883 F.2d 1540 | 11th Cir. | 1989
Lead Opinion
Louis Miller was convicted of conspiracy to knowingly distribute cocaine in violation of 21 U.S.C. § 846 and the knowing distribution of cocaine in violation of 21 U.S.C. § 841. Miller was sentenced to fifteen years on each count to run concurrently. He appeals his sentence on several grounds. Since we find that the district court committed reversible error by admitting extrinsic evidence under Federal Rule of Evidence 404(b), we do not address the other issues.
FACTS
On January 7, 1987, Special Agent Regina Bledsoe of the Drug Enforcement Administration (DEA) and a confidential informant contacted Labron Lyons at his mother’s house in order to buy some cocaine. Lyons accommodated them by calling the beeper number of his supplier. Lyons told Agent Bledsoe and the informant that “Louis” would bring the cocaine to him. Agent Bledsoe then waited in her car in the driveway of the house. Several minutes later, a blue Pontiac Grand Am pulled in behind her. After obtaining the money from Agent Bledsoe, Lyons entered the passenger seat of the Grand Am. The car pulled away, drove around the block, and returned and dropped Lyons off. When Lyons returned he had an ounce of cocaine.
Miller was charged with the January and September transactions in separate indictments. The district court denied defense counsel’s motion to join the two offenses in a single trial. At the trial for the January offense, the government introduced evidence concerning the September transaction. Over defense counsel’s objection, the district court held it admissible under Fed. R.Evid. 404(b). Both Lyons and Special Agent Bledsoe testified as to the events in September. In addition, the tape of the phone call setting up the delivery was played to the jury. Finally, testimony that Miller was carrying cocaine and eight hundred dollars was also admitted. Miller was convicted and this appeal ensued. This appeal concerns only Miller’s conviction for the January offense; Miller’s appeal from his conviction for the September offense is pending in this court.
DISCUSSION
Miller contends that the trial court committed reversible error when it admitted the evidence concerning the September incident under Fed.R.Evid. 404(b). We agree. The admission of evidence under rule 404(b) is entrusted to the sound discretion of the trial judge. On the unique facts presented in this case, the court abused its discretion in admitting the evidence.
Rule 404(b) states:
Other crimes, wrongs, or acts. Evidence of other crimes, wrongs, or acts is not admissible to prove the character of a person in order to show that he acted in conformity therewith. It may, however, be admissible for other purposes, such as proof of motive, plan, knowledge, identity, or absence of mistake or accident.
The rule codifies the “venerable principle that evidence of extrinsic offenses should not be admitted solely to demonstrate the defendant’s bad character.” United States v. Beechum, 582 F.2d 898 (5th Cir.1978) (en banc), cert. denied, 440 U.S. 920, 99 S.Ct. 1244, 59 L.Ed.2d 472 (1979). Beechum created a two part test to determine the admissibility of evidence under rule 404(b). First, the evidence must be relevant to an issue other than the defendant’s character. Relevance is determined by the issue to which the extrinsic evidence is addressed.
The government argues that the evidence was relevant to modus operandi. It contends that the two transactions were sufficiently similar to demonstrate the defendant’s modus operandi and to corroborate Lyons’ testimony. This argument misconstrues rule 404(b). Not only does it come dangerously close to admitting that the only purpose for introducing the evidence was to prove that the defendant acted in conformity with a bad character, it also implies that the only purpose for organizing the September transaction was to obtain a conviction for the January offense.
The government’s argument misunderstands the modus operandi exception to rule 404(b)’s ban on character evidence. The modus operandi exception does not allow the government to admit evidence of the defendant’s acts to show that he usually did things a certain way. To allow that would be tantamount to allowing proof that the defendant acted in conformity with his bad character. Instead, evidence demonstrating a modus operandi is admissible for a much more limited purpose. In some cases, a modus operandi might constitute circumstantial proof of identity. See C. Wright & K. Graham, 22 Federal Practice & Procedure: Evidence, § 5246, at 513-14 (1978). In such cases, however, “the physical similarity [of the acts] must be such that it marks the offenses as the handiwork of the accused. In other words, the evidence must demonstrate a modus oper-andi.” Beechum, 582 F.2d at 912 n. 15 (citing United States v. Goodwin, 492 F.2d 1141, 1154 (5th Cir.1974)). Indeed, “a much greater degree of similarity between the charged crime and the uncharged crime is required when the other crime is introduced to prove identity than when it is introduced to prove a state of mind.” Id. (quoting United States v. Myers, 550 F.2d 1036, 1045 (5th Cir.1977)).
In this case, because defense counsel alluded to an alibi defense in his opening statement, identity clearly was at issue. However, for evidence of the September transaction to be admissible to prove identity, the similarity between the two acts must be so great as to show a distinctive modus operandi. The extrinsic evidence fails this part of the Beechum test. It is true that the two incidents shared similarities, Lyons “beeped” his supplier who drove to the house, picked up Lyons and then drove around the block to transact the deal. There were, however, several important differences between the two transactions. In the January transaction, the seller drove a blue Grand Am. In the September transaction, Miller was driven by a third man in a blue Monte Carlo. Thus the only similarities between the two offenses were the use of a beeper and delivery in a car. But since the government orchestrated both acts, these similarities are attributable to the government’s intervention. Moreover, there is no showing that these
We also reject the argument that the evidence was admissible to corroborate Lyons’ testimony. First, corroboration is not one of the purposes of the rule. Neither the rule nor caselaw supports the government’s argument. Second, the testimony at most showed that Lyons bought drugs in this manner; it shed no light on whether he was being truthful that Miller sold him the drugs in January. There is no doubt that the government’s case concerning the January transaction was weak. Although Agent Bledsoe and Special Agent John Harvey, who was maintaining surveillance of the car, identified the driver as Miller, neither had a good view of the suspect. Agent Bledsoe only saw him briefly through her rearview mirror and Agent Harvey stayed fifty to sixty feet from the car when he was following it and only got one glimpse of the driver’s face. In addition, the car was not under continuous surveillance. Therefore the government’s case relied solely on Lyons. As the government concedes, Lyons was far from a good witness. He had been convicted of cocaine trafficking previously and had a history of drug use. Moreover, he agreed to cooperate in return for the dismissal of the substantive count and the government’s agreement not to prosecute him on other unrelated drug charges. Thus while the government had a legitimate desire to corroborate Lyons’ testimony, this evidence does not do so.
The government also argues on appeal that the evidence is admissible to prove Miller’s intent. Although the government did not advance this justification at the trial level, we note that the district court’s jury instructions contained the pattern jury charge for when extrinsic evidence is introduced to prove intent.
The September transaction is clearly relevant to intent because it required the same state of mind as the charged offenses. Beechum, 582 F.2d at 911. However, that is only the first step in the Beechum analysis. We must also consider whether the probative value of the evidence is not substantially outweighed by the prejudice. This balancing requires a “commonsense assessment of all the circumstances surrounding the extrinsic offense.” Id. at 914. In determining the probative value of the evidence, the court must consider the strength of the government’s case, whether the issue of intent was contested, the similarity between the charged and extrinsic offenses, and the remoteness of time separating the two acts. Id. at 915; see Dorsey, 819 F.2d at 1060; United States v. Parr, 716 F.2d 796, 805 (11th Cir.1983); United States v. Mitchell, 666 F.2d 1385, 1390 (11th Cir.), cert. denied, 457 U.S. 1124, 102 S.Ct. 2943, 73 L.Ed.2d 1340 (1982). These factors are weighed against the danger “that the jury may convict the defendant not for the offense charged but for the extrinsic offense.” Beechum, 582 F.2d at 914. The weighing of these factors is entrusted to the discretion of the trial judge. Id. Applying these factors to the case at hand, we find that the district court abused its discretion in finding that the probative value of the evidence was not outweighed by the prejudice.
Intent was at issue in the case because of Miller’s plea of not guilty. United States v. Elliott, 849 F.2d 554, 558-59 (11th Cir.1988); United States v. Nahoom, 791 F.2d 841, 845 (11th Cir.1986). When the issue dominates the case, the probity of the evidence is apparent. Dorsey, 819 F.2d at 1060. On the other hand, when the government has a strong case on intent, the extrinsic offense adds little and therefore is readily excludable. Beechum, 582 F.2d at 914; see Parr, 716 F.2d at 805 (when government presented a “substantial but not overwhelming case on intent,” additional evidence was not unnecessary).
There are significant factors decreasing the probative value of the extrinsic evidence and increasing the potential for undue prejudice. This extrinsic evidence concerned an act which the government planned and executed approximately nine months after the charged offense. The fact that the act occurred approximately nine months after the charged offense does not render it automatically inadmissible. The remoteness of time, however, may decrease the probative value of extrinsic evidence. Compare United States v. Jimenez, 613 F.2d 1373, 1376 (5th Cir.1980) (one year lapse between charged offense — possession of heroin — and subsequent extrinsic act — possession of cocaine — deprived extrinsic act of any relevance which was not outweighed by prejudice) with United States v. Terebecki, 692 F.2d 1345, 1349 (11th Cir.1982) (fifteen months between extrinsic act and charged act which were similar fraudulent business transactions did not deprive extrinsic act of relevance). The effect of the remoteness of time on the admissibility of the extrinsic evidence depends in part on the similarity between the offenses. As noted above, the only similarity between the transactions was that they involved a similar means of delivering co
Moreover, the possibility of prejudice is greatly increased because of the government’s role in creating and executing the September transaction. Although intent was at issue, the key issue at trial was the identity of the supplier in the January transaction. Before the jury could consider the evidence of the September transaction as relevant to the intent of the defendant, it had to find that Miller had committed the January offense. As pointed out above, the government had a weak case on the identity of the January transaction but an almost airtight case against Miller for the September transaction. It is possible that the jury considered the evidence of the September transaction to establish that Miller had committed the January offense. This is especially true since more evidence was admitted relating to the September transaction than the charged offense. In addition, the AUSA asked the jury to consider the evidence as relevant to identifying Miller as the supplier in the January offense.
We recognize that the judge instructed the jury that it should only consider the extrinsic evidence as relevant to intent. We conclude, however, on the unique facts of this case, that these instructions were not sufficient to cure the tremendous potential for prejudice. First, we note that the trial court gave these instructions during jury instructions; no limiting instructions were given when the evidence was introduced. Moreover, the AUSA argued that the evidence had been admitted to prove identity and to corroborate Lyons’ identification of Miller and structured her case accordingly.
We recognize that the weighing of the probative value and prejudicial impact of such evidence is entrusted to the sound discretion of the trial judge. We do not take the prospect of overturning such a discretionary decision lightly. However, in this case, we think the district court abused its discretion by failing to consider the circumstances behind the extrinsic act. The fact that the extrinsic event was orchestrated by the government nine months after the charged offense decreases the probative value of the evidence and dramatically increases the possibility of prejudice.
REVERSED.
. Noticeably absent from the dissent is any citation to Beechum. This absence explains why the dissent contends that the law of this circuit is that extrinsic evidence is admissible if it corresponds to the charged offense in "important particulars.” Infra at 1549. The dissent does not acknowledge that Beechum held that “similarity and hence relevancy, is determined by the inquiry or issue to which the extrinsic offense is addressed.” 582 F.2d at 911. Indeed, the case cited for the dissent’s proposition, United States v. Alston, 460 F.2d 48, 55 (5th Cir.), cert. denied, 409 U.S. 871, 93 S.Ct. 200, 34 L.Ed.2d 122 (1972), was decided before Beec-hum (indeed it was even decided before the Federal Rules of Evidence were adopted. Pub.L. No. 93-595, 88 Stat. 1926 (1974)). Once this flaw in the dissent’s analysis is recognized, it is clear that the September transaction was not admissible to prove identity.
. The flaw in the dissent’s argument is its belief that similarity rather than uniqueness is the predicate for admissibility of extrinsic evidence to prove “identity, method of operations, intent, or anything other than bad character[.]" Infra at 1549. The caselaw establishes that uniqueness rather than similarity is the touchstone of admissibility of extrinsic acts to prove identity because of the danger that the jury will consider the extrinsic acts solely as proof of propensity. As Weinstein explains
There are many instances when details of the crime show an individuality that, if repeated, are highly probative of the conclusion that they were committed by the same person.... Proof of the commission of the same type of crime is not sufficient on this theory unless the particular modus operandi is analyzed. A defendant cannot be identified as the perpetrator of the charged act simply because he has at other times committed the same commonplace variety of criminal acts except by reference to the forbidden inference of propensity. The question for the court is whether the characteristics relied upon are sufficiently idiosyncratic to permit an inference of pattern for purposes of proof.
J. Weinstein & M. Berger, 2 Weinstein's Evidence, ¶ 404[16], at 128-29. There was absolutely no showing in this case that the two transactions were "idiosyncratic.”
. This conclusion does not conflict with the holding of a recent panel of this court in United States v. Stubbins, 877 F.2d 42, 44 (11th Cir.1989). In Stubbins, the government introduced testimony that the defendant had attempted to sell "crack” cocaine at a certain address on an occasion prior to the incident for which the defendant was being tried. Id. at 43. The district court held that the evidence was admissible to prove identity because the fact that the offenses were perpetrated at the same place made it distinctive. The panel, however, held the evidence admissible because it helped to explain why the police officer knew that the first address written in its report was incorrect. Id. at 44. The error in the police report had been repeatedly highlighted during cross-examination. Id. at 43. In Stubbins, therefore the evidence was not used as circumstantial evidence to prove identity; it was used as direct evidence to explain the police officer's identification of the dwelling. See generally, 2 Weinstein's Evidence, ¶ 404[15], at 114-15 (discussing use of extrinsic evidence as direct evidence of identity); 22 Federal Practice & Procedure: Evidence, § 5246, at 1515 (same).
. The jury instructions stated:
Now, during the course of the trial, you know from the evidence that came in, you heard that at a time other than the time charged in the indictment in this case the defendant committed acts similar to the acts charged here. You may consider such evidence not to prove that the defendant did the facts charged in this case but only to prove the defendant’s state of mind or plan, that is that the defendant acted as charged in this case with the necessary intent and not through accident or mistake.
Therefore, if you find that the government has proved beyond a reasonable doubt that the defendant did, in fact, commit the acts charged in the indictment and that the defendant also committed similar acts at other times, then, you may consider those other acts in deciding whether the defendant committed the acts charged here willfully and not through accident or mistake.
Record, Vol. 2, at 186.
. The dissent contends that the evidence of the January transaction was "substantial, perhaps overwhelming,” infra at 1548. This argument, however, cuts against the government's position in this case. The strength of the government’s case is an important factor in determining the probative value of the extrinsic evidence. Extrinsic evidence is only relevant if it has a tendency to prove a fact that is at issue; thus, if the government’s case is strong, there is no need for the evidence. Beechum, 582 F.2d at 915 (“Probity in this context is not an absolute; its value must be determined with regard to the extent to which the defendant's unlawful intent is established by other evidence, stipulation, or inference. It is the incremental probity of the evidence that is to be balanced against its potential for undue prejudice.’’) (emphasis added). If the evidence was as overwhelming as the dissent contends, there was no need for the government to use the evidence of the September transaction at trial. The fact that the government relied so heavily on the evidence of the September transaction to convict the defendant of the January sale, therefore, reflects the weakness of the government’s case.
. During rebuttal argument, AUSA Stewart stated
Although he is not charged with an offense occurring in September, it is appropriate for you to look at that. Consider it in terms of the modus operandi, how it shows intent, how it shows doing the same thing again, how it helps to identify that the person who got the cocaine on the 7th was, indeed, Mr. Miller, who arrived in September and the deal was done virtually in the identical manner. Mr. Lyons calls up, orders it up, they get in the car, they drive around, he gets the cocaine in the car, he comes back, he delivers it to the agent, and then in September there's an arrest, and Mr. Miller is arrested with the cocaine.
Record, Vol. 2, at 174.
. During closing argument, AUSA Stewart told the jury that it should believe Lyons' testimony because of the September transaction. She said:
... I submit to you that the best way in which Mr. Lyons' testimony that the cocaine came from Mr. Miller is corroborated is from the other actions of Mr. Miller himself. From January to September, you heard the tape, both Agent Bledsoe who had listened to Mr. Miller talk before, and Mr. Lyons, who had obviously talked to Mr. Miller before on several occasions, said that was his voice on the tape. What were they talking about? Ordering up some cocaine.
... And where did they meet? At Labron Lyons' mamma’s house over on Avon Avenue, the place the first deal was. How does the deal take place? The same way the first deal took place. Car drives up. This time he’s not alone; he’s got somebody with him. Labron hops in the car. They drive around the area. They come back. Labron has got two ounces of cocaine that he did not have when he got into that car....
The arrest signal was given. Mr. Lyons and Agent Bledsoe are arrested, too, to make it look like Mr. Miller is arrested. And what does Agent Dolan find on his person? Another package of cocaine stuffed down in the sock.
Ladies and Gentlemen, I submit to you that corroborates and reinforces exactly what La-bron Lyons said happened the first day. Why? It happened the second day, too, and he was arrested at the scene the second day.
Record, Vol. 2, at 158-59.
. Implicit in our holding is a determination that the error in admitting the extrinsic evidence was not harmless. The dissent faults us for impermissibly evaluating the credibility of the witnesses. The dissent misunderstands the role of this court in determining the harmlessness of a trial court’s error. An error is only harmless if "the error had no substantial influence on the verdict and there was sufficient evidence to support the verdict apart from the error. United States v. Martinez, 700 F.2d 1358, 1367 (11th Cir.1983); United States v. Heller, 625 F.2d 594, 599 (5th Cir.1980) (error harmless "if reviewing court is sure, after viewing the entire record, that the error did not influence the jury or had a very slight effect on its verdict.”) Because the September transaction was the linchpin of the government’s case on identity, we cannot say that the extrinsic act did not affect the jury’s verdict. The dissent's contrary conclusion relies on pure speculation. See infra, at 1549 ("the jury may well have found believable [Lyons’] testimony of the events of January 7, 1987.”)
. We do not mean to suggest that the government may not use a cooperating individual to build a case against a suspected drug dealer or that the government is required to charge separate acts in a single indictment. The problem in this case is that the government used the second act, for which it had an airtight case, to prove that the defendant committed a similar offense several months earlier. While in some cases, it may be appropriate for the government to introduce extrinsic evidence of subsequent acts, in this case, the probative value of the evidence was substantially outweighed by the potential for undue prejudice.
Dissenting Opinion
dissenting:
Although the majority pays lip service to the clear abuse of discretion standard for reviewing evidentiary rulings, it reduces to a mere form of words our highly deferential touchstone. By reevaluating the credibility of witnesses and reweighing the evidence, the opinion usurps the role of the jury and diminishes to the point of extinction the substantial evidence against Miller for the January 7, 1987, drug deal, while the evidence of the September 3, 1987, sale is somehow transformed into a monster.
The majority opinion holds that the extrinsic evidence was not admissible to show identity, method of operations or to corroborate Lyons’ testimony. It concedes that the extrinsic offense was probative of intent but concludes that the district court clearly abused its discretion in admitting it, primarily because (1) the government’s case concerning the January transaction was “weak;” (2) the government “orchestrated” the September drug buy; and (3) the two crimes were not sufficiently similar. I would hold that the district court did not clearly abuse its discretion, because the evidence of the September, 1987, violation was admissible to prove identity, mo-dus operandi and intent.
Central to the panel’s reversal of this conviction is its insistence that the government’s case was weak. According to the opinion, the government’s case “relied solely on Lyons,” and the Rule 404(b) material constituted the prosecution’s “entire case.” In my view, this conclusion distorts the record. The testimony of three witnesses puts Miller at the scene of the January, 1987, crime beyond a reasonable doubt. Agent Harvey recounted that during the course of a surveillance on January 7, 1987, he received a radio transmission advising him to look for a blue Pontiac Grand Am. He spotted the car at 2:00 P.M.: “I observed the car in the neighborhood just to the north of [Lyons’ mother’s] house and observed Labron Lyons in the front passenger seat of the Grand Am, and I observed Louis Miller driving the Grand Am. I just saw them for a brief minute.” R. 2-30. A couple of minutes later, after the car had returned to the house, Agent Harvey resumed surveillance of the Grand Am. R. 2-30-31. In the agent’s words,
At that point, when I was following it again, it was only Louis Miller in the car. Myself and several other cars followed Miller for several miles. We also had an airplane on surveillance which was mainly calling out the direction of travel. We stayed on the car for about a half hour. At one point we observed Miller stop, I observed him stop and make a telephone call from a pay phone on Campbellton Road just west of 285. We later followed him on to 285. He went south, and we eventually lost him in traffic.
R. 2-31. Agent Harvey positively identified Miller in court as the driver of the Grand Am. R. 2-31. On cross-examination, he declared that when he first saw Miller at the wheel of the car, he was “very close,” so close that the agent became concerned about being “burned,” i.e. discovered by the targets of the investigation. R. 2-34.
Incredibly, the majority dismisses Agent Harvey’s testimony as unreliable because he did not have “a good view of the suspect.” Of course this inaccurate assessment of the evidence overlooks the fact that Agent Harvey saw Miller get out of the Grand Am and make a telephone call minutes after the drug deal had transpired, and that the entire transaction took place in broad daylight. In determining that Agent Harvey’s identification of Miller was not credible, the majority assumes a function the law has wisely assigned to the jury or factfinder. Having before us only a cold record, this court is not competent in most cases to decide which witnesses had a good view. There is absolutely no reason to second guess the jury’s assessment of Agent Harvey’s testimony.
Agent Bledsoe also identified Miller as the cocaine supplier. She recalled that on January 7, 1987, while buying cocaine in an undercover capacity, Lyons told her that he would beep “Louis,” the man who would bring the contraband. R. 2-66. Lyons also informed Agent Bledsoe that Louis would be driving either a Mercedes, a blue Chevrolet Blazer or a Grand Am. R. 2-67. Agent Bledsoe said that Louis Miller arrived at Lyons’ mother’s house at 2:00 in the afternoon, the driver and only occupant of a blue Grand Am. R. 2-67. She identified Miller in court as the driver of the vehicle. R. 2-67. Agent Bledsoe testified that after the cocaine was delivered to her on January 7, 1987, Lyons let her know that Louis would have higher quality cocaine for sale later in the day. R. 2-96. As the majority points out, Agent Bledsoe revealed on cross-examination that she observed Miller by looking into her vehicle’s rear view mirror. R. 2-99. However, she expressed no uncertainty about the identity of the Grand Am’s driver, explaining, “I immediately observed Mr. Miller and recalled I knew him from a previous meeting at the ‘Electric Blue.’ ” R. 2-96. According to Agent Bledsoe, the Electric Blue is a disco in Greensboro, Georgia, and there she and her partner had approached Miller one night in 1986 to discuss purchasing cocaine. R. 2-86-87. While Agent Bledsoe had stood by watching, her partner had spoken with Miller on that occasion. R. 2-93. She identified Miller in court as the individual
Finally, the testimony of Labron Lyons, Miller’s co-defendant, established that Miller participated in the January, 1987, sale. Lyons stated unequivocally that Louis Miller supplied the cocaine for sale to Agent Bledsoe on January 7, 1987. R. 2-43. His testimony fully corroborated that of Agent Bledsoe as to the particulars of that transaction. The defense attacked Lyons’ veracity, uncovering the details of the witness’ drug use, prior conviction and plea bargain. This material, though, properly was weighed by the jury, whose role was to decide whether Lyons “was far from a good witness.” While the majority clearly was swayed by the defendant’s attack on Lyons’ credibility, the jury may well have found believable his testimony of the events of January 7, 1987.
The panel barely acknowledges that extrinsic act evidence is admissible for purposes other than to prove a person’s bad character, such as to show a method of operation or identity. It would grudgingly allow other act evidence on such grounds only where the extrinsic offense and the charged crime are somehow unique. In the majority’s view, where the charged crime and the other crime are “routine” drug transactions,
The key which unlocks Rule 404(b) and opens the door to extrinsic act evidence showing identity or method of operations is the high degree of similarity between the charged and uncharged offenses, not, as is suggested in the opinion, the relative uniqueness of the crimes. See United States v. Beechum, 582 F.2d 898, 912 n. 15 (5th Cir.1978) (en banc), cert. denied, 440 U.S. 920, 99 S.Ct. 1244, 59 L.Ed.2d 472 (1979). Ironically, the majority misreads the Beechum dicta it venerates, using selective quotation to mischaracterize the meaning of the relevant passage. What the former Fifth Circuit Court of Appeals actually said was:
The identity of the defendant may be established by evidence of offenses extrinsic to the indictment. In this in*1550 stance, the likeness of the offenses is the crucial consideration. The physical similarity must be such that it marks the offenses as the handiwork of the accused. In other words, the evidence must demonstrate a modus operandi. Thus, “[a] much greater degree of similarity between the charged crime and the uncharged crime is required when the evidence of the other crime is introduced to prove identity than when it is introduced to prove a state of mind.”
Id. (citations omitted; emphasis added).
Illustrative of this circuit’s view is United States v. Messersmith, 692 F.2d 1315 (11th Cir.1982). In that case, the court held admissible evidence of another crime, observing:
The drug smuggling operation which these defendants conducted two weeks before the crime charged is clearly relevant to their identity as the individuals who committed this crime. The operation which the informant described has the same unusual features as the crime charged. Both trips were financed by [the same third party], both trips involved the two defendants, both trips involved a departure from Costa Rica to Colombia with an intended fuel stop at Caicos, and both trips were made for the purpose of picking up large quantities of quaaludes. Given these similarities, as well as the short space of time between the two incidents, the court properly found that the probative value of the evidence was not outweighed by its prejudice to the defendants.
Id. at 1320-21. Certainly, the drug smuggling operations in Messersmith were no more unique than the crimes in this case. Quaaludes are not rare, and smuggling them into this country from South America by plane is common.
Another example of this circuit’s focus on similarity as opposed to novelty is the court’s recent decision in United States v. Stubbins, 877 F.2d 42 (11th Cir.1989). In Stubbins, the defendant was convicted for selling crack cocaine to DEA Agent Ronald Boston on July 16, 1987. During the transaction, the defendant had expressed reservations about doing the deal, explaining to the undercover agent that he was already under indictment. Afterward, Agent Boston wrote in his report that the offense had taken place at 1808 24th Avenue. He later learned the correct address to be 1808 23rd Avenue. Over defense objection, the district court allowed Tampa Police Officer Paul Miller to testify that on April 9, 1987, he had seen the defendant attempt to sell crack on the street outside 1808 23rd Avenue. As of July 16, 1987, charges still were pending against the defendant as a result of the prior offense. Id. at 43.
This court affirmed the defendant’s conviction, “holding ... that the extrinsic offense [was] sufficiently similar so as to be relevant to show identity.” Id. at 44. Discussing its reasons for so deciding, the court wrote:
In this case, the distinctive feature of both offenses was that they occurred at the same address. As the district judge stated in his ruling on the evidence question, “both offenses were involved with or originated in or were a part of the use of the same premises and ... that factor is sufficiently unusual and distinctive as to constitute a modus operandi ...” The defense was mistaken identity and counsel made much of the fact that Agent Boston’s report had a different address from the one he gave in his testimony. The extrinsic evidence helped to explain how Boston later discovered,*1551 from conversations with Miller, the incorrect address on the report and supported his identification of the defendant as the person who sold him the cocaine. Further, the person who sold cocaine to Agent Boston stated that he was already under indictment. This further served to indentify defendant as the person involved in the cocaine sale.
Id. (emphasis added).
Nothing in the Stubbins opinion suggests that the crack sales were in any way unique. Actually, the transactions seem to have been rather ordinary. The court’s emphasis was on the likeness of the offenses, as it should have been. The majority opinion thus is plainly inconsistent with the Stubbins decision. In the instant case, one of the many similarities between the two crimes was that they occurred at the same address. In Stubbins, the fact that both incidences had occurred at the same location sufficiently marked the offenses as the handiwork of the accused. Stubbins therefore compels the conclusion that the evidence of the extrinsic act in the present case demonstrated a modus operandi and was properly admitted to show identity.
Because the majority cannot harmonize its holding with that of the panel in Stub-bins, the opinion attempts an unreasonable restatement of the Stubbins decision. First, no fair reading of Stubbins permits the conclusion that this court rejected the district court’s reasoning that the fact that the crimes had occurred at the same place made them sufficiently distinctive to constitute a modus operandi. Rather, the panel approved the district court’s statement. This must be so, because as far as one can tell from the opinion, the only similarities between the two crimes were the common site and the sale of same illegal substance. The court explicitly rejected the notion that any prior crack offense would always be admissible, id. at 44. Therefore, the fact that the two crimes had occurred at the same address was the key similarity, the feature which let in the extrinsic act evidence. Second, the majority’s assertion that the extrinsic crime in Stubbins was not used as circumstantial evidence of identity but rather “was used as direct evidence to explain [Agent Boston’s] identification of the dwelling” cannot be taken seriously. Certainly Officer Miller’s testimony corroborated Agent Boston’s and thereby strengthened the agent’s identification of the defendant. But Miller’s testimony did not directly connect the defendant with Agent Boston, nor Agent Boston with 1808 23rd Avenue. Cf. United States v. Clemons, 676 F.2d 122, 123 (5th Cir. Unit B 1982) (in drug prosecution, deputy’s testimony about subsequent drug purchase was admissible, where testimony helped establish defendant’s identity and connected him with confidential informant); Sanchez v. United States, 341 F.2d 565, 566 (5th Cir.), cert. denied, 382 U.S. 860, 86 S.Ct. 119, 15 L.Ed.2d 98 (1965) (evidence that defendant had been seen in detective’s office two months prior to bank robbery relevant to detective’s identification of defendant). It did so indirectly, and thus the extrinsic offense was circumstantial evidence of the defendant’s identity and of the scene of the crime. Not only does the majority’s interpretation of Stubbins contradict that panel’s express holding, its alternative ratio decidendi is legally unsound under the facts of that case.
In the instant case the two transactions are described as being almost dissimilar. Yet the two drug deals were identical in all important particulars, and the majority’s contrary claim that “the only similarity between the transactions was that they involved a similar means of delivering cocaine” ignores the reality contained in the record. Both the September transaction and the January transaction share certain noteworthy features. In both instances, Agent Bledsoe posed as a buyer of cocaine. Both times Labron Lyons brokered the deal, contacting Miller by using a beeper. Both transactions involved similar amounts of cocaine — one ounce in the first sale and two ounces in the latter one. The site of both sales was the home of Lyons’ mother. Each operation unfolded in a similar manner. Miller arrived at the house shortly after being beeped by Lyons. Lyons entered Miller’s vehicle, and they drove around the block for several minutes. While in the car, money and drugs changed hands, Lyons tendering the cash and Miller
The majority admits that the extrinsic evidence was relevant in this case to show intent, but, after applying the Rule 403 balancing test, it concludes that the district court abused its discretion because the probative value of the evidence was substantially outweighed by the danger of unfair prejudice to the defendant. In so holding, the majority advances several arguments which largely echo those made earlier in its opinion. The first is the contention that the similarities between the January and September transactions are attributable to the government’s “orchestration” of the second sale. Even if I agreed that the government “orchestrated” the September transaction, the majority’s faulty premise, which it repeats no less than five times in its opinion, finds no support in our cases, and is wholly inconsistent with the former Fifth Circuit’s decision in Alston, a case which involved at least as much “orchestration” as is found here.
A second factor considered by the majority is the length of time, nine months, separating the charged crime and the extrinsic offense. The opinion hints that this “remoteness,” coupled with the “dissimilarity” between the offenses, renders the other act evidence less probative. I would disagree and cite United States v. Hitsman, 604 F.2d 443, 448 (5th Cir.1979) (where offenses were two to three years apart, prior drug crime was probative of intent).
Lastly, the majority attackes the government’s “weak” case. I have already referred to this issue in detail but reemphasize that I find the government’s evidence of the January transaction to be quite strong.
Accordingly, I cannot conclude that the district court abused its discretion in allowing the government to introduce the evidence of the September, 1987, transaction for the purpose of showing the defendant’s state of mind. I disagree that the facts of this case are unique. It is precisely because district courts frequently face such evidentiary questions that I vigorously oppose the majority decision. I fear that the opinion, despite its invocation of the clear abuse standard, will unreasonably fetter the broad discretion of the district courts in this area, and I fear that probative, relevant evidence will be excluded unnecessarily as a result.
Even assuming the district court did abuse its discretion in admitting the extrinsic act evidence, the error was at most harmless, for the other evidence against Miller was substantial.
. Whether extrinsic act evidence is admissible to corroborate the testimony of a government witness whose credibility has been attacked by the defense is an open question in this circuit. United States v. Williford, 764 F.2d 1493, 1498 n. 1 (11th Cir.1985). The majority’s sweeping dictum to the contrary should be disregarded and the issue settled in an appropriate case. Were other avenues of admissibility not available here, this case might well be an appropriate vehicle for making that determination.
. The majority writes that the operation here was a "common method of drug trafficking.” Since there is absolutely no evidence in the record to support this assertion, I must assume that the court is utilizing its common knowledge of the commerce in contraband. If, however, this transaction were in fact routine, it seems to me that a case or two could be found presenting similar facts to support the statement. I am skeptical that the transaction at issue was routine, but resolution of this issue is not necessary to my analysis.
. In Bonner v. City of Prichard, 661 F.2d 1206, 1209 (11th Cir.1981), the Eleventh Circuit adopted as precedent the decisions of the Fifth Circuit rendered prior to October 1, 1981.
. See, e.g., United States v. O’Neill, 767 F.2d 780 (11th Cir.1985); United States v. Nabors, 707 F.2d 1294 (11th Cir.1983), cert. denied, 465 U.S. 1021, 104 S.Ct. 1271, 79 L.Ed.2d 677 (1984); United States v. Hawkins, 661 F.2d 436 (5th Cir. Unit B 1981), cert. denied sub nom., McCain v. United States, 456 U.S. 991, 102 S.Ct. 2274, 73 L.Ed.2d 1287 (1982); United States v. Stone, 659 F.2d 569 (5th Cir. Unit B 1981); United States v. Kragness, 830 F.2d 842 (8th Cir.1987); United States v. Bonadonna, 775 F.2d 949 (8th Cir.1985); United States v. Roper, 716 F.2d 611 (4th Cir.1983); United States v. Burchinal, 657 F.2d 985 (8th Cir.), cert. denied, 454 U.S. 1086, 102 S.Ct. 646, 70 L.Ed.2d 622 (1981).
. The majority quotes my statement out of context to support its contention that my conclusion "relies on pure speculation.” The majority insists upon putting an astonishing “defendant’s spin” on the government's case, discounting or simply disregarding the testimony of Agents Harvey and Bledsoe and co-defendant Lyons as to Miller’s participation in the January crime. Despite the strong evidence, I disagree with the opinion's argument that "there was [therefore] no need for the government to use the evidence of the September transaction at trial.” The prosecution is obligated to present the strongest case possible under the law. Our role is neither to tell the government how to try its case nor to second guess the district court’s reasonable evi-dentiary ruling where admissibility was a close question.