United States v. Andre C. Joseph

POSNER, Chief Judge. André Joseph pleaded guilty to bank robbery and was sentenced to 95 months in prison. The appeal argues that two of his previous sentences should not have been counted separately; if this is right, the sentence he received for the bank robbery was too long. The federal sentencing guidelines require, for purposes of calculating a defendant’s criminal history, that previous sentences be counted as separate sentences if they are imposed in “unrelated eases,” but as one sentence (the sum of the separate sentences if they are consecutive, the longest if they *402 are concurrent) if imposed in “related cases.” U.S.S.G. § 4A1.2(a)(2). A note states that “prior sentences are considered related if they resulted from offenses that ... were consolidated for trial or sentencing,” unless the offenses “were separated by an intervening arrest.” Application Note 3. The effect of counting two sentences as two rather than as one is, of course, to lengthen the defendant’s criminal record. The two sentences that the district judge counted separately were imposed in the following circumstances. On November 23, 1989, Joseph, who was in the Clinton County, Pennsylvania jail, helped another prisoner escape. Even though Joseph was in jail, a warrant for his arrest was issued; but it is unclear whether he was ever arrested. On December 6, 1989, a criminal complaint was filed that charged him with conspiracy, to which he later pleaded guilty. On July 17, 1990, which was after the charge of conspiracy had been lodged against him, he escaped from custody while being transported from the jail (we do not know to where) by Clinton County sheriffs deputies. He hid in a private garage but was recaptured immediately and charged the same day with criminal trespass. Again an arrest warrant was issued, and again we do not know whether it was executed, and again he eventually pleaded guilty. Joseph was sentenced for both crimes — helping the other prisoner escape, and his own escape — on the same day, November 7, 1990, and he was given prison sentences of slightly different lengths. The record does not disclose whether they were to be served concurrently or consecutively. There was no order consolidating the cases for sentencing. The district judge at the federal sentencing hearing refused to decide whether the two Pennsylvania sentences, the one for conspiracy and the one for trespass, had been consolidated and if so whether they were separated by an arrest, that is, whether Joseph had been arrested for the conspiracy before he committed the trespass. The district judge thought that these determinations were unnecessary because the two offenses were unrelated and since they were unrelated the sentences for them had to be counted separately. This was error. “Related sentences” is a defined term, defined in an application note to be sure but such notes are authoritative when, as in this case, they explain a guideline. United States v. Hill, 48 F.3d 228, 231 (7th Cir.1995). Sentences in two cases are “related” when the eases were consolidated for trial or sentencing, unless there was an intervening arrest. It doesn’t matter how unrelated the crimes really are. One could be for unauthorized use of the slogan “Smokey the Bear,” the other for having laundered drug money five years earlier. If the sentences were consolidated, they are to be treated as a single sentence for purposes of calculating the defendant’s criminal history. The district judge’s error was not consequential if the two Pennsylvania cases were not consolidated for sentencing or there was an intervening arrest. The second issue is analytically simpler. We are not disposed to accept the government’s invitation to treat the filing of a criminal complaint as the equivalent of an arrest, just because “arresting” a person already in custody is a pointless act. Some criminal proceedings begin with (or include) an arrest; others do not. There certainly is no legal requirement that a criminal defendant be arrested. Lem Woon v. Oregon, 229 U.S. 586, 590, 33 S.Ct. 783, 784-85, 57 L.Ed. 1340 (1913). Some defendants appear voluntarily to be arraigned, without having been arrested. Some are tried in absentia, sometimes having successfully eluded capture from the start. There is no requirement that they have ever been in custody, only that they were informed of the possible consequences of failing to appear for trial. Commonwealth v. Sullens, 533 Pa. 99, 619 A.2d 1349, 1353 (1992); People v. Badia-Almonte, 188 A.D.2d 1001, 592 N.Y.S.2d 157 (1992); People v. Watson, 109 Ill.App.3d 880, 65 Ill.Dec. 360, 441 N.E.2d 152 (1982). For some minor crimes, the arrest of the defendant is actually prohibited. See, e.g., People v. Bland, 884 P.2d 312, 315 (Colo.1994); State v. Sassen, 240 Neb. 773, 484 N.W.2d 469, 471-72 (1992). When federal prisoners are placed in segregation pending the determination of disciplinary charges against them, this is not considered “arrest.” Unit *403 ed States v. Mills, 810 F.2d 907, 909 (9th Cir.1987); United States v. Smith, 464 F.2d 194, 196 (10th Cir.1972). The Sentencing Commission could not have been unaware of all this. It is not our place to rewrite the guidelines (and we must treat the application note as a guideline). There is play in their interpretive joints, but to interpret “arrest” to mean “arrest or charge” is a little too adventurous for our taste in the absence of a compelling reason, which has not been offered, for believing that the choice of words was a slip of the pen. Still, there was an arrest warrant. Were the issue of intervening arrest critical, the government could make redoubled efforts to discover whether Joseph was “arrested” after the first offense in some meaningful sense of the word. It is not critical, as we are about to see. With regard to whether the two Pennsylvania cases were consolidated for sentencing, the government intimates that, the sine qua non of consolidated sentencing is a formal order of consolidation. One ease so holds. United States v. Elwell, 984 F.2d 1289, 1296 n. 7 (1st Cir.1993). One leaves the issue open. United States v. Davis, 929 F.2d 930, 935 n. 3 (3d Cir.1991). A couple of cases seem to assume that it is required. United States v. Klein, 13 F.3d 1182, 1185 (8th Cir.1994); United States v. McComber, 996 F.2d 946, 947 (8th Cir.1993). But most, including the cases in this court, treat the absence of a formal order as merely probative and not conclusive. E.g., United States v. Russell, 2 F.3d 200, 204 (7th Cir.1993); United States v. Coleman, 964 F.2d 564, 566-67 (6th Cir.1992); United States v. Garcia, 962 F.2d 479, 482-83 (5th Cir.1992). The government has not explained to us why everything must depend on the presence of an order. Consider the following two pairs of cases. In both pairs a defendant is given concurrent sentences of three years on the same day for crimes X and Y committed six months apart, but in one pair there is a formal order consolidating the cases and in the other there is not. On the government’s view, in computing the criminal history of the first defendant the district judge is to count one sentence of three years but in computing the criminal history of the second defendant he is to count two sentences each of three years. This can make a big difference in the length of the defendant’s federal sentence. Is it a senseless difference? To answer that question, one would like to know why the two sentences had been consolidated in one case but not the other. Perhaps the hypothetical is unrealistic. Perhaps if the pairs of offenses are identical, both would be consolidated or neither. We do not know. The practice of consolidation may differentiate between pairs (or other multiples) of cases for reasons unrelated to the character of the defendant’s criminal activity. The government, while emphatic that there was no consolidation here, offers no description or rationale of the practice and we have not found any on our own. The sentencing guideline at issue either endorses arbitrary differences in sentencing for the sake of administrative convenience (and so without an order the defendant loses), which we do not understand the government to be arguing even though it does seem to think that the presence of an order is both a necessary and a sufficient condition for treating two sentences as one; or it assumes that a decision to consolidate two cases for sentencing, with or without a formal order, reflects a judgment that the cases are (in contrast to our hypothetical pairs of cases) so far related that to sentence the defendant for them separately would create an exaggerated sense of the defendant’s criminality. If administrative convenience is the rationale, then Joseph must lose, because there was no order of consolidation. If the second, the penological, rationale is the true one, then Joseph has a stronger case but still is wrong to argue that the fact that a defendant is sentenced for two or more crimes on the same day means that the cases must be deemed consolidated, in the relevant sense of consolidation, for sentencing. E.g., United States v. Woods, 976 F.2d 1096, 1102 n. 6 (7th Cir.1992); United States v. Alberty, 40 F.3d 1132, 1135 (10th Cir.1994); United States v. McAdams, 25 F.3d 370, 374 (6th Cir.1994); United States v. Klein, supra, 13 F.3d at 1185; United States v. Lopez, 961 F.2d 384, 387 (2d Cir.1992). The same-day sentencing *404 could be a coincidence; or a convenience to the defendant or to prison authorities. It need not imply a judgment that the defendant is not so wicked or dangerous as his having been charged with separate crimes in separate eases might suggest. This is true even if he receives concurrent sentences, e.g., United States v. Sexton, 2 F.3d 218, 219 (7th Cir.1993) (per curiam); United States v. Klein, supra, 13 F.3d at 1185; United States v. Garcia, supra, 962 F.2d at 482-83, since the length of each of the concurrent sentences may have been augmented because of a belief that these really were separate rather than related offenses. United States v. Boulahanis, 677 F.2d 586, 591 (7th Cir.1982). Concurrent sentences, moreover, need not be and often are not imposed in the same proceeding. Courts have criticized Application Note 3 for overlooking possibilities of unrelatedness in cases in which an order of consolidation is entered, since that too might be done for administrative convenience rather than to signify that the defendant deserved a lighter punishment because his previous crimes had been related ones. E.g., United States v. Woods, supra, 976 F.2d at 1100-01; United States v. Butler, 970 F.2d 1017, 1029-30 (2d Cir.1992). Yet these decisions assume that if an order of consolidation is entered, the issue is concluded; the defendant is entitled to the benefit of Application Note 3; and so of course the note strongly implies. Even so— even if as we agree the court is not to look behind an order of consolidation — it does not follow that the absence of such an order should be deemed conclusive against finding consolidation, at least in the absence of any indication of the criteria used by the sentencing jurisdiction in deciding whether to consolidate cases pending at the same time against the same defendant. Indeed, if as appears a formal order of consolidation is sometimes entered for reasons unrelated to the relatedness of the defendant’s offenses, this makes it all the more likely that a formal order of consolidation is sometimes withheld even though the offenses are in fact closely related. This would be functional consolidation, and while it is arguable that the Sentencing Commission did not want the courts to make so qualitative a judgment as is entailed in finding two sentences functionally consolidated, we have held, to the contrary, that functional consolidation satisfies Application Note 3. United States v. Russell, supra, 2 F.3d at 204; see also United States v. Chapnick, 963 F.2d 224, 228-29 (9th Cir.1992); cf. United States v. Villarreal, 960 F.2d 117, 119-21 (10th Cir.1992). We are not asked to reexamine that holding. The district judge did not determine whether the two cases against Joseph had been consolidated in the functional sense for sentencing. He did however find that the two crimes were unrelated, which seems plain from the facts — they were different crimes committed months apart, the interval alone making it unlikely that they would be “consolidated” for sentencing, though there might be a single sentencing hearing for the sake of administrative convenience — concretely, for the sake of not having to haul Joseph and the prosecutors before the court on two separate occasions. We do not see . how the criteria for what we are calling functional consolidation could be thought satisfied. There was no order of consolidation, and there is no reason to think that the judge who sentenced Joseph treated the two crimes as one in any sense. No purpose would be served by a remand for resentencing in these circumstances, and this regardless of how the tangled issue of the standard of appellate review of a finding of consolidation is resolved. That issue, left open in United States v. Russell, supra, 2 F.3d at 204; cf. United States v. Ali, 951 F.2d 827, 828 (7th Cir.1992); United States v. Woods, supra, 976 F.2d at 1099; United States v. Garcia, supra, 962 F.2d at 481, has divided the other circuits. For example, compare United States v. Lopez, supra, 961 F.2d at 386, with United States v. Mau, 958 F.2d 234, 236 (8th Cir.1992). We need not try to settle the issue here. Whatever the standard of review applicable to this ease, Joseph cannot show that his sentences were consolidated. The judge’s error was harmless, and the judgment is therefore AFFIRMED.