Summary
holding that § 2113(e)"duplicates the general federal felony-murder statute"
Summary of this case from United States v. McDuffyOpinion
No. 13–1812.
2014-08-19
UNITED STATES of America, Plaintiff–Appellee, v. David VANCE, Defendant–Appellant.
Sharon R. Fairley, Office of the United States Attorney, Chicago, IL, for Plaintiff–Appellee. Geoffrey S. Goodman, Foley & Lardner LLP, Chicago, IL, Thomas L. Shriner, Jr., Foley & Lardner LLP, Milwaukee, WI, for Defendant–Appellant.
Sharon R. Fairley, Office of the United States Attorney, Chicago, IL, for Plaintiff–Appellee. Geoffrey S. Goodman, Foley & Lardner LLP, Chicago, IL, Thomas L. Shriner, Jr., Foley & Lardner LLP, Milwaukee, WI, for Defendant–Appellant.
Before POSNER, FLAUM, and MANION, Circuit Judges.
POSNER, Circuit Judge.
David Vance, along with Alton Marshall and Henry Bluford, were charged with having committed two bank robberies in Chicago in 2007, 18 U.S.C. § 2113(a), as well as related crimes, such as conspiring to rob the banks. 18 U.S.C. § 371. Vance also was charged with the more serious offense of killing a person during the commission of a bank robbery. 18 U.S.C. § 2113(e).
Marshall and Bluford, but not Vance, pleaded guilty. Vance was tried by a jury, and Marshall testified at the trial. The jury convicted Vance, and the judge sentenced him to terms of years on all counts except the one charging the killing; for that the judge sentenced him to life in prison. The judge made the sentences on two of the counts—sentences to fixed terms of prison of 7 and 25 years, respectively—consecutive to the sentences on all the other counts, including therefore the sentence of life imprisonment. The judge imposed conditions of supervised release, which of course do not take effect until the defendant is released. The imposition of sentences consecutive to a life sentence and of conditions of supervised release when a life sentence is imposed presents issues that we discuss briefly at the end of the opinion.
Why it's taken seven years from the indictment to the argument of the appeal is unexplained, though we can piece together some of the reasons from the record. Marshall and Bluford didn't plead guilty till 2009; after that there probably was a protracted back and forth with Vance to see whether he would cooperate and plead guilty. Also, the parties filed numerous motions. The government for example moved to disclose grand jury testimony and the defense asked the district court for money to pay expert witnesses yet refused to submit a budget. The defense also twice asked the district judge to delay the trial. And the pre-sentencing report from the probation service was not submitted to the court more than a year after the jury had found Vance guilty, though it appears that the delay in submitting the report was attributable to post-trial motions rather than to any lassitude by the probation service.
According to Marshall's testimony, the night before the first of the two bank robberies Vance suggested to him that they commit robberies that night. Marshall agreed, and that night the pair robbed a seafood restaurant, a Mexican restaurant, and a diner. They wore ski masks (the same ones used later in the bank robberies) and gloves. Vance carried a long-barreled .44 caliber revolver, the same model later used in both bank robberies. Vance told Marshall that he (Vance) would approach the counter to get the restaurant's money while Marshall “watch[ed] his back,” and that was how they conducted two of the restaurant robberies. In the third, the robbery of the Mexican restaurant, both men approached the counter and ordered the employees there to open the register.
Vance argues that Marshall's testimony about the three restaurant robberies should not have been admitted; that if believed all it proved was that Vance had a propensity to commit robberies, and propensity evidence is not admissible, lest it prejudice the jurors against the defendant, causing them to convict even if the evidence of his guilt of the crime that he is being tried for is weak.
But Rule 404(b) of the Federal Rules of Evidence, the source of the prohibition against introducing evidence of a defendant's propensity to commit crimes, allows (in subsection (b)(2)) the admission of evidence of other crimes besides the one the defendant is charged with if that evidence pertains to, among other things, “identity,” in this case Vance's identity as one of the masked bank robbers. Similarities between the restaurant robberies and the bank robberies supported an inference that if, as Marshall testified, Vance had been one of the restaurant robbers he probably had been one of the bank robbers as well. Remember that in two of the restaurant robberies Vance had rushed the counter where the money was kept while Marshall watched the patrons (though, in the third both had rushed the counter). There had been a “rusher” in the bank robberies as well, and if Vance had been the rusher in the restaurant robberies this made it more likely that he had also been the rusher in the bank robberies—and whoever rushed the counter in the second bank robbery was also the teller's killer. Vance brandished a .44 caliber revolver in the restaurant robberies, as he did in one of the bank robberies. And Marshall and Vance trusted each other enough as accomplices in robbery—as shown by their joint commission of the three restaurant robberies—to make it seem likely that Vance was also a participant in the bank robberies, to which Marshall had already confessed.
This body of evidence was not conclusive in identifying Vance as one of the bank robbers, but it was not so flimsy that it had to be excluded on the ground that its probative value was outweighed by its prejudicial effect. Weakest was the caliber evidence, since .44 caliber revolver ammunition is common. But there are a number of equally or more common calibers of such ammunition, such as .22, .32, .357. 38, and .45, so the fact that .44 is common does not negate an inference that Vance is likely to have been armed with the same weapon in both robberies (rather than that another of the robbers just happened to have a weapon of the identical caliber), thus increasing the likelihood that he was the killer in the bank robbery. Moreover, when several pieces of evidence point in the same direction, the probability that it's the right direction is greater than the probability that any one of the pieces is accurate. Suppose the probability that the first piece of evidence is accurate—and thus that Vance was indeed one of the bank robbers—is .50, that the probability that the second is accurate is .40, but that the probability that the third is accurate is only .10. Still, the likelihood that none of the evidence is accurate—and Vance therefore was not one of the bank robbers after all—is lower when the third piece of evidence is considered than when it is not; excluding it in our example would raise the probability of erroneous identification of Vance as one of the bank robbers from 27 to 30 percent. ((1 - .5) x (1 - .4) x (1 - .1) = .27; (1 - .5) x (1 - .4) = .30.)
It could be (though is not) argued that since Marshall was the source of the evidence of Vance's participation in the restaurant robberies and also the evidence of his participation in the bank robberies, the restaurant-robberies evidence had no probative value; if the jury thought Marshall was telling the truth about the bank robberies, that was enough to nail Vance; if they disbelieved him, his testimony about the restaurant robberies was irrelevant because Vance wasn't on trial for those robberies. But that analysis is incomplete. The restaurant evidence if believed showed that Marshall and Vance were partners in crime and thus increased the likelihood that Vance had participated in the bank robberies, “rushed the counter,” and shot the teller with a .44 caliber pistol.
When other-crimes evidence is used for a proper purpose, such as to determine identity, rather than for the improper purpose of demonstrating that the defendant has a propensity to commit crimes (he committed a prior crime, so probably he committed the crime he's currently accused of), it is important both that the proof value of the evidence not be substantially outweighed by its prejudicial effect and that the jury be carefully instructed to limit its consideration to the effect of the evidence to determining identity. As this court explained in United States v. Gomez, No. 12–1104, p. 24 (7th Cir. Aug. 18, 2014) (en banc), regarding the first of these requirements, the relevance of the evidence for a proper purpose “must be established through a chain of reasoning that does not rely solely on the forbidden inference that the person has a certain character and acted in accordance with that character on the occasion charged in the case. If the proponent [the prosecutor] can make this initial showing, the district court must in every case assess whether the probative value of the other-act evidence is substantially outweighed by the risk of unfair prejudice and may exclude the evidence under Rule 403 if the risk is too great. The court's Rule 403 balancing should take account of the extent to which the non-propensity fact for which the evidence is offered actually is at issue in the case.”
Vance's identity as one of the bank robbers is of course at issue in this case, and the evidence that he was one of them does not depend on inferring that if he had committed robberies in the past he was likely to have committed the robbery that he was charged with in the present case.
The judge twice gave limiting instructions, telling the jury it could not use the evidence of the restaurant robberies to infer that Vance had a propensity to commit crimes, specifically robberies. The need for such instructions arises from the danger that jurors who are given evidence of a defendant's prior crimes will conclude from it that the defendant is a bad, dangerous person who should be convicted even if the proof that he committed the crime that he's being tried for is weak. The restaurant robberies showed that Vance was indeed a robber, seemingly a compulsive one; a jury might think Vance so dangerous that it should resolve any doubts about his guilt of the bank robberies against him rather than in his favor.
Hence the need for limiting instructions—but there are grounds for skepticism about their efficacy in preventing juries from reasoning in that way. Krulewitch v. United States, 336 U.S. 440, 453, 69 S.Ct. 716, 93 L.Ed. 790 (1949) (Jackson, J., concurring) (“the naive assumption that prejudicial effects can be overcome by instructions to the jury all practicing lawyers know to be unmitigated fiction”); Maus v. Baker, 747 F.3d 926, 927–28 (7th Cir.2014); United States v. Mazzone, 782 F.2d 757, 764 (7th Cir.1986); United States v. Jones, 455 F.3d 800, 811 (7th Cir.2006) (concurring opinion); Nash v. United States, 54 F.2d 1006, 1006–07 (2d Cir.1932) (L. Hand, J.); Edith Greene & Mary Dodge, “The Influence of Prior Record Evidence on Juror Decision Making,” 19 Law & Human Behavior 67, 76–77 (1995); Roselle L. Wissler & Michael J. Saks, “On the Inefficacy of Limiting Instructions,” 9 Law & Human Behavior 37, 38–39 (1985).
And the second limiting instruction, given at the end of this case, was murky. It stated: “You have heard evidence that the defendant was involved in conduct other than that charged in the indictment. You may consider this evidence only in deciding questions of defendant's preparations or plans to commit the charged offenses or the identity of the persons involved in the charged offenses. You should consider this evidence only for these limited purposes.” As the government explains in a footnote to its brief, the reference to Vance's other conduct included not only the restaurant robberies but also “the car thefts [that is, the theft of the cars used in the bank robberies] as proof of defendant's preparation and planning.” The judge should have separated out the restaurant robberies, which had nothing to do with the preparations for the bank robberies, from the bank robberies that were the subject of the trial.
The instructions also were too abrupt, in failing to explain to the jury why it could not use the evidence to demonstrate propensity. The reason is not obvious and there is a danger that if given a counterintuitive command without an explanation a jury will ignore it. “Do this because I say so” may work with children but is unlikely to work with adults. To quote from the Gomez opinion again (at p. 26), “we see no reason to keep the jury in the dark about the rationale for the rule against propensity inferences. Lay people are capable of understanding the foundational principle in our system of justice that ‘we try cases, rather than persons.’ ... The court's limiting instruction would be more effective if it told the jurors that they must not use the other-act evidence to infer that the defendant has a certain character and acted ‘in character’ in the present case because it does not follow from the defendant's past acts that he committed the particular crime charged in the case. Finally, the instruction would be improved by tying the limiting principle to the prosecution's burden of proof. The jurors should be reminded that the government's duty is to prove beyond a reasonable doubt every element of the specific crime charged, and it cannot discharge its burden by inviting an inference that the defendant is a person whose past acts suggest a willingness or propensity to commit crimes” (emphasis in original).
None of these things was done in this case, but Vance does not complain about the limiting instructions and so there is no issue of their validity.
Nor does he complain that the judge or the jury improperly balanced probative value and prejudicial effect. Rather he takes the extreme position that the evidence of the restaurant robberies had only “very small” or even zero probative value, mainly because Marshall testified, and Vance did not deny, that they were long-time friends and so the evidence concerning the restaurant robberies merely confirmed that acknowledged fact. Actually it did more; it pointed to Vance as the probable triggerman in the bank robbery. In all three restaurant robberies Vance was the robber who rushed the counter while Marshall watched his back (although Marshall did join him at the counter in one of the robberies). The fact that when Vance and Marshall robbed together (at least when they robbed together on the night before the first bank robbery) Vance rushed the counter makes it more likely that he also rushed the counter at the second bank, where the teller was killed; video surveillance evidence showed that whoever rushed the counter at the second bank shot the teller moments later. Furthermore, there is a difference between partnership in robberies and a personal friendship. If Vance and Marshall were partners in robbery, there is at least some likelihood that in a robbery at which Marshall is conceded to have been present (the bank robbery), one of the other robbers was indeed Vance.
Granted, the evidence of the restaurant robberies did not have a great deal of probative value—but for a reason that does not help Vance's defense. It had very limited incremental probative value because the other evidence of Vance's guilt (discussed next) was very strong. But by the same token the incremental prejudicial effect of the evidence regarding the restaurantrobberies must also have been slight, because for an effect to be prejudicial means that it may actually have swayed the jury's decision. If conviction was a forgone conclusion, the evidence of the restaurant robberies was icing on the cake; the government didn't need it to convict Vance.
The other evidence of Vance's guilt included Marshall's eyewitness testimony about the bank robberies, DNA evidence implicating Vance, the identification of Vance as one of the robbers by a teller at the first bank robbed and also by Vance's own girlfriend, and testimony by Marshall's girlfriend that further implicated Vance. Taken all in all, this evidence, wholly apart from the evidence about the restaurant robberies, was compelling. Even without that evidence, no reasonable jury could have acquitted Vance.
The teller testified that she recognized Vance, whom she had known since childhood, as the robber who approached her and demanded money. Marshall's girlfriend testified that Vance, Marshall, and Bluford arrived at her apartment the evening before the day planned for the second robbery (although that robbery did not actually take place for another two days) and that Vance was carrying bags of guns and ammunition. She also testified that Marshall had confessed to her that he had committed the second bank robbery along with Vance and Buford and that Vance had been “the man with the gun going over the counter,” and thus the killer. Vance's girlfriend testified that she recognized him in a surveillance photograph taken during the first bank robbery.
Vance's counsel tries to pick apart the government's evidence—other than that of Marshall's girlfriend, which by itself would have been sufficient to convict Vance, since Marshall's confession to her, made before he had cut a plea deal with the government in exchange for testifying against Vance, was recorded by the wire that the FBI had given her to wear. Counsel notes, however, that Marshall had lied repeatedly in his dealings with the prosecutors and claims that this proved him to have been a “pathological liar” none of whose testimony should have been believed. Counsel also argues that the DNA evidence should not have been admitted and that the female teller and Vance's girlfriend could not have identified him because the robber whom they identified as Vance was masked.
The jury was entitled to reject these arguments. Regarding Marshall's lies, falsus in uno, falsus in omnibus (false in one thing—or for that matter more than one thing—therefore false in everything) is not a doctrine of American law, United States v. Edwards, 581 F.3d 604, 612 (7th Cir.2009), and isn't even true. When Mary McCarthy famously said of Lillian Hellman “that every word she writes is a lie, including ‘and’ and ‘the,’ ” Hellman sued her for defamation. (She died before judgment, and her estate dropped the suit, even though New York had abrogated the common law rule that an action for defamation does not survive the victim's death. N.Y. Estates, Powers & Trusts Law § 11–3.2(b).) The jury was entitled to believe Marshall's testimony that Vance was one of his accomplices in the bank robberies, as well as in the restaurant robberies, and also that Vance had admitted to him after the second robbery to having shot the teller who died.
Vance's DNA was found on latex gloves near the getaway car used in the second bank robbery; and Marshall testified that the robbers had worn such gloves. Vance's counsel complains that the police investigator put the gloves in a plastic bag rather than a paper bag, which is a no-no because such a bag traps moisture,which tends to dissolve DNA. E.g., National Institute of Justice, “Evidence Collection and Preservation,” www. ncjrs. gov/ nij/ DNAbro/ evi. html (visited Aug. 7, 2014). But as the government points out, this error could not morph someone else's DNA into a match for Vance's; it could only make Vance's DNA more difficult to find on the gloves—yet it was found, as explained by the government's expert witness. Vance's counsel complains that the witness should have been subjected to a Daubert hearing. But counsel's complaint about the DNA evidence—that the gloves shouldn't have been put in plastic bags—was of no consequence to the validity of the evidence. And it was the error not of this witness but of the FBI investigator who had found the gloves that was being challenged; the expert witness's competence to match DNA taken from Vance to DNA found on the gloves was not in question. Furthermore, the judge conducted several pretrial hearings, at which hundreds of pages of reports prepared by the expert witness, and other relevant documents, were presented and considered. The hearings taken as a whole constituted an adequate Daubert hearing, albeit the expert did not testify at them.
Vance's counsel further complains that the expert demonstrated her incompetence by denying the possibility of secondary transfer of DNA (for example, one person touches a phone, then another, and DNA from the first person is transferred via the phone to the second person). She did not deny it; rather, she testified that she had no reason to think it had occurred in this case—no reason to think Vance's DNA was on the latex gloves by reason of having been transferred to the gloves by someone or something innocently touched by Vance before the robbery.
As for the testimony by the teller who had known Vance since childhood, Vance's counsel argues that she could not have identified Vance as one of the robbers because he was masked. But a photograph taken by a surveillance camera at the bank (reproduced below) shows that Vance's mask did not cover his entire face. That he could be identified from such a photograph, though partially masked, by a person who had known him for many years was entirely plausible. United States v. Williams, 698 F.3d 374, 378 (7th Cir.2012). His girlfriend also testified that he was the robber in the surveillance photograph.
IMAGE
So much for guilt; on to punishment. Vance challenges his sentence of life imprisonment on the ground that 18 U.S.C. § 2113(e)—the statute on which the sentence was based—is a hopeless muddle. It provides, so far as relates to this case, that a person who in the course of committing a bank robbery in violation of federal law “kills any person, or forces any person to accompany him without the consent of such person, shall be imprisoned not less than ten years, or if death results shall be punished by death or life imprisonment.” It's irrelevant that Vance appears not to have intended to kill the teller in the second bank robbery. He shot him in a buttock, and the bullet hit the iliac artery, causing him to bleed rapidly to death. It is reasonably clear that the language we quoted from the statute dispenses with any requirement of proving intent to kill, and in this respect duplicates the general federal felony-murder statute, 18 U.S.C. § 1111(a). See United States v. Jackson, 736 F.3d 953, 957–58 (10th Cir.2013), and cases cited there; United States v. Allen, 247 F.3d 741, 782–83 (8th Cir.2001), vacated on other grounds, 536 U.S. 953, 122 S.Ct. 2653, 153 L.Ed.2d 830 (2002).
But that's all that's clear from the statutory text; for, issues of criminal intent to one side, read literally the statute makes no sense unless the victim happens to be the Biblical Lazarus. For read literally it says that a person who kills another person shall be punished by death or life imprisonment only if death results. But death is always the consequence of being killed. Even Lazarus was dead for four days before (according to the Gospel of John) Jesus Christ restored him to life. John 11:17.
It's silly for the government to invoke the statute's “plain language” and “plain text” and insist that therefore “the sole function of the court [ ] is to enforce it according to its terms,” or to say that “the ‘if death results' language may be considered redundant to the phrase, ‘kills any person’ ” (emphasis added). The statute is a mess. Nevertheless it is apparent what the drafters of the statute and the Congress that enacted it and the President who signed it intended, or if asked (for it is uncertain who ever read the provision, buried as it is in a 356–page statute, Pub.L. 103–322, 108 Stat. 1796 (Sept. 13, 1994)), would have said they intended: if in the course of committing a bank robbery the robber abducts someone, the minimum punishment is ten years in prison, but if he kills someone the minimum is life in prison. United States v. Parks, 700 F.3d 775, 778–79 (6th Cir.2012); cf. United States v. Turner, 389 F.3d 111, 120–21 (4th Cir.2004). We know this is what was intended because the confusing “if death results” passage was substituted for “punished by death if the verdict of the jury shall so direct,” Pub.L. 103–322, § 60003(a)(9), 108 Stat.1969 (Sept. 13, 1994), in more than a dozen provisions of the federal criminal code, in order to eliminate the possibility that a jury would impose a death sentence for a bank robbery in which no one died. See H.R.Rep. No. 103–466, at pp. 12–15 (March 25, 1994). The change had nothing to do with abduction.
What else could the muddled statute mean? Vance's interpretation is that if, as in this case, the person killed in a bank robbery is not abducted, the minimum sentence is 10 years; only if he is both abducted and killed is it life. It would make no sense, however, to allow so lenient a sentence when death occurs, yet make the fact of abduction jack up the minimum sentence to life. Notice how broadly what we're calling “abduction” is defined: “forces any person to accompany him” (him being the robber). This could mean forcing a teller to lead the robber to or toward the vault (which in fact happened here, which makes us wonder why there is any question about the legality of the life sentence)—hardly a circumstance that would warrant jacking up the minimum sentence from 10 years to life.
Vance's counsel invokes the rule of lenity. That rule of statutory interpretation can be understood in various ways: as backing up the prohibition against retroactive criminal punishment by requiring that a criminal statute be at least minimally clear; as underscoring the fact that there are no common law federal crimes and therefore judges are not to create crimes by interpreting ambiguous criminal statutory language as imposing criminal liability; and, least plausibly, as providing “fair warning” of potential criminal liability (unrealistic because most criminals are not familiar with the text of criminal statutes). The statute in this case is not actually vague or ambiguous; the text is nonsensical, presenting the familiar example of a statute or other legal document that can't mean what it says—that read literally is non-sense—yet the meaning of which is obvious; and in such a case the obvious meaning is the legal meaning. “No rule of construction necessitates our acceptance of an interpretation resulting in patently absurd consequences.” United States v. Brown, 333 U.S. 18, 27, 68 S.Ct. 376, 92 L.Ed. 442 (1948); see also United States v. Kirby, 74 (7 Wall.) U.S. 482, 486–87, 19 L.Ed. 278 (1868).
The defendant raises additional issues, but they are peripheral and resolving them in his favor would not alter the outcome. One further point, however, unmentioned by the parties, deserves attention. It concerns the imposition of conditions of supervised release to take effect, as all such conditions do, when the defendant is released from prison after serving his prison sentence. What can it mean to impose on the defendant's post-release conduct restrictions that because he's been sentenced to life in prison will not take effect until his death? Yet this peculiar feature of Vance's sentence is not unique; conditions of supervised release are routinely imposed in life-sentence cases. See, e.g., United States v. Cavender, 228 F.3d 792, 797 (7th Cir.2000); United States v. Rodríguez–Berríos, 573 F.3d 55, 60 n. 1 (1st Cir.2009). Because the practice is routine, we are disinclined to criticize the district judge. And we are aware that the guidelines provisions relating to supervised release do not make an exception for life sentences, though such an exception could, as a matter of common sense, be thought implicit.
When we raised this matter with the government's lawyer at the oral argument, she pointed out that the defendant received multiple sentences, only one of which was a sentence of life imprisonment. Were that sentence to be reversed, whether on appeal or in a post-conviction proceeding or by presidential pardon or commutation, the defendant would be serving a sentence for a fixed number of years and so the conditions of supervised release might someday take effect. Fair enough. Yet Vance's sentence states that among the counts of conviction to which the conditions of supervised release apply is the felony-murder count for which the life sentence was imposed, which will be the only sentence Vance serves if his convictions of the other counts in the indictment are vacated.
A life sentence might be dismissed or commuted and replaced by a term of years. But that is far from certain to happen. The judge imposed 14 non-mandatory conditions of supervised release on Vance, which might in the unusual circumstances of conditions imposed on a defendant sentenced for life be excessive. For future reference, we emphasize the importance of care in the supervised-release stage of sentencing, which sometimes gets short shrift from busy judges. See United States v. Siegel, 753 F.3d 705 (7th Cir.2014); United States v. Bryant, 754 F.3d 443 (7th Cir.2014).
A similar feature of the sentence was to make the term sentences run consecutive to the life sentence, although again it could be argued that this was done in case the life sentence should be commuted to a term of years, or vacated entirely. Obviously the judge did not mean that after dying in prison Vance is to begin serving those other sentences.
As we said, Vance's lawyer has made no issue of the conditions of supervised release imposed as part of the sentence, or for that matter the imposition of sentences to run consecutively to Vance's life sentence. So we shall let the entire judgment stand.
Affirmed.