prove the commerce element in the case. The statute prohibits a prior felon from possessing a firearm “in or affecting commerce.” 18 U.S.C. § 922(g). Here the government offered evidence from an ATF expert that both of the firearms sold by Acosta were made in specific states other than Rhode Island. The jury was therefore entitled to conclude that both weapons had traveled in interstate commerce before Acosta possessed them.
Acosta makes no constitutional claim but argues that the statute should be read leniently to require that a defendant possess the weapon, if not actually during its interstate travel, at least close in time to such travel. But this court has already held that the terms “affecting commerce” were used as “jurisdictional words of art” reflecting an intent to exercise the commerce clause power broadly. United States v. Gillies, 851 F.2d 492, 493-95 (1st Cir.), cert. denied, 488 U.S. 857, 109 S.Ct. 147, 102 L.Ed.2d 119 (1988). Given Congress' inferred intent, it is hard to doubt Gillies was correct and, in any case, it is binding.
Second, Acosta objects to the trial court's handling of a note from the jury. On May 27, 1994, a few hours after the jury began deliberations, the court advised counsel for both sides that the jury had submitted a note reading: “If we play the tape/conversations that were not introduced during the trial, are heard, are they allowed?” At the request of both sides, the court summoned the jury back to the courtroom to inquire further, it not being evident that there were any taped conversations except those introduced at trial.
Before counsel could reassemble, the jury sent word that it had reached a verdict. Prior to taking the verdict, the court asked the jury in open court whether the jury no longer wished to have its written question answered. The foreman stated that that was correct. Then the verdicts were received and the jury was discharged, without objection by either side, either before or immediately after the taking of the verdicts. Four days later, Acosta's counsel filed a motion to recall the jury for voir dire regarding the note.
The district judge's denial of the motion to recall is now challenged on appeal, but the denial was plainly correct. While the jury's inquiry is puzzling, there is no proof that the jury considered anything outside the evidence. More important, if defense counsel wanted a further inquiry, the time to ask here-as with any curable defect or doubt-plainly was before the verdict. United States v. Mosquera, 63 F.3d 1142, 1156 & n. 7 (1st Cir.1995) Trial judges must manage juries in the face of all kinds of problems and perplexities. When trial counsel acquiesce in a proposed solution, it is rare indeed that an appeals court will engage in second guessing.
Third, the district court sentenced Acosta to an enhanced sentence under the Armed Career Criminal Act, because he had three previous convictions “for a violent felony or a serious drug offense,” specifically three drug trafficking convictions that met the statutory definition of “a serious drug offense.” 18 U.S.C. § 924(e)(1). Acosta urged that one of the prior convictions was invalid because no inquiry was made in that case as to the factual basis for his guilty plea. On procedural grounds the state court refused to set the conviction aside.
Acosta admits that under Custis v. United States, 511 U.S. 485, 114 S.Ct. 1732, 128 L.Ed.2d 517 (1994), a defendant has no right collaterally to attack his prior convictions during a sentencing under the Armed Career Criminal Act. But, he says, Custis does not prevent the trial court from considering such a collateral attack as a matter of discretion. We think that the reasons given by the Supreme Court in Custis apply with equal force, whether the reexamination of the state conviction is sought by the defendant or the trial judge. See 511 U.S. at ----, 114 S.Ct. at 1738-39.
One further word is in order. Whether or not the government unduly encouraged Acosta to commit the offense is a close call, but it is the kind of close call that a jury is equipped to make. What may be even more troublesome in cases of this kind is the possibility of undue encouragement to the informant, as a result of compelling government inducements (here, money; in Beal, dismissal of criminal charges) to overstep the bounds in the field, or in the courtroom, or both.
In his dual role as both instigator and witness, the informant has a special capacity-as well as strong incentive-to tilt both the event itself and his testimony about it. If the government is going to use its informants in a role just short of provocateur, it would be well advised to consider devising restrictions that will at least lessen the likelihood for abuse. Otherwise, the lesson of history is that the courts themselves are likely to take precautions and their adjustments are usually more rigid and far-reaching.
BOUDIN, Circuit Judge.