Before Campbell, Chief Judge, Bownes and Selya, Circuit Judges.
SELYA, Circuit Judge. Barry J. Griffin, defendant/appellant, was indicted by a federal grand jury in the United States District Court for the District of Maine. As more clearly appears infra, Griffin proceeded to trial and was found guilty by a jury on three of six counts. After sentence was imposed, he prosecuted this appeal. We affirm.
Due in part to the multiple loci of the venture and the plentiful cast of characters, the factual predicate of this case is somewhat sprawling. Rather than attempt to recreate the entire scenario, we will explicate the facts only to the extent which we deem necessary to place into perspective the issues which we must consider on appeal.
Beginning in 1983 (and continuing in 1984) three loads of marijuana were imported into Maine at the apparent behest of one Michael Gillis. These shipments were stored at three different sites, located in the bucolic environs of Yarmouth, Sweden, and Naples, Maine, respectively. On the prosecution's theory of the crimes, Gillis was the mastermind of the enterprise. He arranged to have the contraband distributed, at least in part, through a network managed by Louis Distasio, Sr. (Distasio) and manned by two of Distasio's sons (among others). In the government's view, the appellant was Gillis's adjutant and aide-de-camp, assisting him in the orchestration of the illicit operation.
The grand jury treated Griffin and Gillis like Gog and Magog, returning a single indictment which charged each of them with the same six counts of criminal activity. The odd-numbered counts alleged participation in three separate conspiracies to possess large quantities of marijuana (in excess of 1,000 pounds) with the intent to distribute the same. 21 U.S.C. § 846. The even-numbered counts tracked the conspiracy counts, charging the defendants in each instance with the substantive offense of illegal possession of the marijuana. 21 U.S.C. § 841(a)(1). To grind the mill more finely, Counts I and II related to the marijuana which had been brought to Yarmouth. Counts III and IV referred to the marijuana deposited in Sweden. And, Counts V and VI replicated the charges with regard to the Naples stash.
When the balloon went up, Gillis was nowhere to be found. (From aught that appears of record, he is still a fugitive from justice.) The appellant stood trial alone. He was convicted on the first and second (Yarmouth) counts, but of the lesser-included offense of conspiracy to distribute and illegally to possess less than 1,000 pounds of marijuana. He was also convicted, as charged, of the Naples conspiracy (Count V). The jury found him not guilty of the remaining three counts.
II. PLAIN ERROR
As from the vasty deep of the trial record, the appellant has surfaced a small flotilla of issues for our consideration. But, most of these must be allowed to recede into the bathyal zone without specific comment. The appeal has, in these numerous respects, been irreparably holed by procedural default.
Griffin attempts to raise several points which we believe fall in this category. These relate, inter alia, to the prosecutor's summation, the reception and treatment of certain evidence, and the district court's instructions to the jury. We see no need to all-encompassing in listing items which have, largely, gone by the boards. We do, however, set forth a representative sampling of the day's catch in the margin.1
Griffin's trial counsel interposed no objections to the prosecutor's closing argument; he proffered only two objections, both entirely unrelated to the errors assigned on appeal, to the jury instructions; and he preserved none of the evidentiary exceptions which we group under this rubric. The governing precepts are straightforward and unambiguous. In general, error may not be predicated upon rulings admitted or excluding evidence, Fed. R. Evid. 103, or on purported exceptions to the charge, Fed. R. Crim. P. 30, or in regard to an entire gamut of orders, Fed. R. Crim. P. 51, unless the party putatively aggrieved makes his complaint known to the trial court in due season, thus preserving his rights. These rules are no mere formalities, no judicial trap for the unwary litigant. They are essential to the balanced and orderly functioning of our adversarial system of justice. They serve several wild purposes. Among these is the idea that a party cannot sleep upon perceptible rights, but has an obligation to alert the district judge to error-in-the-making when and as the occasion arises. Placing this obligation on the litigant gives both the court and the party's opponent fair warning and a timely opportunity to acknowledge bevues and correct them so that cases can be decided squarely on merit (or the lack of it). If a party shirks this duty -- as Griffin has done -- he forfeits much of his opportunity thereafter to complain about ensuing mistakes. Such points can be reviewed on appeal only for the existence of what courts have come to term "plain error."
Plain errors are those "affecting substantial rights." Fed. R. Crim. P. 52(b). Put another way, they are "'particularly egregious errors' . . . that 'seriously affect the fairness, integrity or public reputation of judicial proceedings'." United States v. Young, 470 U.S. 1, 15, 84 L. Ed. 2d 1, 105 S. Ct. 1038 (1985) (citations omitted) (quoting United States v. Frady, 456 U.S. 152, 163, 71 L. Ed. 2d 816, 102 S. Ct. 1584 (1982), United States v. Atkinson, 297 U.S. 157, 160, 80 L. Ed. 555, 56 S. Ct. 391 (1936)). When we evaluate plain error "we are not . . . concerned with technical error or with prejudicial error . . . ." McMillen v. United States, 386 F.2d 29, 35 (1st Cir. 1967), cert. denied, 390 U.S. 1031, 88 S. Ct. 1424, 20 L. Ed. 2d 288 (1968). See also United States v. Rosa, 705 F.2d 1375, 1380-81 (1st Cir. 1983) (per curiam) (similar). Nor can we trifle with the tactical decisions of counsel, with errors reflecting only mere inadvertence, or with matters that are, in their poorest light, only arguably wrong. The plain error doctrine, in short, does not permit us to consider the ordinary backfires -- whether or not harmful to a litigant's cause -- which may mar a trial record. The doctrine focuses our attention only on blockbusters: those errors so shocking that they seriously affect the fundamental fairness and basic integrity of the proceedings conducted below. It follows that such errors are to be noticed only in "exceptional cases or under peculiar circumstances to prevent a clear miscarriage of justice." Nimrod v. Sylvester, 369 F.2d 870, 873 (1st Cir. 1966). See also Elwood v. Pina, 815 F.2d 173, slip op. at 6 (1st Cir. Mar. 31, 1987); Polansky v. United States, 332 F.2d 233, 235 (1st Cir. 1964).
Accordingly, we evaluate the appellant's procedurally defaulted claims against the entire record in order to determine whether this "high standard", United States v. Previte, 648 F.2d 73, 81 (1st Cir. 1981), has been met. We need not recount this odyssey step by step. Suffice it to say that our painstaking review of the record convinces us that the appellant received a trial in which fairness remained paramount and in which no plain error transpired.
The prosecutor's closing argument -- whether or not a textbook model -- was neither vulpine nor unduly inflammatory. There are no signs of prosecutorial misconduct so grossly improper as to warrant our intervention at this state; nothing was sufficiently amiss as "was likely to have affected the trial's outcome", United States v. Fuller, 768 F.2d 343, 347 (1st Cir. 1985), or which would require appellate intervention "to deter future misconduct." Id. The district court's charge to the jury, taken in its entirety, was comprehensive and generally accurate. Apart from (or perhaps including, see infra Part III) the two evidentiary matters discussed below, the exclusion and reception of proffered proof was unremarkable. The defendant's constitutional rights were fully and faithfully preserved. In fine, there is nothing in the record to suggest that, singly or in combination, errors of the sort which are now belatedly protested before us infected the general good health of the trial. Plain error is plainly absent.
III. EVIDENTIARY RULINGS
This pronouncement does not, however, complete our inquiry. In two areas, Griffin has assigned errors on appeal which we believe were sufficiently preserved at his trial to justify further examination. We proceed, therefore, to consider them.
A. The Relevancy/Prejudice Calculus
The appellant, having objected throughout on the basis that the testimony of Richard Hatem was irrelevant and unduly prejudicial, now urges that it was reversible error for the district court to permit Hatem to testify. As a subset of the same point, Griffin asserts that the court should have granted his timely motion to strike Hatem's testimony in its entirety, or allowed an alternative motion for a mistrial. To put these contentions into proper context, it should be observed that Hatem was a self-confessed coconspirator who had entered into a plea agreement with the government. Hatem -- living proof of the lack of honor among thieves -- proceeded to breach the agreement by concealing and misrepresenting material information. He was thereupon indicted for perjury and convicted. Later, he repented and again became the prosecutor's helpmate.
At Griffin's trial, Hatem testified about certain of the events surrounding the Sweden conspiracy. Most significantly, he corroborated Distasio's testimony that the two of them met a third man at a restaurant in Bridgton, Maine, near the Sweden stash house, in the course of, and for purposes related to, the operation. Among other tidbits, the witness testified that the person they met ("X") took Hatem's truck, left for a time, and returned with a load of marijuana. Although Hatem could not identify X, in other respects his version of these events squared with the account given by Distasio. The distributor's rendition of the "third man" theme differed from Hatem's, however, in that Distasio flatly identified Griffin as X. The defense placed the story of the Bridgton meeting in considerable doubt -- so much so, we note as an aside, that Griffin was acquitted on the two Sweden charges (Counts III and IV). It became correspondingly important for the government to attempt to bolster what slim evidence it had of the appellant's involvement in the Sweden scheme, so that the corroboration which Hatem furnished was not unnecessarily cumulative in this instance.
The district court found Hatem's testimony relevant in at least three respects: establishing the existence of the Sweden conspiracy, establishing Distasio's role in the distribution network (which, on the government's theory, comprised an integral part of the conspiracy), and in substantiating Distasio's account of the Bridgton affair. The finding of relevance is so clearly correct that it is scarcely worth discussing. Under Fed. R. Evid. 401, evidence is relevant if it has "any tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without that evidence." We have consistently recognized "that the district courts have broad discretion as to discerning the relevancy vel non of evidence . . . ." United States v. Tierney, 760 F.2d 382, 387 (1st Cir.), cert. denied, 474 U.S. 843, 106 S. Ct. 131, 88 L. Ed. 2d 108 (1985). The proof which Hatem supplied easily qualified under these generous guidelines. Indeed, Griffin's appellate counsel, to his credit, conceded as much during oral argument before us.
Once evidence is determined to be relevant, the only remaining issue is whether the district court correctly balanced the probative value of the evidence against its prejudicial impact under Fed. R. Evid. 403.2 As with Rule 401, the district courts have considerable discretion in calibrating the Rule 403 scales. United States v. Moreno Morales, 815 F.2d 725, slip op. at 35-36 (1st Cir. 1987); United States v. Kadouh, 768 F.2d 20, 21 (1st Cir. 1985); United States v. Tierney, 760 F.2d at 387-88; United States v. Kepreos, 759 F.2d 961, 964 (1st Cir.), cert. denied, 474 U.S. 901, 106 S. Ct. 227, 88 L. Ed. 2d 227 (1985). "The question is one of 'unfair' prejudice -- not of prejudice alone." Moreno Morales, slip op. at 36. Only in exceptional circumstances will we reverse the exercise of a district court's informed discretion vis-a-vis the relative weighting of probative value and unfairly prejudicial effect. Tierney, 760 F.2d at 388; Kepreos, 759 F.2d at 964. As we have stated,
the trial judge is Johnny-on-the-spot; he has savored the full taste of the fray, and his considerable discretion must be respected so long as he does not stray entirely beyond the pale.
Tierney, 760 F.2d at 388.
In this instance, no fault can be found with the district court's discretionary rulings. For the reasons which we have noted, Hatem's testimony was plainly probative. The aspects in which Griffin claims it to have been impermissibly prejudicial are at best conjectural. He argues, first, that because Hatem could not identify X, his testimony unfairly tied the appellant to the Sweden conspiracy. Yet, this ignores the "fit" between Hatem's testimony and that of Distasio. Moreover, the evidence obviously lacked powerful impact; after all, the jury acquitted the defendant on Counts III and IV.
The second contention is more sophisticated. Griffin speculates that, because Hatem had been prosecuted for perjury when, after agreeing to cooperate, he proved to be insufficiently forthcoming, the jurors would think that the government's accomplice witnesses would not dare to lie. But, evidence of the perjury prosecution cuts both ways. To be sure, as Griffin contends, it might have impressed the jury with the seriousness of cooperation agreements generally, with some indirect benefit to the credibility of the government's other plea-bargained witnesses. On the other hand, it is just as likely that Hatem's acknowledged perjury underscored for the jurors that the mere existence of a plea agreement was no guarantee that a cooperating witness was testifying truthfully.
It is significant, too, that the district judge took prudent and timely steps to alleviate any possible unfair prejudice. He instructed the jury, for example, that accomplice testimony should be "received with caution, . . . weighed with great care, and subjected to careful scrutiny." Indeed, he remarked that such testimony should be examined "with greater care and caution than the testimony of other witnesses." He admitted the plea agreements which each of these witnesses had signed as full exhibits, complying with the guidelines set out in that wise in United States v. Dailey, 759 F.2d 192, 200 & n.8 (1st Cir. 1985). except for Hatem's plea agreement, see supra, no objection to their admission was interposed. And in Hatem's case, the objection was part of the overall relevancy objection vis-a-vis his testimony, and did not focus specifically on the promise to tell the truth.3
In fine, Hatem's testimony was of admitted relevance and, if believed, had positive probative power on the issues in the case. Its sum and substance was prejudicial in that it tended to prove Griffin's guilt as to at least two of the counts -- but it was not unfairly so. Given that the district court wisely undertook prophylactic measures to dilute any overlay of unfair prejudice associated with this evidence, we do not find any abuse of discretion.
B. Rulings In Limine: Ripeness
The remaining issue derives from certain rulings made by the district court in connection with the examination of a prosecution witness, one Allan Mineart. Some background is helpful. Prior to appellant's trial, Mineart had been convicted of participation in the Naples conspiracy and had turned state's evidence. In the Homeric phrase, however, there is many a slip 'twixt the cup and the lip. Despite his pledge of cooperation, Mineart balked. Though he spilled some of the beans, he withheld relevant information from the investigators for over a year. The undisclosed data apparently included knowledge anent Griffin's role in the enterprise. Mineart at long last decided to make a clean breast of things. After he did so, the United States attributed his earlier reticence to a threat originating with Gillis.
Mineart was called as a government witness at the appellant's trial. At an appropriate stage during direct examination, the AUSA requested a bench conference. At sidebar, the prosecutor explained that he intended to forestall a predictable attack on Mineart's credibility by eliciting from him the reason for his delay in telling the whole truth. That explanation would, according to the AUSA, involve acquainting the jurors with the supposed threat (which, the government conceded, was not engineered by Griffin). Defense counsel objected vigorously to any revelation of Gillis's warning, reasoning that such evidence would introduce an element of violence into the case. Citing Fed. R. Evid. 403, see supra n.2, he urged that the prejudicial impact of such information would, as to Griffin, far overbalance any legitimate probative value which the evidence might possess.
Faced with these competing concerns, the district court sustained the objection, concluding that Mineart should be foreclosed from mentioning any such supposed threat in the government's examination in chief. The court went on to rule that, if defense counsel attempted to impeach Mineart's credibility because of his earlier failure to cooperate fully, then the AUSA would be permitted to question the witness on redirect as to the threat. No voir dire was held; the ostensible threat was never described with particularity; and Mineart never acknowledged that he had in fact been braced. Griffin was thus presented with a workable choice: he could attack Mineart based upon his original (incomplete) version of the events (and risk that the government would demonstrate the reason for the lapse), or he could refrain from savaging Mineart on this basis (thereby keeping the lid firmly shut on any possible showing that Gillis had attempted to intimidate the witness). Mineart's direct examination ended without further incident and, in cross-questioning Mineart, Griffin's trial counsel shied away from any reference to the witness's tardiness in telling the whole story.
We hold that the appellant's unease concerning the district judge's proposed handling of Gillis's saber rattling never ripened into an appealable matter. To be sure, the judge made a conditional ruling in limine -- but, because Griffin's trial counsel elected not to cross-examine Mineart about why he procrastinated in implicating the defendant, the actual issue which the appellant seeks to have use decide never arose. On the record as it stands, the district court merely sustained Griffin's Rule 403 objection and excluded evidence of the threat -- if any existed -- from the government's case in chief. Although the court telegraphed when its ruling was likely to be if defense counsel opened the door, the latter never knocked. And, we will not venture to pass upon issues such as this in a vacuum.
The parallel between this case and Luce v. United States, 469 U.S. 38, 83 L. Ed. 2d 443, 105 S. Ct. 460 (1984) is, we think, compelling. In Luce, the district court, on advance motion, refused to preclude the government from using a criminal defendant's prior conviction as impeachment material. The defendant did not take the stand and was convicted. He assigned as error the district court's unwillingness to bar the prosecution from using his criminal record in cross-examination. The Court decided that any possible harm flowing from the trial court's in limine ruling was "wholly speculative." Id. at 41. Speaking of the weighing of probative value against prejudicial effect which Fed. R. Evid. 609(a)(1) demands,4 Chief Justice Burger noted:
To perform this balancing, the court must know the precise nature of the defendant's testimony, which is unknowable when . . . the defendant does not testify.
Luce, 469 U.S. at 41 (footnote omitted).
So here, Fed. R. Evid. 403, see supra n.2, necessitates much the same genre of comparative analysis as Rule 609(a)(1). The Rule 403 calculus, like that of Rule 609(a)(1), "is subject to change when the case unfolds . . . ." Luce, 469 U.S. at 41. When, as here, the predicate examination has not been essayed, the trial court cannot be held to its tentative forecast. As in the Luce context, it is too great a handicap to bind a trial judge to a ruling on a subtle evidentiary question, requiring the most delicate balancing, outside a precise factual context.
Moreover, because the cross-examiner did not broach the subject of Mineart's delay, we have no way of knowing whether Gillis really uttered a threat, and if so, whether the government would actually have tried to show it. That situation, too, has its parallel in the Rule 609 context -- for when the defendant does not testify, the court never knows precisely what he would have said, or if the prosecutor would have followed through on his professed intention of using the conviction to impeach. Exactly that sort of uncertainty led the Court to characterize the point at issue in Luce as a "matter of conjecture." Id. at 42. The same is true here.
The speculative nature of this situation is highlighted dramatically by the proceedings below concerning another witness, Richard Robinson. According to the prosecution, Robinson, like Mineart, had been menaced by Gillis. As a result, Robinson, like Mineart, was less than candid at first. Later, he implicated Griffin. The defendant was given much the same option regarding cross-examination of Robinson as he had been offered vis-a-vis Mineart. In Robinson's case, the appellant's counsel chose to cross-question him about the omissions. Nevertheless, when the time came for redirect, the government decided to forgo any exploration of any threat and to rehabilitate Robinson in other ways. The same possibility existed, of course, as to Mineart; without a definite ruling in the vibrant context of live testimony, there is no reliable way of telling whether the government would have sought to rehabilitate Mineart by bringing out Gillis's supposed warning -- or whether, if the prosecution opted to do so, Mineart would have risen to the bait. Nor can we tell whether, if both these contingencies had come to pass, the district court would have allowed the evidence.5
It is no answer to say that everyone had placed their cards on the table, face up, when the district judge made his advance ruling. As the Court noted in Luce, 469 U.S. at 41 n.5, proffers can differ from actual testimony in material ways. Here, the threat itself was never spelled out. And, "even if nothing unexpected happens at trial, the district judge is free, in the exercise of sound judicial discretion, to alter a previous in limine ruling." Id. at 41-42.
Lastly, we note that permitting a defendant to raise on appeal a conditional Rule 403 forecast which results from an advance ruling in limine creates much the same danger of encouraging a defendant, as a trial tactic, to plant reversible error that is present in the precincts patrolled by Rule 609(a)(1). This concern was important to the Court in Luce, id. at 42, and it is important to us. The defendant is not entitled to the "windfall of automatic reversal. . . ." Id.
There is no principled way to distinguish Griffin's point from that confronted in Luce, supra. We hold that to raise and preserve for review the claim of improperly constructing the Rule 403 balance, a party must obtain the order admitting or excluding the controversial evidence in the actual setting of the trial. This does not mean, of course, that the evidence must always be revealed to the jury; as we have said in different context, "the cure should not itself become the carrier of the disease." Curran v. Department of Justice, 813 F.2d 473, 475 (1st Cir. 1987). Counsel should cooperate with the district court in exercising restraint and in employing the prophylaxis of the sidebar, where appropriate. See Fed. R. Evid. 103(c). In more complex situations, counsel may request that the jurors retire, or in exceptional cases, that the actual testimony be screened voir dire in the jury's absence. What matters is that the district judge have an opportunity to make his rulings in an actual context (or, at least, a verisimilitudinous enactment of an actual context), and that both the trial court and the appellate court "have . . . a complete record detailing the nature of [the] testimony, the scope of the cross-examination, and the possible impact of the [evidence] on the jury's verdict." Luce, 469 U.S. at 41. Only in that way can the trial judge meaningfully "perform the required balancing in the concrete context of actual question and answer . . . ." United States v. Mazza, 792 F.2d 1210, 1223 (1st Cir. 1986), cert. denied, 479 U.S. 1086, 55 U.S.L.W. 3569, 94 L. Ed. 2d 147, 107 S. Ct. 1290 (1987).6
We are not alone in concluding that the reasoning of Luce is not narrowly to be confined to situations under Rule 609. In United States v. Weichert, 783 F.2d 23, 25 (2d Cir. 1986) (per curiam), the court applied Luce to an in limine ruling involving Fed. R. Evid. 608(b),7 reasoning that "if impeaching questions are permitted under Rule 608(b), the trial court still is required to balance probative value against prejudice under Fed. R. Evid. 403, and this balancing is as dependent on the specific factual context as it is in Rule 609 cases." Id. See also United States v. Dimatteo, 759 F.2d 831, 832-33 (11th Cir. 1985) (per curiam) (similar). In United States v. Johnson, 767 F.2d 1259, 1270 (8th Cir. 1985), the Eighth Circuit applied Luce to in limine rulings under Fed. R. Evid. 404(b), a rule dealing with the admissibility of evidence of "other crimes, wrongs, or acts." "Although Luce was decided under Fed. R. Evid. 609(a)(1), its logic applies with equal force to motions under Rule 404." Johnson, 767 F.2d at 1270 (footnote omitted). Luce held sway, according to the Johnson court, because both Rule 609(a)(1) and Rule 404(b) "require the court to determine whether the danger of undue prejudice outweighs the probative value of the evidence." Id. at 1270 n.9. See Fed. R. Evid. 404(b) advisory committee note (explaining requisite balancing test by reference to factors appropriate under Rule 403). See also United States v. Studnicka, 777 F.2d 652, 660 (11th Cir. 1985) (objection to in limine ruling waived when concrete context did not materialize; unclear whether contested evidence to be offered substantively or for impeachment); cf. Coursen v. A.H. Robins Co., 764 F.2d 1329, 1342 (9th Cir. 1985) (in limine rulings concerning the use of sexual history evidence unreviewable). But cf. Adams v. Fuqua Indus., Inc., 806 F.2d 770, 773-74 (8th Cir. 1986) (in civil case, granting of motion in limine preserved exclusion of evidence for review where later objection would have been a mere formality).
Under the applicable standard, Griffin did not properly preserve the error of which he now complains for appellate scrutiny. Accordingly, we need not consider the fashion in which the district judge constructed the Rule 403 equation for purposes of his ruling in limine.8
As we noted previously, the appellant has endeavored to interest us in manifold other assignments of error. Most are fatally flawed by procedural default; none have discernible merit; all are, by this reference, rejected without further comment.
The record before us reveals that Griffin was fairly tried and fairly convicted. We need go no further.
1. Such issues include Griffin's claim that prosecutorial misconduct inhered in the Assistant United States Attorney's (AUSA's) comments, during closing argument, about the perjury conviction of the witness Richard Hatem (see text infra Part III-A). No contemporaneous objection was taken. Griffin also laments the court's failure to give an elaborate cautionary instruction concerning Hatem's testimony. None was requested. The appellant criticizes the court's charge regarding the law of conspiracy and the testimony of plea-bargained accomplice witnesses. He also castigates the AUSA's summation as it pertained to such witnesses. The alleged errors were not preserved below. And, Griffin complains to us, for the first time, that the trial court's brief explanation of the appellate process to the jury -- undeniably invited by defense counsel's final argument -- was erroneous.
2. Rule 403 of the Federal Rules of Evidence provides in pertinent part that: Although relevant, evidence may be excluded if its probative value is substantially outweighed by the danger of unfair prejudice. . . .
3. In any event, we have recently held that, as a general rule, it is within the discretion of the district judge to admit into evidence in a criminal case an unredacted plea agreement in which the witness covenants to testify truthfully. United States v. Martin, 815 F.2d 818, slip op. at 5-6 (1st Cir. 1987) (only by viewing the entire agreement can the jury properly assess "the probable motives or interests the witnesses could have in testifying truthfully or falsely") See also United States v. Miceli, 446 F.2d 256, 261 (1st Cir. 1971) (prosecutor's statement that witnesses had made an arrangement with the government to testify truthfully "fell short of plain error").
4. Fed. R. Evid. 609(a)(1) provides: For the purpose of attacking the credibility of a witness, evidence that he has been convicted of a crime shall be admitted if elicited from him or established by public record during cross-examination but only if the crime (1) was punishable by death or imprisonment in excess of one year under the law under which he was convicted, and the court determines that the probative value of admitting this evidence outweighs its prejudicial effect to the defendant . . . .
5. In order to raise and preserve for review the claim of the improper ruling under Fed. R. Evid. 403 in the circumstances presented here, several additional steps would have been needed. Griffin would have had to cross-examine Mineart about his initial (incomplete) cooperation with the government. On redirect, the AUSA would have had to attempt to introduce the evidence of Gillis's threat. Then, the defendant would have had to interpose a timely objection. The district court, faced with an actual situation, would have had to overrule the objection and determine that the evidence was admissible. Finally, Mineart would have had to testify as the prosecutor predicted he would. It would be foolhardy -- and wrong -- for us to assume that all of these necessary events would inevitably have occurred as described.
6. In Mazza, we followed Luce in the Rule 609 context. See also United States v. Wolfe, 766 F.2d 1525, 1526-27 (11th Cir. 1985) (per curiam) (similar), cert. denied, 475 U.S. 1066, 106 S. Ct. 1379, 89 L. Ed. 2d 604 (1986).
7. Fed. R. Evid. 608(b), in general terms, governs the admissibility of specific instances of conduct, other than criminal convictions, "for the purpose of attacking or supporting [a witness's] credibility. . . ."
8. It should not be implied that we have any fault to find with the trial judge's resolution of the problem. On the face of things, he appears to have considered the relevant criteria carefully and to have acted well within the bounds of his discretion.