Appeal from the United States District Court for the Western District of Wisconsin. No. 06 CR 105-Barbara B. Crabb, Chief Judge. Appeal from the United States District Court for the Southern District of Illinois. No. 05-30196-WDS-William D. Stiehl, Judge.
The opinion of the court was delivered by: Posner, Circuit Judge.
Before POSNER, RIPPLE, and TINDER, Circuit Judges.
We have consolidated for decision two appeals presenting overlapping issues under Fed. R. Evid. 404(b) (admissibility of evidence of previous crimes or other prior "bad acts"). Tyreese Taylor and Samuel Hogsett were convicted in separate trials of distributing crack (Hogsett was also convicted of a gun offense) and sentenced to 240 and 355 months in prison, respectively.
Rule 404(b) forbids the use of prior convictions or other evidence of bad acts to prove that the defendant has a propensity to commit crimes. But it allows such evidence to be presented (in the discretion of the trial judge, balancing probative value against prejudice to the defendant under Fed. R. Evid. 403) to prove other, material facts, including criminal intent, identity, and absence of mistake. Taylor's lawyer told the judge before the trial began that he was going to request an instruction that would permit the jury to convict his client of the lesser offense of possession of crack for personal use rather than for sale. The judge ruled that the request opened the door to the government's presenting evidence of Taylor's prior conviction of possession of crack with intent to distribute it. But after the lawyers' opening statements to the jury, Taylor's lawyer withdrew the request for a lesser-included instruction, and the judge told the prosecutor that he could introduce evidence of prior bad acts, to rebut an inference that Taylor possessed drugs only for his personal use, only if the defendant opened the door to such evidence in some other way during the trial. The defendant did not do that.
The judge did allow the persons who had bought crack from Taylor in transactions that he was accused in this case of having made to testify that they knew from prior dealings with him that he was indeed a crack dealer. That evidence, like the prior conviction, related to his intent to distribute crack rather than to possess it just for personal use. He did not make an issue of intent, as we have just seen, but the buyers' evidence of prior dealings with him also related to identity; the evidence explained how they knew and thus could identify him. Although no one questioned these witnesses' ability to identify him as the person from whom they had bought crack in the transactions charged by the prosecution, we hesitate to pronounce the admission of their testimony of prior dealings with him a violation of Rule 404(b). The fact that a defendant pleads not guilty does not provide many clues to the specific attacks that he will mount against the government's case. Unless the government is allowed to present some evidence about previous transactions between the government's witnesses and the defendant, the transactions alleged in the government's current case could be challenged in the defendant's closing argument as unworthy of belief, especially since the buyer witnesses would be criminals who might be hoping for lenity by testifying for the government. It would be too late for the government to attempt to rehabilitate those witnesses. (We shall give another and clearer example of this sandbagging concern when we discuss Hogsett's appeal.)
The evidence that Taylor was known to be a seller of illegal drugs also explained why the govern-ment's informants had identified him as a potential seller for a controlled buy and how they knew who he was and what car he drove; the police officer who had observed the drug scene, being outside the house in which two of the three controlled buys took place, did not see Taylor hand over the drugs. The buyers' previous knowledge about him related to the accuracy (hence absence of mistake) of their testimony concerning the controlled buys that provided crucial evidence for the government's case.
But clearly the judge should not have allowed the officer who arrested Taylor to testify that he had recognized him as a result of having known him "throughout [the officer's] career as a police officer and as a drug and gang officer" (emphasis added). There was no doubt of the identity of the arrested person.
The government offers two other reasons for the admission of this damaging testimony, which implied that Taylor had a long history of drug and gang activity. First, the officer had arrested him after observing illegal tinting on his car windows and an illegal tinted cover on his rear license plate. But the officer testified that he had not made the decision to arrest for so trivial a traffic offense until he recognized that it was Taylor's car. He also testified that he knew there was an outstanding warrant for Taylor's arrest (on what charge the jury was not told).
That evidence was at once irrelevant and damaging, as was the officer's testimony about his prior professional knowledge of Taylor. It is not as if the government had to try to justify the arrest on the basis not of the traffic offenses but of suspicion that Taylor was a drug dealer. Not only was the legitimacy of the arrest for the traffic offenses not questioned; it was an issue for the judge rather than for the jury to decide. United States v. McKinney, 919 F.2d 405, 414 (7th Cir. 1990); see Jones v. United States, 362 U.S. 257, 264 (1960); United States v. Nunez-Rios, 622 F.2d 1093, 1098-99 (2d Cir. 1980); Fed. R. Crim. P. 12(b)(3)(C).
The government appeals (in Hogsett's case as well) to the principle that bad-act evidence "inextricably intertwined" with admissible evidence may be admitted without regard to the specific exceptions in Rule 404(b), and argues that all the bad-act evidence in Taylor's case was of that character. Although many cases recite the "inextricably intertwined" formula, see, e.g., United States v. Luster, 480 F.3d 551, 556-57 (7th Cir. 2007), and cases cited in United States v. Bowie, 232 F.3d 923, 927-28 (D.C. Cir. 2000), it is unhelpfully vague. Courts do not agree on whether it refers to evidence "intrinsic" to the charged crime itself, in the sense of being evidence of the crime, or whether though evidence of another crime it may be introduced in order to "complete the story" of the charged crime. As explained in the Bowie opinion, neither formulation is satisfactory: to courts adopting the former, "inextricably intertwined evidence is intrinsic, and evidence is intrinsic if it is inextricably intertwined," while "the 'complete the story' definition of 'inextricably intertwined' threatens to override Rule 404(b). A defendant's bad act may be only tangentially related to the charged crime, but it nevertheless could 'complete the story' or 'incidentally involve' the charged offense or 'explain the circumstances.' If the prosecution's evidence did not 'explain' or 'incidentally involve' the charged crime, it is difficult to see how it could pass the minimal requirement for admissibility that evidence be relevant. See Fed. R. Evid. 401 and 402." 232 F.3d at 928.
What is true, but irrelevant to this case, is that if a defendant commits two criminal acts at the same time and is charged with only one, the evidence of the charged crime may unavoidably reveal the uncharged one, as in Ignacio v. People of Territory of Guam, 413 F.2d 513, 519-20 (9th Cir. 1969), and United States v. Persico, 425 F.2d 1375, 1384 (2d Cir. 1970). In such a case-for example where the defendant assaults a person in the course of buying illegal drugs from him but is prosecuted only for possessing drugs with intent to distribute them, or where he commits an armed robbery but is charged only with being a felon in possession-the evidence of the "other" crime can't be disentangled from the evidence of the charged crime. Or if a defendant makes an issue of his criminal intent, as Taylor initially did by his later-abandoned request for an instruction on simple possession, his previous activities may become relevant to inferring his state of mind with regard to the current charges. If in the past he possessed small quantities of crack with intent to sell rather than merely to consume, this would be some evidence that his current modest inventory also was not just for his own consumption. Moreover, as we said, the fact that a defendant's buyers had dealt with him previously could explain how they were able to identify him, why they picked him for the controlled buy, and why he was willing to deal with them.
But intent and absence of mistake are express exceptions to the Rule 404(b) bar; there is no need to spread the fog of "inextricably intertwined" over them. Almost all evidence admissible under the "inextricably interwoven" doctrine is admissible under one of the specific exceptions in Rule 404(b), or under the judge-made "no confusion" exception discussed later in this opinion. The objection to "inextricably interwoven" is that its vagueness invites prosecutors to expand the exceptions to the rule beyond the proper boundaries of the exceptions.
The arresting officer's testimony about the real reason for his arresting Taylor does not connect to any of the exceptions. It was just a way of telling the jury that the officer knew Taylor to have been a drug offender and gang member for a long time and that at the time of the arrest Taylor was a wanted criminal. But the violation of Rule 404(b) was harmless because the evidence of Taylor's guilt was overwhelming. Two confidential informants testified that they had bought crack from Taylor in controlled buys set up by the police. A third witness saw one of those drug deals taking place. The informants' accounts were corroborated by a police officer, while another officer testified that Taylor had had a bag of crack in his possession when he was arrested. Taylor called no witnesses of his own and offered no evidence of innocence. No adverse inference can be drawn from his failure to testify, ...