UNITED STATES of America, Plaintiff, Appellee, v. Henry LOMBARD, Jr., Defendant, Appellant.
This court earlier upheld the convictions of defendant Henry Lombard; but the court vacated the life sentence imposed on one of the counts and remanded for resentencing, holding that the district court had authority to depart downward. United States v. Lombard, 72 F.3d 170, 187 (1st Cir.1995) (“Lombard I ”). On remand, the district court reimposed the original sentence. This new appeal raises a constitutional claim that Lombard made earlier but was not decided on the initial appeal.
The facts are set forth at length in Lombard I, 72 F.3d at 172-76, and only the briefest summary is needed to set the stage. Lombard and his half-brother, Hubert Hartley, were tried in Maine state court for murdering two acquaintances as they slept in Hartley's Maine cabin on Thanksgiving morning in 1990. Despite something close to eyewitness testimony from Hartley's girlfriend, both men were acquitted by juries in separate trials in 1992.
A federal grand jury then indicted Lombard and Hartley for different crimes relating to the same episode. Lombard and Hartley were charged with conspiracy, 18 U.S.C. § 371, the conspiracy having multiple objectives: to possess a firearm in violation of the felon in possession statute, 18 U.S.C. § 922(g), to travel interstate to avoid prosecution, 18 U.S.C. § 1073, and to remove evidence to prevent seizure, 18 U.S.C. § 2232(a). Lombard was also charged substantively under the felon in possession statute and Hartley with aiding and abetting this crime.
The defendants were tried together on the federal charges in 1993. Much of the evidence concerned the commission of the same killings for which they had been acquitted, the evidence being relevant inter alia to the flight and removal of evidence charges. Hartley pled guilty at the close of the government's case. Lombard was convicted on both of the counts directed against him: conspiracy and felon-in-possession. Lombard's convictions were sustained in Lombard I and are not now before us.
At sentencing, Lombard-without regard to the murders-was subject to a statutory sentence of 15 years to life because his prior convictions brought him within the armed career criminal statute. 18 U.S.C. § 924(e). Under the Sentencing Guidelines, again without reference to the murders, the guideline sentencing range would have been roughly between 20 and 30 years. U.S.S.G. § 4B1.4; id. ch. 5, pt. A.1 However, Lombard had so many criminal history points over the number needed for the highest criminal history category that an upward departure might have been imposed. U.S.S.G. § 4A1.3.
However, the ordinary guideline computation went by the boards. The felon in possession guideline provides that where the firearm is used in connection with another offense, the base level should be that of the “object” offense. U.S.S.G. §§ 2K2.1(c)(2), 2X1.1. The base level for premeditated murder requires a life sentence. Id. § 2A1.1; ch. 5, pt. A. Because the district court found by a preponderance of the evidence that Lombard had participated in the premeditated murders, the court imposed a life sentence on Lombard.
On appeal in Lombard I, this court took note of several unusual circumstances, including the impact on the sentence of the uncharged murders, Lombard's prior acquittal of those murders, the qualitative difference between murder and the offense of conviction, and the extreme penalty of life imprisonment. Expressing but not resolving constitutional concerns, the court then held that these special facts gave the district court discretionary authority to depart downward, U.S.S.G. § 5K2.0, and remanded to permit the district court to consider such a departure. 72 F.3d at 184-85.
At the resentencing, the district court said that it fully understood (and had understood previously) its authority to depart downward. But the court remained convinced that “the appropriate sentence in this case is the sentence that was imposed initially,” and it reimposed the life sentence. Lombard now appeals again, stating that the single question presented is whether the district court violated his “due process right to proof beyond a reasonable doubt” as to the murders when it reimposed the life sentence.
At the threshold, the government asserts, somewhat to our surprise, that “appellate jurisdiction does not exist.” Its stated reason is that a discretionary decision by the sentencing judge declining to depart from the guideline range is not subject to appeal. While the premise is generally sound, United States v. Romolo, 937 F.2d 20, 22 (1st Cir.1991), Lombard has explicitly declined to challenge the refusal to depart; rather, he wants to renew his constitutional challenge to the use of the murders to establish the guideline range for his sentence.
There is nothing outre about the distinction. Lombard is challenging his sentence, and the sentence-as many do-depended on several determinants: here, the armed career criminal statute, various decisions made in applying the guidelines including the finding that Lombard had participated in the murders, and lastly a discretionary decision by the district judge not to depart from the guideline range. That this last decision is unreviewable hardly precludes review of other parts of the equation.
A challenge to the constitutionality of the guidelines as applied is certainly a permitted subject for an appeal, 18 U.S.C. § 3742(a), and presents an issue that we consider de novo. United States v. Carson, 988 F.2d 80, 82 (9th Cir.), cert. denied, 510 U.S. 847, 114 S.Ct. 142, 126 L.Ed.2d 105 (1993). Of course, there might be a law-of-the-case bar to the appeal, although not a jurisdictional one, if this court had fully rejected the constitutional claims in Lombard I. But Lombard I plainly said that constitutional concerns did exist but might be mooted by the remand. 72 F.3d at 184-85.
For reasons we will address in due course, Lombard does not place much weight on the element in this case that would strike non-lawyers as the most troubling: that he has been given a life sentence based on a finding that he committed the two murders of which he was earlier acquitted. Instead, he argues that the district court erred by using a “preponderance of the evidence” standard to determine that he had in fact committed the prior murders and then by using this finding to sentence Lombard as if he had committed the murders.
The framework for federal sentencing is familiar. Departures aside, the guidelines require the district court to calculate the guideline range based not only on the conduct comprising the federal crime of conviction but also on “relevant” albeit “uncharged” conduct-here, the murders-that the sentencing court finds actually occurred in connection with that crime. U.S.S.G. §§ 1B1.3; 2K2.1(c)(2). And ordinarily the facts at sentencing need be proved only by a preponderance of the evidence. McMillan v. Pennsylvania, 477 U.S. 79, 91, 106 S.Ct. 2411, 2418, 91 L.Ed.2d 67 (1986); United States v. Carrozza, 4 F.3d 70, 80-81 (1st Cir.1993), cert. denied, 511 U.S. 1069, 114 S.Ct. 1644, 128 L.Ed.2d 365 (1994).
The use of uncharged conduct at sentencing stems from the longstanding view that the judge should employ all relevant information that helps to decide where, within the broad range usually fixed by statute, this defendant should be sentenced. United States v. Tucker, 404 U.S. 443, 446, 92 S.Ct. 589, 591, 30 L.Ed.2d 592 (1972); 18 U.S.C. § 3577. As for the lower standard of proof, courts sometimes say that “guilt” is the crucial event that alone requires proof beyond a reasonable doubt, and sometimes that more procedural constraints would bog down sentencing. See, e.g., McMillan, 477 U.S. at 92 n. 8, 106 S.Ct. at 2419 n. 8.
In the face of historical practice and judicial precedent, a frontal attack on these practices would be difficult, and Lombard does not attempt it. Rather, quoting “the tail that wags the dog” metaphor in McMillan, 477 U.S. at 88, 106 S.Ct. at 2417, Lombard says that due process requires the use of a “beyond a reasonable doubt” standard in cases where, as here, the finding that the uncharged crime occurred has so severe an effect on the sentence. He adds that in view of the jury's prior acquittal, there must be a reasonable doubt in this case.
This is not an argument that would likely have had much success prior to the guidelines, cf. Patterson v. New York, 432 U.S. 197, 214, 97 S.Ct. 2319, 2329, 53 L.Ed.2d 281 (1977), but the guidelines provide some basis for reconsidering the issue. In the past, uncharged conduct was merely a background fact, like the defendant's criminal record or his habit of kicking his dog, that a judge might consider in making the highly discretionary, and largely ad hoc, decision as to sentence. The guidelines have altered matters in at least one significant respect.
We now have a regime that, aiming to provide equal treatment, requires the sentencing judge to make findings as to relevant uncharged conduct and absent a departure requires the judge to sentence on that basis within a very narrow range. See 18 U.S.C. § 3553(b); U.S.S.G. §§ 1B1.1, 1B1.2, 5C1.1(a). Thus a finding of an uncharged crime at sentencing, and the compulsory fixing of the sentence on that basis, makes the sentencing itself now look somewhat more like a conviction for that uncharged crime-but without the benefit of the criminal standard of proof beyond a reasonable doubt or, for that matter, an indictment or jury trial on that crime.
Most courts have been less, or not at all, troubled by use of the uncharged conduct that has only a limited effect on the sentence or is qualitatively the same crime as the offense of conviction (such as other related drug sales) or both. See, e.g., United States v. Wright, 873 F.2d 437, 441-42 (1st Cir.1989). As the impact and qualitative difference grow, courts become more concerned. The reference in McMillan, 477 U.S. at 88, 106 S.Ct. at 2417, to the risk of the “tail” (sentencing) “wagging the dog” (the substantive offense) has often been taken to suggest that the Supreme Court might endorse some outer limit.2
The guidelines' substantive provisions were, in the main, intended to impose sentencing results very much like those that prevailed in the pre-guidelines era. See Stephen Breyer, The Federal Sentencing Guidelines and the Key Compromises Upon Which They Rest, 17 Hofstra L.Rev. 1, 8 (1988). The pertinent change, as we have noted, is that the guidelines are compulsory. But for Lombard the latter element has been largely removed by our decision in Lombard I, which restored to the district court its pre-guideline discretion to decide whether and how far to give weight to the murders.
True, the district court may (and here did) still choose to give weight to the uncharged offenses in fixing the sentence within the statutory range if it finds by a preponderance of evidence that they occurred; but this was always permitted by longstanding practice and explicit Supreme Court authority. Wisconsin v. Mitchell, 508 U.S. 476, 485, 113 S.Ct. 2194, 2199, 124 L.Ed.2d 436 (1993); Williams v. New York, 337 U.S. 241, 246, 69 S.Ct. 1079, 1082, 93 L.Ed. 1337 (1949). There is no indication that the Supreme Court has altered its position on this issue. If anything, McMillan reinforced that position in upholding a mandatory sentence enhancement based on uncharged conduct.
Some may think that even the status quo ante is at odds with due process and that uncharged conduct should never be considered without criminal-trial safeguards. But the choice then may be, in substance, between turning the sentencing into a new criminal trial or ignoring provable facts that most people think relevant in deciding who deserves more and who less punishment. See Breyer, supra, at 9-12. If the Constitution is now taken to forbid “real offense” sentencing unless criminal-trial procedures are applied, that ruling must come from a higher court.
The only circuit court squarely to impose a higher standard of proof in certain sentencings is the Third Circuit. There, in United States v. Kikumura, 918 F.2d 1084 (3d Cir.1990), the court held that an upward departure due to uncharged conduct should be based on clear and convincing evidence where the finding had an extraordinary impact-there, “a twelve-fold, 330-month departure from the median of an applicable sentencing range.” Id. at 1102. This ruling was premised on a reading of the guidelines informed by due process concerns and has been much discussed but generally not followed. See United States v. Masters, 978 F.2d 281, 286 (7th Cir.1992).
Lombard does not urge this halfway house in the present case. In truth, most judges are unlikely to see a great gulf between a preponderance and “clear and convincing” evidence. Based upon the sentencing transcripts and the trial evidence in this case, the district court would probably find, as to Lombard, that the latter standard had been amply met. By contrast, the “beyond a reasonable doubt” standard is widely regarded as making a substantial difference and, for this very reason, courts have been very cautious in extending it to new realms. Cf. Masters, 978 F.2d at 286-87.
In all events, given Supreme Court precedents, we conclude that the Constitution does not require a heightened proof standard in a case such as ours. Policy is a different matter: on this score, one can argue about imposing greater safeguards for sentencing decisions that severely affect the defendant. But if you asked trial judges, most would be likely to say that what they mainly needed was more latitude and fewer constraints. Cf. Koon v. United States, 518 U.S. 81, ---- - ----, 116 S.Ct. 2035, 2046-47, 135 L.Ed.2d 392 (1996); United States v. Rivera, 994 F.2d 942, 951-52 (1st Cir.1993).
Finally, we think it essential to say a word about the matter that troubled the court in Lombard I but that Lombard has chosen not to stress, at least in his legal argument: the use of acquitted conduct to enlarge his sentence. This certainly accorded with the guidelines. In their present form they draw no distinction between relevant conduct that is uncharged and relevant conduct of which the defendant has actually been acquitted. Absent a departure, all must be given the weight assigned by the guidelines. U.S.S.G. § 1B1.3; United States v. Mocciola, 891 F.2d 13, 16 (1st Cir.1989). The question is why.
The explanation for including acquitted conduct has the usual charm of lawyer's logic. It is said that there is no technical inconsistency between a prior acquittal and the use of the very same acquitted conduct at sentencing to enlarge the sentence, because the jury merely found that the defendant had not been proved guilty “beyond a reasonable doubt”; the sentencing judge, by contrast, finds by a preponderance of the evidence that the acquitted conduct did occur. See, e.g., United States v. Isom, 886 F.2d 736, 738 & n. 3 (4th Cir.1989).
This syllogism has been expressly adopted by the Supreme Court in another context, Dowling v. United States, 493 U.S. 342, 349, 110 S.Ct. 668, 672, 107 L.Ed.2d 708 (1990), and is regularly followed by the lower courts, including ours, in sentencing and elsewhere. See, e.g., Rossetti v. Curran, 80 F.3d 1, 5-6 (1st Cir.1996). Presumably, it is because of these precedents, emphasized in Lombard I, that Lombard has not challenged the syllogism or stressed the fact that the uncharged conduct in this case is also acquitted conduct. As a matter of constitutional law, the syllogism is “rational” enough (as well as binding upon us).
Yet, many judges think that the guidelines are manifestly unwise, as a matter of policy, in requiring the use of acquitted conduct in calculating the guideline range. See United States v. Lanoue, 71 F.3d 966, 984 (1st Cir.1995). A lawyer can explain the distinction logically but, as a matter of public perception and acceptance, the result can often invite disrespect for the sentencing process. This threat is aggravated insofar as the guidelines compel-rather than merely permit-the practice.
Certainly situations exist where the sentencing court might persuasively explain the use of acquitted conduct. For example, a defendant might be acquitted because of reliable evidence suppressed by a Fourth Amendment exclusionary rule or the defendant might later be shown to be guilty by co-conspirator testimony not available at the time of the earlier trial. But the present regime commands that acquitted conduct be taken into account and severely limits the court's ability to disregard it. That a practice is constitutional does not make it wise.
1. Although the sentencing took place in September 1994, the district court applied the November 1990 edition of the guidelines in order to avoid any ex post facto problems. United States v. Prezioso, 989 F.2d 52, 53-54 (1st Cir.1993). All references are to that edition.
2. The reference was in fact directed to a problem rather different than our own, namely, the alleged danger that a state legislature might “tailor [ ]” its substantive crime to shift into the sentencing phase an element that was traditionally part of the crime. Id.
BOUDIN, Circuit Judge.