978 F.2d 972
UNITED STATES of America, Plaintiff-Appellee,
James CAPUTO, Defendant-Appellant.
United States Court of Appeals,
Argued June 3, 1992.
Decided Oct. 28, 1992.
Robert L. Michels (argued), Office of U.S. Atty., Crim. Div., Barry R. Elden, Asst. U.S. Atty., Office of U.S. Atty., Crim. Receiving, Appellate Div., Chicago, for plaintiff-appellee.
Adam Bourgeois (argued), Chicago, for defendant-appellant.
Before CUDAHY, POSNER, and KANNE, Circuit Judges.
POSNER, Circuit Judge.
James Caputo pleaded guilty to federal drug offenses and was sentenced under the federal sentencing guidelines to 120 months in prison. His appeal challenges the sentence on several grounds but only one has sufficient merit to warrant discussion. It concerns the computation of his criminal history. He received six criminal history "points" and as a result was classified in criminal history Category III, putting him in the 97-121 month guideline range. He argues that he should have received only three points, which would have put him into Category II and by doing so have reduced his guideline range to 87-108 months, so that his sentence would have been at least 13 months shorter unless the judge made an upward departure from the guidelines range, a possibility not discussed. The argument was not made in the district court, so we can use it to reverse the judgment only if there was "plain error."
Rule 52(b) of the criminal rules says that "plain errors or defects affecting substantial rights may be noticed although they were not brought to the attention of the court." Were the passage more liberally punctuated, there would be commas around "defects," so that the rule would read "plain errors, or defects, affecting substantial rights may be noticed...." United States v. Young, 470 U.S. 1, 15 n. 12, 105 S.Ct. 1038, 1046 n. 12, 84 L.Ed.2d 1 (1985). Rule 52(a) defines a "harmless error" (which the rule says "shall be disregarded") as one that "does not affect substantial rights." So putting the two subsections together, and adding our clarifying punctuation of (b), we might suppose that Rule 52(b) allowed the correction of errors not brought to the attention of the court that were at once clear and not harmless.
But that is not how the rule is interpreted. A plain error that may be noticed by the court is not an error that merely is plain and not harmless. It is an error that is plain and "likely to have made a difference in the judgment, so that failure to correct it could result in a miscarriage of justice, that is, in the conviction of an innocent person or the imposition of an erroneous sentence." United States v. Newman, 965 F.2d 206, 213 (7th Cir.1992). "[L]ikely to have made a difference in the judgment" is at least a tad stronger than not harmless, and some of our cases draw the distinction more starkly: "we determine whether the alleged error resulted in a miscarriage of justice of such magnitude that the defendant probably would have been acquitted absent the error." United States v. Reiswitz, 941 F.2d 488, 493 (7th Cir.1991); see also United States v. Blythe, 944 F.2d 356, 359 (7th Cir.1991); United States v. Silverstein, 732 F.2d 1338, 1349 (7th Cir.1984). These formulations, the shade of verbal difference between which is not to be taken too seriously, are supported by the Supreme Court's opinion in United States v. Frady, 456 U.S. 152, 163 and n. 13, 102 S.Ct. 1584, 1592 and n. 13, 71 L.Ed.2d 816 (1982), which explains that Rule 52(b) authorizes the correction only of "particularly egregious errors," where "a miscarriage of justice would otherwise result." The error must be such that "the trial judge and prosecutor were derelict in countenancing it." Id.; see also United States v. Young, supra, 470 U.S. at 15, 105 S.Ct. at 1046. The "appellate court [is required] to find that the claimed error not only seriously affected 'substantial rights' "--and thus was not harmless, Fed.R.Crim.P. 52(a)--"but that it had an unfair prejudicial impact on the jury's deliberations." United States v. Young, supra, 470 U.S. at 17 n. 14, 105 S.Ct. at 1045 n. 14. A harmless error is one that could not have affected the jury, a plain error one that not only could have done so but probably did so. As a practical matter, no doubt, the difference is small, United States v. Silverstein, supra, 732 F.2d at 1349, but it is discernible.
Behind the narrow interpretation of the plain-error rule is a recognition that it is inconsistent with the premises of an adversary system, id. at 1349--which is why the civil rules contain no counterpart to Rule 52(b) of the criminal rules and why this circuit has been unwilling to create a common law plain error rule for civil cases, Bogan v. Stroud, 958 F.2d 180, 184 (7th Cir.1992); Maul v. Constan, 928 F.2d 784, 787 (7th Cir.1991), unlike some other circuits. See Polys v. Trans-Colorado Airlines, Inc., 941 F.2d 1404, 1408 (10th Cir.1991), and cases cited there and earlier cases cited in City of Newport v. Fact Concerts, Inc., 453 U.S. 247, 256 n. 13, 101 S.Ct. 2748, 2754 n. 13, 69 L.Ed.2d 616 (1981); see generally Hobson v. Wilson, 737 F.2d 1, 32 n. 96 (D.C.Cir.1984). What could justify the anomaly in the criminal sphere? It is the injustice of allowing the conviction of an innocent person, or an unlawful sentence imposed upon a guilty person, to stand. Such injustices can be avoided, with relatively minor perturbation in the orderly process of justice, despite the defendant's failure to have brought the error that precipitated his conviction or sentence to the trial court's attention, when the appellate court can find and correct the error without any entanglement in contested or unknown facts, or in obscure or contestable rules of law. The benefit of departing from the ordinary processes of adversary justice is maximized when the departure is necessary to save an innocent person. The cost is minimized when the error can be picked out with relatively little difficulty.
Both the existence of injustice and the ease of its correction depend on certainty that error was committed. An error is not plain in the sense of egregious if it isn't even clear that it was an error, and if that isn't clear the appellate court is likely to encounter difficulty in determining that an error occurred at all. If it is uncertain whether the trial court committed an error, it will be difficult to say that there is a substantial danger that an innocent man was convicted, unless the appellate court invests substantial resources in determining whether there was error.
We should note an exception to the formula that we are expounding: if the error consists of a certainly erroneous failure to consider a factual question that if resolved in the defendant's favor would require his acquittal, the plain-error rule is applicable even though a determination of whether there was in fact a miscarriage of justice must await the additional factfinding in the trial court. With this exception (which does not entail any entanglement by the appellate court in the factual issue), the certainty of the error's actually being error, and, what is closely related, whether the error is straightforward in the sense that the appellate court can readily determine that it is an error, are, along with severe prejudice to the defendant, essential elements of a proper reversal under Rule 52(b). But we do not think, as the government argued, that the error must be plain in the further sense that it leaps out at the reader of the district court's decision--that it is obvious in the sense of being a tyro's error, an error that only a first-year law student would have made, a ludicrous error, an error that condemns the lawyer who failed to bring it to the judge's attention of professional incompetence and the judge of judicial incompetence for having failed to notice it. If the plain-error doctrine were so confined it could almost never be invoked successfully, if only because it would be virtually coextensive with the doctrine of ineffective assistance of counsel. So, while the error must be straightforward, it can be so in hindsight. United States v. Silverstein, supra, 732 F.2d at 1349; cf. United States v. McCall, 915 F.2d 811, 814 (2d Cir.1990). But it must be sufficiently certain and sufficiently prejudicial that the trial judge and prosecutor were derelict in countenancing it. United States v. Frady, supra, 456 U.S. at 163, 102 S.Ct. at 1592. It cannot be subtle, arcane, debatable, or factually complicated. It must be--plain; but it needn't be blatant.
There can be plain error in sentencing as well as in convicting. United States v. Newman, supra, 965 F.2d at 213. To impose an unlawful sentence is a miscarriage of justice, and often the determination of the lawfulness of a sentence does not depend on a factual determination. Does this mean that a criminal defendant can raise all his purely legal objections to a sentence for the first time on appeal, on the theory that any error in the length of the sentence brings about a miscarriage of justice? No, because subtle, obscure, debatable errors are (to repeat) excluded. And there is a further control, which lies in the resolution of a double issue that we have not yet discussed: Can only plain errors be noticed for the first time on appeal? And must plain errors be noticed for the first time on appeal? The answer to the first question is a qualified yes, and to the second a qualified no.
Although Rule 52(b) does not say in so many words that a court may not notice an error unless it is plain, the negative implication--that it may not, in general, notice a nonplain error even if it is prejudicial (that is, not harmless and hence not within the bar of Rule 52(a))--seems inescapable. The interest, which courts stress, in bringing criminal proceedings to a speedy conclusion, Flanagan v. United States, 465 U.S. 259, 264-66, 104 S.Ct. 1051, 1054-55, 79 L.Ed.2d 288 (1984); Illinois v. F.E. Moran, Inc., 740 F.2d 533, 536 (7th Cir.1984), would be impaired if appellate judges felt entirely free to reverse criminal judgments on grounds that had not been raised in the trial court. But the qualification "in general" is important. There are other grounds besides plain error on which courts forgive waivers in criminal as in other cases. One is where the appellee has waived waiver. Taylor v. Gilmore, 954 F.2d 441, 450 n. 5 (7th Cir.1992), cert. granted, --- U.S. ----, 113 S.Ct. 52, 121 L.Ed.2d 22 (1992); United States v. Leichtnam, 948 F.2d 370, 375 (7th Cir.1991); Fagan v. Washington, 942 F.2d 1155, 1157 (7th Cir.1991). Another is where the appellant could not have presented the issue to the district court, as in a case in which the defendant's appellate counsel argues that trial counsel rendered ineffective assistance--a ground that trial counsel himself was hardly likely to have raised in the district court. United States v. Castillo, 965 F.2d 238, 243 (7th Cir.1992). Jurisdictional defects can be noticed at any stage in a proceeding whether or not raised in the trial court, Fed.R.Civ.P. 12(h)(3), and, likewise, delicate issues involving federalism or comity. E.g., Younger v. Harris, 401 U.S. 37, 41, 91 S.Ct. 746, 749, 27 L.Ed.2d 669 (1971); Fortino v. Quasar Co., 950 F.2d 389, 391 (7th Cir.1991). We need not attempt an exhaustive catalog. See Peretz v. United States, --- U.S. ----, ----, 111 S.Ct. 2661, 2679, 115 L.Ed.2d 808 (1991) (dissenting opinion); Thomas v. Indiana, 910 F.2d 1413, 1415 (7th Cir.1990).
If an error is plain, must the appellate court notice it and reverse the district court? We think not. Authority is sparse but what there is supports this conclusion. Phillips v. Lane, 787 F.2d 208, 212 (7th Cir.1986), and cases cited there. Rule 52(b) uses "may," not "shall," but that semantic point is not decisive; "may" in statutes and rules is often interpreted in a mandatory rather than permissive sense. Sonicraft v. NLRB, 814 F.2d 385, 386 (7th Cir.1987). The principal reason we think the word permissive in Rule 52(b) is a practical one: if read as mandatory it would require the appellate court to look for errors that hadn't been brought even to its attention, let alone to the trial court's. Or errors brought to its attention after the case had been argued, or perhaps after it had been decided but before the decision had become final. And we suppose too that if the defendant sought to bring to the appellate court's attention for the first time a provision of the sentencing guidelines that had not been mentioned in the trial court and that would at most have shortened the defendant's sentence by a few weeks the appellate court could refuse to reverse notwithstanding the technical miscarriage of justice.
This case is not of that type. The issue the defendant seeks to raise for the first time on appeal would, if resolved in his favor, shorten his sentence by more than a year. And at least as presented by the defendant (the significance of this qualification will appear), the issue is one of law. But the error--if there was one--is far from straightforward, and therefore could not be used to reverse the judgment and to order Caputo resentenced. Actually there was no error. Rule 52(b) does not prevent us from pointing that out; it prevents us only from reversing on the basis of an error that does not come within the scope of the rule.
Of the six criminal-history points assigned to Caputo, three are a consequence of two convictions for crimes that he committed while on probation and that he claims fall within exclusions to the criminal-history provisions of the guidelines. One was a one-year sentence of "conditional discharge" for resisting a peace officer, the other a sentence for using a false driver's license. The guidelines exclude "hindering or failure to obey a police officer," and that would seem to knock out the one-year sentence but for an exception to the exclusion for "a term of probation of at least one year." U.S.S.G. § 4A1.2(c)(1). Caputo didn't receive probation, he received "conditional discharge," and the question is whether these are the same thing, as the government argues and previous cases dealing with conditional-discharge sentences hold, Pedigo v. Commonwealth, 644 S.W.2d 355, 357-58 (Ky.App.1983), or assume, United States v. Osborne, 931 F.2d 1139, 1160 (7th Cir.1991). Probation means that the convicted defendant is not incarcerated but must comply with various conditions set by the sentencing court and monitored by a probation officer, such as periodic reporting, keeping out of trouble, and keeping clean of drugs. Conditional discharge is the same except that there is no probation officer, and in fact was introduced into Illinois law (we conjecture, with some support in Council Commentary, Smith-Hurd Ill.Stat.Ann. ch. 38, p 1005-1-4, and Jay M. Hanson, "The Conditional Discharge--A New View in Misdemeanor Sentencing," 62 Ill.Bar J. 280 (1974)) as a way of coping with the shortage of probation officers. It is probation without the probation officer and that is a distinction without a difference so far as the purposes of the guideline exception is concerned. Were Caputo right, then even though hindering a police officer would count in a defendant's criminal history if the defendant had received a year's probation, it would count for nothing in that history if he received five years of identical restrictions only called conditional discharge because the state had a shortage of probation officers. Such a difference in treatment would make no sense. Nor can it be assumed that in deciding between probation and conditional discharge Illinois state judges always assign the more serious or the more dangerous offender to the first. They may assign a less serious, less dangerous offender to probation because they think he would benefit more from continuous supervision than a somewhat worse offender, or a more serious one to conditional discharge because the assignment of a probation officer would be impractical or unnecessary for one reason or another. Lawrence X. Pusateri & Robert Kent Scott, "Illinois' New Unified Code of Corrections," 61 Ill.Bar J. 62, 70, 72 (1972).
At argument the question arose whether a person on conditional discharge would ever be returned to custody, without a probation officer to monitor his activities; if not, then in practice conditional supervision would be much the lighter penalty and perhaps should be so treated under the guidelines. We had supposed that the usual way in which a probationer is returned to custody is that he commits a crime while on probation, and when he is arrested his probationary status is discovered, which would be equally true of conditional discharge. But we seem to have been mistaken. According to Administrative Office of the Illinois Courts, 1989 Annual Report to the Supreme Court of Illinois 237, more than half the probationers convicted for violating probation had committed violations of the terms of probation rather than new offenses, and presumably that high fraction results from the monitoring of probationers by probation officers, which has no counterpart in conditional discharge. So probation may in fact be a more severe penalty, but Caputo made nothing of this until oral argument, which was too late.
He did argue that the Sentencing Commission knows the different types of punishment and if it had meant "probation or conditional discharge" it would have said so. Even the Sentencing Commission is not omniscient, however, as witness the many appellate court decisions that struggle with the meaning of particular provisions and the many amendments that the Commission has made in its original and successive versions of the guidelines. Only five states have conditional-discharge procedures similar to Illinois. (The others are Connecticut, Kentucky, New Hampshire, and New York.) The Commission could easily have overlooked a relatively esoteric state procedure. In a provision of the guidelines not in issue in this case the Commission did mention "unsupervised probation," a punishment method, found in a number of states, that is functionally equivalent to conditional discharge--and in that provision it equated unsupervised to supervised probation. Application Note 4 to U.S.S.G. § 4A1.1; United States v. Bailey, 955 F.2d 28 (8th Cir.1992). Since unsupervised probation equals conditional discharge, we think the latter should be equated to probation.
An offense is excludable from one's criminal history if it is "similar to" a list of offenses that are expressly excludable, among which are "driving without a license or with a revoked or suspended license" and giving "false information to a police officer." U.S.S.G. § 4A1.2(c)(1). While there is a resemblance, using a false driver's license is categorically more serious and should not be ignored in computing a defendant's criminal history as the listed offenses are. This is especially clear with respect to the "similar" offense of driving without a valid license. The old distinction between misfeasance and nonfeasance is not entirely arbitrary. The driver who fails to obtain a driver's license, or who continues driving after his license has been revoked or suspended or has expired, is guilty of failing to expend resources (albeit modest ones) of time and money that the state requires him to expend as a condition of being permitted to drive. His is a wrongful inactivity, but often it is the wrongfulness of irresponsibility rather than of calculation. The driver who expends resources to obtain forged or otherwise fraudulent documentation to enable him to drive crosses the line from inactivity to activity and by doing so reveals himself to be a person willing to incur expense to commit a crime, presumably in anticipation of compensating profit. This willingness suggests a more calculating, a more resourceful, and a more dangerous criminal. In some cases, including the present one, he will be more dangerous for an additional reason; he will be more difficult to apprehend. A person who has no documentation will be apprehended the first time he is stopped by a policeman and a license check run on him. Not so a person who, as we understand happened here, obtains a license under a false name from the state, so that the name on the license corresponds to the name in the state's records. The intermediate case, driving with someone else's license plates, was assimilated to driving without a license in United States v. Mitchell, 941 F.2d 690 (8th Cir.1991)--but all the court said was that the district judge's determination that these were similar offenses was not clearly erroneous.
The resemblance between what Caputo did and giving a police officer false oral information is closer, but there is still a difference. The giving of false information, as where a person arrested for speeding claims not to have known the speed limit or to have a malfunctioning speedometer, is often spontaneous; and the human proclivity for self-exculpation, which is what usually triggers a tender of false information to a police officer, is well-nigh universal. The element of calculation is greater when an individual goes to the bother of obtaining a false document in advance of any confrontation with an officer.
We find no errors, plain or otherwise, and the judgment is therefore