United States v. Robert F. Carrozza, United States of America v. Raymond J. Patriarca, United States of America v. Raymond J. Patriarca

LEVIN H. CAMPBELL, Senior Circuit Judge. Raymond J. Patriarca pled guilty to one count of conspiring to violate the Racketeer Influenced and Corrupt Organizations Act (“RICO”), 18 U.S.C. § 1962(d), one count of violating RICO, 18 U.S.C. § 1962(e), four counts of interstate travel in aid of racketeering, 18 U.S.C. § 1952 (the “Travel Act”), and one count of conspiring to violate the Travel Act. He was sentenced by the United States District Court for the District of Massachusetts to a prison term of 97 months, three years of supervised release, a $50,000 fine, $122,344 costs of incarceration, and $3,954 costs of supervision. Pursuant to 18 U.S.C. § 3742(b), the government appeals from the district court’s determination that the relevant conduct for sentencing purposes in this RICO ease is limited to just the predicate Travel Act violations charged against Patr-iarca and conduct relating directly to those charged predicates. Patriarca appeals from the district court’s upward departure under U.S.S.G. § 4A1.3 and from the district court’s imposition of the costs of incarceration and supervision under U.S.S.G. § 5E1.2®. Robert F. Carrozza appeals from a 228-month sentence imposed by the district court after Carrozza pleaded guilty to 49 counts of racketeering-related offenses. Carrozza argues that the district court’s decision to “assume” that his base offense level should be adjusted upwards for his role in the offense constituted plain error. I. Patriarca’s Sentence A. Background Count One of the indictment charged Patr-iarca and seven codefendants with participation in a criminal conspiracy to violate the RICO statute. Count Two charged the same defendants with a substantive violation of the RICO statute. The remaining 63 counts charged related racketeering acts involving different defendants, including — in Count 30 — a conspiracy to violate the Travel Act. The RICO charges alleged that the Patr-iarca Family had committed illegal activities over a period of 14 years. They identified the defendants as members of a nationwide criminal organization known as La Cosa Nos-tra, and described Patriarca’s role, after July 1984, as the boss and ultimate supervisor of the Patriarca Family. The RICO counts alleged that the Patriarca Family, named as the RICO enterprise, acted in conformity with the rules of La Cosa Nostra, including the requirement that members commit murder at the direction of their superiors. It was further alleged that members of the Patriarca Family were required to obey their superiors and commit criminal acts at their direction, including murder. Members of the Patriarca Family were allegedly required to share their illegal profits with their superiors. *73 The indictment alleged that the Patriarca Family was in the business of extortion, narcotics trafficking, loansharking, gambling, and murder. The indictment charged the commission of a total of 68 separate, predicate acts, most of them by defendants other than Patriarca. The predicate racketeering acts in which Patriarca was personally named were five violations of (and conspiracy to violate) the Travel Act, four of which were also charged as substantive violations against Patriarca in Counts 81, 36, 38 and 39. Prior to Patriarca’s entry of a guilty plea, the government informed the court and Patr-iarca that it would seek to include specific acts of relevant conduct, pursuant to U.S.S.G. § IB 1.3, in determining Patriarca’s base offense' level, and would further seek upward departures pursuant to U.S.S.G. § 4A1.3 and § 5K2.0. As an example of relevant conduct, the government then cited Patriarca’s involvement in the narcotics trafficliing of Patriarca’s associate, Salvatore Michael Ca-ruana. As an example of conduct justifying an upward departure, the government cited the murder of Vincent James Limoli, which was charged against one of Patriarca’s code-fendants. On December 3,1991, Patriarca pled guilty without having entered into any agreement with the government. In the sentencing proceedings that ensued, the government asked the court to consider seven instances of relevant conduct, along with the charged c5n-duct, in determining Patriarca’s base offense level for his RICO offenses. These instances were (1) Patriarca’s involvement in the drug trafficking of Caruana; (2) Patriarea’s efforts to harbor Caruana as a fugitive; (3) the murder of Limoli; (4) the murder of Theodore Berns, which was committed by Carua-na purportedly because Berns was involved with Caruana’s wife; (5) the narcotics activities charged against codefendant Robert Carrozza; (6) Patriarea’s alleged authorization of an attempt to murder Vincent Fer-rara; and (7) the harboring of La Cosa Nos-tra member, Alphonse Pérsico, while he was a fugitive from justice. Of these acts, only the Limoli murder and Carrozza’s drug trafficking had been mentioned in the indictment, these two acts having been charged as predicate acts against Patriarca’s codefend-ants (not Patriarca himself). The government acknowledges that Patriarca had direct personal involvement only in the Caruana drug trafficking and the harboring of Carua-na as a fugitive. But it also argues that all seven activities were reasonably foreseeable to Patriarca and were committed during, and in furtherance of, the RICO conspiracy after Patriarca had joined it as its chief. The government asserted that holding Patriarca responsible for the Limoli or the Berns murder would increase his base offense level to 43, but that this level should then be reduced by three levels because Patriarca’s role was minimal or minor under U.S.S.G. § 3B1.2. The guideline range for an offense level of 40 and Criminal History Category I is 292-365 months in prison. The government recommended a sentence of 292 months. After numerous evidentiary hearings, the district court announced its decision to sentence Patriarca to 97 months imprisonment. This was an upward departure from the court’s calculated guideline range of 63 to 78 months. 1 The court concluded that relevant, conduct in a RICO case was, as a matter of law, limited to the specific predicate acts charged' against the defendant (here, as to Patriarca, the Travel Act violations) and conduct relating to the charged predicates. The court observed that the base offense level for RICO is the greater of 19 or “the offense level applicable to the underlying racketeering activity.” U.S.S.G. § 2El.l(a). Because § 2E1.1 specifies more than one base offense level, the court determined that § 1B1.3 requires the proper base offense to be ascertained by the inclusion of relevant conduct. The core question, in the court’s view, was whether “underlying racketeering activity” within the meaning of § 2E1.1(a)(2) referred only to the predicate racketeering acts charged against Patriarca himself, or whether it also embraced other racketeering acts *74 including those of Patriarca’s RICO cocon-spirators committed in the course of the RICO conspiracy. In opting for the former construction, the court relied upon three principles it felt were key: (1) the guidelines are primarily a “charge offense” system; (2) the guidelines are generally intended to duplicate nationwide past practices; and (3) the guidelines are intended to establish a sentencing system which is both administratively workable and fair. Regarding the first, the court noted that none of the seven instances of conduct cited by the government had been charged against Patriarca personally in the indictment. As to the second reason, the court noted that there are no reported pre-guide-line RICO cases in which a defendant was sentenced and punished for an uncharged murder. With respect to the third, the court stated that the government’s position was administratively unwieldy: weeks or months of evidentiary hearings could be required to decide if a defendant committed the uncharged relevant conduct. Finally, the court was concerned about the procedural fairness of punishing a defendant for an uncharged murder without indictment, trial by jury, and proof beyond a reasonable doubt. The court reasoned that adoption of the government’s position would raise serious constitutional questions which the district court’s interpretation would avoid. Treating the Limoli or Berns murder as relevant conduct would, the court believed, have the effect of raising the maximum penalty for the RICO violations from 20 years to life imprisonment. The RICO penalty provision, 18 U.S.C. § 1963(a), provides for a maximum sentence of 20 years unless “the RICO violation is based on racketeering activity for which the maximum penalty includes life imprisonment,” in which case the maximum sentence is life. The guideline penalty for murder, which is a level 43 offense, is life imprisonment. Because the district court, unlike the government, thought a three-level reduction for a minor or minimal role in the offense was unlikely, the court reasoned that if Patriarca was held responsible for the Li-moli or Berns murders, his probable sentence would be life imprisonment. In the court’s view, therefore, treating the murders as relevant conduct (thereby increasing the guideline range from 63-78 months to life) would violate the due process clause of the constitution by permitting the relevant conduct determination “to be a tail which wags the dog of the substantive offense.” McMillan v. Pennsylvania, 477 U.S. 79, 88, 106 S.Ct. 2411, 2417, 91 L.Ed.2d 67 (1986). The government appeals from this determination. B. The Government’s Appeal: Relevant Conduct “The legal determination as to the proper interplay among related guidelines is subject to plenary review.” United States v. Schultz, 970 F.2d 960, 962 (1st Cir.1992), cert. denied, — U.S. -, 113 S.Ct. 1020, 122 L.Ed.2d 167 (1993). Therefore, we review de novo the district court’s application of the relevant conduct guideline, U.S.S.G. § 1B1.3, to the RICO guideline, U.S.S.G. § 2E1.1. We conclude that the district court erred when it limited relevant conduct to conduct in furtherance of the predicate acts charged against Patriarca. We hold that relevant conduct in a RICO case includes all conduct reasonably foreseeable to the particular defendant in furtherance of the RICO enterprise to which he belongs. We agree with the government that the language of the relevant conduct section, § 1B1.3 2 , and its application to the RICO *75 guideline, § 2E1.1, are clear, and hence'must be applied. Section 1B1.3 calls for determining the following guideline elements on the basis of relevant conduct as defined: (1) the base offense level, where the guideline specifies more than one base offense level, (2) specific offense characteristics, (3) cross references in Chapter Two, and (4) adjustments in Chapter 3. The RICO guideline, § 2E1.1 3 , specifies more than one base offense level, including a cross reference to “the offense level applicable to the underlying racketeering activity.” See United States v. Masters, 978 F.2d 281, 284 (7th Cir.1992) (reference to “underlying racketeering activity” in § 2E1.1(a)(2) is a cross reference), cert. denied, — U.S. -, 113 S.Ct. 2333, 124 L.Ed.2d 245 (1993); U.S.S.G. § 1B1.5, application note 1 (cross references may be general, such as “to the guideline for the ‘underlying offense’”). Therefore, § 1B1.3 requires the base offense level for § 2E1.1 to be determined on the basis of relevant conduct as that term is described in § lB1.3(a)(l). Section 1B1.3 states that “in the case of a jointly undertaken criminal activity,” relevant conduct includes “all reasonably foreseeable acts and omissions of others in furtherance of the jointly undertaken' criminal activity.” U.S.S.G. § lB1.3(a)(l)(B). This is often referred to as the accomplice attribution element of relevant conduct. “Jointly undertaken criminal activity” is defined in § lB1.3(a)(l)(B) as “a criminal plan, scheme, endeavor, or enterprise undertaken by the defendant in concert with others, whether or not charged as a conspiracy.” Id. (emphasis added). Here, the RICO enterprise — the Patriarca Family — was a “jointly undertaken criminal activity.” Thus, Patriarca is potentially liable for the foreseeable criminal acts of others in furtherance of that enterprise even though he did not personally participate in them. The application notes expand on the role of relevant conduct in the case of criminal activity undertaken in concert with others. We quote from application note 2 to § 1B1.3 at length because of the guidance it provides to courts in determining when a defendant is responsible for the conduct of others under the accomplice attribution element of the relevant conduct guideline: In the case of a jointly undertaken criminal activity, subsection (a)(1)(B) provides that a defendant is accountable for the conduct (acts and omissions) of others that was both: (i) in furtherance of the jointly undertaken criminal activity; and (ii) reasonably foreseeable in connection with that criminal activity. Because a count may be worded broadly and include the conduct of many participants over a period of time, the scope of the criminal activity jointly undertaken by the defendant (the “jointly undertaken criminal activity”) is not necessarily the same as the scope of the entire conspiracy, and hence relevant conduct is not necessarily the same for every participant. In order to determine the defendant’s accountability for the conduct of others under subsection (a)(1)(B), the court must first determine the scope of the criminal activity the particular defendant agreed to jointly undertake (i.e., the scope of the specific conduct and objectives embraced by the defendant’s agreement). The conduct of others that was both in furtherance of, and reasonably foreseeable in connection with, the criminal activity jointly un *76 dertaken by the defendant is relevant conduct under this provision. The conduct of others that was not in furtherance of the criminal activity jointly undertaken by the defendant, or was not reasonably foreseeable in connection with that criminal activity, is not relevant conduct under this provision. In determining the scope of the criminal activity that the particular defendant agreed to jointly undertake (i.e., the scope of the specific conduct and objectives embraced by the defendant’s agreement), the court may consider any explicit agreement or implicit agreement fairly inferred from the conduct of the defendant and others. Note that the criminal activity that the defendant agreed to jointly undertake, and the reasonably foreseeable conduct of others in furtherance of that criminal activity, are not necessarily identical. For example, two defendants agree to commit a robbery and, during the course of that robbery, the first defendant assaults and injures a victim. The second defendant is accountable for the assault and injury to the victim (even if the second defendant had not agreed to the assault and had cautioned the first defendant to be careful not to hurt anyone) because the assaultive conduct was in furtherance of the jointly undertaken criminal activity (the robbery) and was reasonably foreseeable in connection with that criminal activity (given the nature of the offense). U.S.S.G. 1B1.3, application note 2. Application note 2 reflects recognition that the accomplice .attribution provision of § 1B1.3 operates to hold a defendant responsible for the conduct of others even though “a count may be worded broadly and include the conduct of many participants over a period of time.” So as to keep the criminal responsibility within bounds, § 1B1.3 requires sentencing courts to ascertain on an individual basis the scope of the criminal activity that the particular defendant agreed jointly to undertake. U.S.S.G. § 1B1.3, application note 2. To do this, the court may consider any “explicit agreement or implicit agreement fairly inferred from the conduct of the defendant and others.” Id.; see United States v. Innamorati, 996 F.2d 456, 488-89 (1st Cir.1993) (members of drug distribution conspiracy may be held accountable at sentencing for different quantities of narcotics, “depending on the circumstances of each defendant’s involvement”); United States v. Collado, 975 F.2d 985, 992 (3d Cir.1992) (“the crucial factor in accomplice attribution is the extent of the defendant’s involvement in the conspiracy”); Wilkens & Steer, Relevant Conduct: The Cornerstone of the Federal Sentencing Guidelines, 41 S.C.L.Rev. 495, 511 (1990) (“liability might be justified for those who are at the top directing and controlling the entire operation”) (quoting 2 W. LaFave & A. Scott, Substantive Criminal Law § 6.8, at 155 (1986)). On remand here, therefore, the district court must determine (1) the scope of the joint criminal activity explicitly or implicitly agreed to by Patriarca jointly with others; (2) whether the criminal acts proffered as relevant conduct were in furtherance of this jointly undertaken criminal activity; and (3) whether the proffered acts were reasonably foreseeable in connection with that criminal activity. These determinations will fix the relevant conduct under § 1B1.3 for purposes of calculating the offense level under § 2E1.1. Such determinations are, of course, all inherently fact-bound. See, e.g. Innamorati, 996 F.2d at 488-89. Rather than applying § 1B1.3 to § 2E1.1 in the straightforward manner discussed above, the district court limited relevant conduct to only those predicate acts that were charged against Patriarca personally — namely, the Travel Act violations. In doing so, the district court improperly treated the term “underlying racketeering activity” in § 2E1.1(a)(2) as if it “otherwise specified” that relevant conduct should not apply to each “offense of conviction” (including the RICO conspiracy count and the substantive RICO count) and instead should apply only to the predicate Travel Act violations. See U.S.S.G. § lB1.3(a) (“Unless otherwise specified, ... cross references ... shall be determined on the basis of ... all reasonably foreseeable acts ... that occurred during the commission of the offense of conviction ... ”) (emphasis added). This was error. “Subsec *77 tion (a) [of § 1B1.3] establishes a rale of construction by specifying, in the absence of more explicit instructions in the context of a specific guideline, the range of conduct that is relevant to determining the applicable offense level....” U.S.S.G. § 1B1.3, Background. The background commentary to § 1B1.3 further makes clear that “[cjonduct that is not formally charged or is not an element of the offense of conviction may enter into the determination of the applicable guideline range.” Section 2E1.1 — specifically the term “underlying racketeering activity” — contains no explicit instructions displacing the general rule in § IB 1.3 that relevant conduct includes uncharged conduct. In a RICO case, there is no justification for limiting “underlying racketeering activity” just to predicate acts specifically charged against one defendant. 4 We, therefore, agree with the government that the term “underlying racketeering activity” in § 2E1.1(a)(2) means simply any act, whether or not charged against defendant personally, that qualifies as a RICO predicate act under 18 U.S.C. § 1961(1) 5 and is otherwise relevant conduct under § 1B1.3. Because the reference to “underlying racketeering activity” is a cross reference, § IB 1.3 comes into play and defines “the range of conduct that is relevant....” See U.S.S.G. § 1B1.3, Background. It follows that the acts of relevant conduct proffered by the government, all of which are racketeering acts that could have been charged as predicate offenses, come under the heading of “relevant conduct” for sentencing Patriarca on the RICO counts of conviction, provided they otherwise meet the accomplice attribution standards of § lB1.3(a)(l)(B). To avoid this conclusion, Patriarca cites to application notes 1 and 5 to U.S.S.G. § 1B1.2. He argues that these application notes show that the term “underlying racketeering activity” should be limited to the specific predicate acts charged against him. We think that neither application note is applicable here. Application note 1 to U.S.S.G. § 1B1.2 states the following: This section provides the basic rules for determining the guidelines applicable to the offense conduct under Chapter Two (Offense Conduct). As a general rule, the court is to use the guideline section from Chapter Two most applicable to the offense of conviction. The Statutory Index (Appendix A) provides a listing to assist in this determination. When a particular statute proscribes only a single type of criminal conduct, the offense of conviction and the conduct proscribed by the statute will coincide, and there will be only one offense guideline referenced. When a particular statute proscribes a variety of conduct that might constitute the subject of different offense guidelines, the court will determine which guideline section applies based upon the nature of the offense conduct charged in the count of which the defendant was convicted. (Emphasis ours.) Patriarca relies on the emphasized portion for the proposition that relevant conduct pertaining to composite crimes, like RICO, must be limited to conduct charged in the indict *78 ment. It is clear, however, from the full text of the application note, that the note is meant to guide courts in the initial selection of the applicable guideline in Chapter Two, not to limit cross references within a particular guideline. There is no question here that the applicable guideline for RICO convictions is § 2E1.1. Thus, application note 1 to § 1B1.2 provides no support for Patriarca’s argument. Application note 5 to § 1B1.2 is equally immaterial to the application of relevant conduct to § 2E1.1. Application note 5 relates specifically to § lB1.2(d), which states that: “A conviction on a count charging a conspiracy to commit more than one offense shall be treated as if the defendant had been convicted on a separate count of conspiracy for each offense that the defendant conspired to commit.” Application note 5 in turn provides the following: Particular care must be taken in applying subsection (d) because there are cases in which the verdict or plea does not establish which offense(s) was the object of the conspiracy. In such eases, subsection (d) should only be applied with respect to an object offense alleged in the conspiracy count if the court, were it sitting as a trier of fact, would convict the defendant of conspiring to commit that object offense. Note, however, if the object offenses specified in the conspiracy count would be grouped together under § 3D1.2(d) {e.g., a conspiracy to steal three government checks) it is not necessary to engage in the foregoing analysis, because § lB1.3(a)(2) governs consideration of the defendant’s conduct. U.S.S.G. § 1B1.2, application note 5 (emphasis added). In arguing that § IB 1.2(d) and application note 5 limit relevant conduct in composite cases, like RICO cases, to conduct “alleged” in the indictment as predicate acts, Patriarca notes similar language in the application notes to § 2E1.1 and the multiple count rules. Application note 1 to § 2E1.1 states that “[wjhere there is more than one underlying offense, treat each underlying offense as if contained in a separate count of convic-tion____” Similarly, the introductory commentary to the multiple count rules provides that “[s]ome offenses, e.g., racketeering and conspiracy, may be ‘composite’ in that they involve a pattern of conduct or scheme involving multiple underlying offenses. The rules in this Part are to be used to determine the offense level for such composite offenses from the offense level for the underlying offenses.” Application note 8 to § 3D1.2 refers specifically back to § lB1.2(d): “A defendant may be convicted of conspiring to commit several substantive offenses and also of committing one or more of the substantive offenses. In such cases, treat the conspiracy count as if it were several counts, each charging conspiracy to commit one of the substantive offenses. See § IB 1.2(d) and accompanying commentary.” U.S.S.G. § 3D1.2, application note 8 (emphasis added). On the basis of this commentary, Patriarca contends that the only way to apply the multiple count section of the guidelines to a RICO conviction is to use the directions for the more commonly applied conspiracy, and hence the rule of § lB1.2(d). 6 We disagree. First, the definition of “offense” contained in the application notes to U.S.S.G. § 1B1.1 is *79 not limited to charged offenses. Instead, “offense” is defined to mean “the offense of conviction and all relevant conduct under § 1B1.S (Relevant Conduct) unless a different meaning is specified or is otherwise clear from the context.” U.S.S.G. § 1B1.1, application note 1{1) (emphasis added). As stated previously, § 2E1.1 does not specify a different meaning; therefore, there is no reason to interpret “underlying offense” to exclude uncharged conduct. Furthermore, although applying § lB1.2(d) to RICO convictions has some superficial appeal, there would be insurmountable obstacles to its practical application. First, by its own terms, § lB1.2(d) is limited to counts “charging a conspiracy.” Therefore, it is difficult to see how § lB1.2(d) could apply to a substantive RICO violation (as opposed to a RICO conspiracy). Even overlooking this language, it would be impossible under application note 5 for a court to determine whether it “would convict the defendant of conspiring to commit” an underlying offense in situations where the defendant is charged with a substantive RICO violation and the underlying offense is not a conspiracy. Thus, § lB1.2(d) is inapplicable to noneonspiracy offenses such as a substantive RICO violation. There are problems with applying § lB1.2(d) to RICO conspiracies as well. It seems clear from the plain text of § lB1.2(d), the application notes, and the official commentary 7 that § lB1.2(d) was enacted to deal with multiple object conspiracies charged in a single count. A RICO conspiracy, however, is considered a single object conspiracy with that object being the violation of RICO. United States v. Ashman, 979 F.2d 469, 485 (7th Cir.1992) (“The goal of a RICO conspiracy is a violation of RICO.”) (quoting United States v. Neapolitan, 791 F.2d 489,496 (7th Cir.), cert. denied, 479 U.S. 940, 107 S.Ct. 422, 93 L.Ed.2d 372 (1986)), petition for cert. filed sub nom. Barcal v. United States, 61 U.S.L.W. 3857 (U.S. April 6, 1993) (No. 92-1804). In enacting RICO, Congress intended that “ ‘a series of agreements that under pre-RICO law would constitute multiple conspiracies could under RICO be tried as a single enterprise conspiracy’ if the defendants have agreed to commit a substantive RICO offense.” 8 United States v. Riccobene, 709 F.2d 214, 224-25 (3d Cir.) (quoting United States v. Sutherland, 656 F.2d 1181, 1192 (5th Cir.1981), cert. denied, 455 U.S. 949, 102 S.Ct. 1451, 71 L.Ed.2d 663 (1982) (internal citation omitted)), ce rt. denied sub nom. Ciancaglini v. United States, 464 U.S. 849, 104 S.Ct. 157, 78 L.Ed.2d 145 (1983). Application notes 1 and 5 to § 1B1.2 are not, therefore, material to determining *80 whether relevant conduct must be limited to predicate acts charged against a defendant. Instead, § 1B1.3 determines the range of conduct that is relevant to cross references such as the term “underlying racketeering activity” in § 2E1.1(a)(2), and the background commentary to § 1B1.3 makes clear that “[c]onduct that is not formally charged ... may enter into the determination of the applicable guideline sentencing range.” Because the application of § 1B1.3 to § 2E1.1 is straightforward and unambiguous, the district court erred in resorting to the general principles underlying the guidelines and the general rule of construction that “courts should construe statutes to avoid decision as to their constitutionality.” See, e.g. United States v. Monsanto, 491 U.S. 600, 611, 109 S.Ct. 2657, 2664, 105 L.Ed.2d 512 (1989). “[C]ourts should strive to apply the guidelines as written, giving full force and effect to the Sentencing Commission’s interpretive commentary and application notes.” United States v. Zapata, 1 F.3d 46 (1st Cir.1993); accord Stinson v. United States, — U.S. -, -, 113 S.Ct. 1913, 1915, 123 L.Ed.2d 598 (1993); United States v. Brewster, 1 F.3d 51 (1st Cir.1993). Absent specific provision in § 2E1.1 that “underlying racketeering activity” includes only charged predicate acts, we see no principled basis to read such a limitation into the provision. Even were the application of relevant conduct to § 2E1.1 less clear than it is, we would have trouble accepting the three principles cited by the district court as the rationale for limiting relevant conduct to the predicate acts actually charged against a defendant. The district court felt that “the Sentencing Guidelines are closer to a ‘charge offense’ system than a ‘real offense’ system of punishment.” Patriarca, 807 F.Supp. at 190; U.S.S.G. Ch. 1, Pt. A, 4(a), p. 5. In the court’s view, the reason the government did not charge the conduct at issue in this appeal as predicate acts in the indictment is because the government had insufficient evidence to sustain a conviction for this conduct. 807 F.Supp. at 191. Because conduct “which the prosecutor can prove in court” is supposed to “impose[ ] a natural limit upon the prosecutor’s ability to increase a defendant’s sentence,” U.S.S.G. Ch. 1, Pt. A, 4(a), p. 5, the court thought that it would be improper for a sentencing court to increase a defendant’s sentence on the basis of uncharged predicate acts. Similar arguments have been rejected by this court and virtually every other circuit court to have addressed the issue. See, e.g., United States v. Mocciola, 891 F.2d 13, 16-17 (1st Cir.1989); United States v. Galloway, 976 F.2d 414, 424 n. 6 (8th Cir.1992) (collecting cases), cert. denied, — U.S. -, 113 S.Ct. 1420, 122 L.Ed.2d 790 (1993). While the district court is correct that “for the most part, the court will determine the applicable guideline by looking to the charge of which the offender was convicted,” United States v. Blanco, 888 F.2d 907, 910 (1st Cir.1989), real offense principles enter into the punishment prescribed in the guidelines through the relevant conduct guideline, § 1B1.3. Breyer, The Federal Sentencing Guidelines and the Key Compromises Upon Which They Rest, 17 Hofstra L.Rev. 1,11-12 (1988). Relevant conduct increases a defendant’s sentence, sometimes very significantly, despite the fact that it was not charged in an indictment, e.g., Blanco, 888 F.2d at 910, and even despite the fact that a jury may have acquitted the defendant for that precise conduct. E.g., Mocciola, 891 F.2d at 16-17; United States v. Rumney, 867 F.2d 714, 719 (1st Cir.) (“traditional sentencing factors need not be pleaded and proved at trial”) (quoting United States v. Brewer, 853 F.2d 1319, 1326 (6th Cir.), cert. denied, 488 U.S. 946, 109 S.Ct. 375, 102 L.Ed.2d 364 (1988)), cert. denied, 491 U.S. 908, 109 S.Ct. 3194, 105 L.Ed.2d 702 (1989). This is because sentencing factors, including the applicability of relevant conduct, need only be proven by a preponderance of the evidence, not beyond a reasonable doubt. Mocciola, 891 F.2d at 16-17; Galloway, 976 F.2d at 424 n. 6. As noted below, in pre-guideline cases courts likewise took into account untried criminal conduct when exercising sentencing discretion. The fact that the government has not charged and proven beyond a reasonable doubt the conduct now asserted as relevant conduct does not prevent the increase in sentence resulting from the relevant conduct guideline. We see no special reason to devi *81 ate from this principle when dealing with a RICO conviction. Nor are we as convinced as the district court that sentencing Patriarca on the basis of uncharged relevant conduct might be so unfair as to raise due process concerns. The district court assumed that if Patriarca was held responsible for either the Limoli or Berns murder, Patriarca would face a potential life sentence under the guidelines and the RICO penalty provision. We believe that the district court was mistaken in this assumption. The RICO statute sets the maximum prison sentence at 20 years unless “the violation is based on a racketeering activity for which the maximum penalty includes life imprisonment.” 18 U.S.C. 1963(a) (emphasis added). We agree with the government that the statutory maximum sentence must be determined by the conduct alleged within the four corners of the indictment. Otherwise, a defendant would not know at the time of his arraignment or change of plea what his maximum possible sentence would be on the charged offenses. The charged conduct, if proven beyond a reasonable doubt, constitutes the “violation” of which a defendant is convicted. Patriar-ca’s charged conduct included no acts such as would result in a life sentence. The predicate acts charged in the indictment weré all violations of the Travel Act, which does not carry a possible life sentence. Therefore, while for sentencing purposes within the 20-year maximum Patriarca is liable for uncharged as well as charged relevant conduct, his maximum penalty is fixed at twenty years for each of the RICO counts. 9 The relevant conduct determination here affects only where, mthin that statutory range, Patriarca should be sentenced. We are also unpersuaded by the district court’s concern .that sentencing Patriarca on the basis of uncharged predicate acts would be “inconsistent with the Sentencing Commission’s intention to set up a system which is' not administratively unwieldy.” Patriarca, 807 F.Supp. at 192. While it is true that considerations of administrative efficiency as well as procedural fairness prompted the Commission to require sentencing courts “to determine the applicable guideline by looking to the charge of which the offender was convicted,” it is also clear that the Commission intended real offense principles to apply to determine the applicability of various adjustments, including cross references. See Blanco, 888 F.2d at 910. The fact that application of real offense principles may burden a sentencing court with additional fact finding is no reason to ignore the Commission’s compromise between “real offense” and “charge offense” sentencing. See id. at 911. Sentencing a RICO defendant on the basis of uncharged predicate acts may not, indeed, prove to be impracticable. Drug conspirators are frequently sentenced on the basis of drug transactions committed by coconspira-tors. In both situations, the court must determine the scope of the criminal activity agreed to by the defendant, the reasonable foreseeability of the conduct proffered as relevant conduct, and whether the relevant conduct was in furtherance of the jointly undertaken activity. To be sure, the wide range of crimes covered by RICO may inject new complexities, but, if so, the remedy lies with the Sentencing Commission. The courts are *82 not empowered to rewrite the relevant conduct guideline. 10 Finally, the district court’s observation that there are apparently no reported pre-guideline cases in which a RICO defendant was sentenced on the basis of ah uncharged murder is not dispositive. There could, of course, have been such eases that went unreported or unappealed. Sentences were not usually the subject of published opinions pri- or to the guidelines. And courts often used material information from many sources in exercising their discretion to set a sentence within the permissible, often very wide, statutory range. See, e.g., Roberts v. United States, 445 U.S. 552, 556, 100 S.Ct. 1358, 1362, 63 L.Ed.2d 622 (1980); Williams v. New York, 337 U.S. 241, 250-51, 69 S.Ct. 1079, 1084-85, 93 L.Ed. 1337 (1949); United States v. Lee, 818 F.2d 1052, 1055 (2d Cir.) (“Any circumstance that aids the sentencing court in deriving a more complete and true picture regarding the convicted person’s background, history, or behavior is properly considered. For that reason, - ..., other crimes for which the defendant was neither tried nor convicted, and crimes charged that resulted in acquittal may be used by the sentencing court in determining sentence”) (citation omitted), cert. denied, 484 U.S. 956, 108 S.Ct. 350, 98 L.Ed.2d 376 (1987). We conclude that the principles put forward by the district court provide no adequate reason for limiting relevant conduct to charged predicate acts in RICO cases. Because the district court incorrectly interpreted the guidelines, it did not reach a host of other arguments raised by Patriarca in an attempt to avoid the straightforward application of the relevant conduct guideline, § 1B1.3, to § 2E1.1. These arguments include: whether due process and the Confrontation Clause require additional procedures, such as a higher standard of proof than preponderance of the evidence, in order to hold Patriarca responsible for the proffered relevant conduct; whether due process requires notice of proffered relevant conduct not otherwise disclosed in the indictment pri- or to the entry of a defendant’s plea of guilty; whether sentencing Patriarca for murders that occurred prior to the effective date of the guidelines violates the Ex Post Facto Clause, even though the RICO offense extended beyond that date; whether the relevant conduct guideline exceeds the Sentencing Commission’s statutory authority; and whether the government should be estopped from arguing that Patriarca is responsible for the proffered relevant conduct. Several of these arguments have been expressly rejected by this circuit and others. See, e.g., United States v. Brewster, 1 F.3d 51 (1st Cir.1993) (“Absent bad faith ... the critical time for disclosure of sentence-related information is not prior to the taking of a plea, but prior to sentencing.”); United States v. David, 940 F.2d 722, 739 (1st Cir.1991) (“It is well established that the guidelines apply to a defendant whose offense begins before the guidelines’ effective date and continues after the effective date.”), cert. denied, — U.S. -, 112 S.Ct. 2301, 119 L.Ed.2d 224 (1992); Galloway, 976 F.2d at 421-22 (rejecting argument that relevant conduct provision is not authorized by the Sentencing Reform Act). However, it is not necessary for us to decide these issues in the first instance. On remand, the district court should consider, and where appropriate, decide those issues that Patriarca chooses to assert again. In a last ditch effort to avoid resentencing, Patriarca contends that the legal issue of how relevant conduct is applied to the RICO guideline has been effectively mooted by the district court’s findings concerning the proffered acts of relevant conduct. Patriarca asserts that the in the course of determining whether an upward departure was warranted pursuant to- § 5K2.0 and § 4A1.3, the court found, as a matter of fact, that the government had not established his criminal liability for five of the seven relevant conduct allegations — the Limoli and Berns homicides, the *83 Carrozza drug dealing, the Ferrara “hit,” or the harboring of Alphonse Pérsico. 11 This contention has no merit. The district court expressly stated that because it found that relevant conduct must be limited to charged predicate acts, it was not deciding “whether the crimes at issue with regard to relevant conduct were within the scope of the defendant’s conspiracy and/or reasonably foreseeable consequences of it.” Patriarca, 807 F.Supp. at 196. In discussing the purported conduct in its upward departure analysis, the district court merely stated that it was not persuaded that Patriarca knew of, or personally participated in, these offenses. However, a defendant can be accountable for the acts of his coconspirators under § IB 1.3 without having been personally involved. The standard is whether the acts of cocon-spirators were in furtherance of the jointly undertaken activity and were reasonably foreseeable to the defendant. The seven acts proffered as relevant conduct must be reexamined in light of this standard. C. Patriarca’s Appeal Patriarca appeals from the district court’s upward departure under U.S.S.G. § 4A1.3 and from the district court’s imposition of the costs of incarceration and supervision under U.S.S.G. § 5E1.2(i). Our holding that the district court must resentenee Patriarca on the basis of his relevant conduct moots the issue of the propriety of the court’s upward departure. The district court departed upwards under § 4A1.3 on the basis of its finding that the government had proved by a preponderance of the evidence that Patriarca had “aided and abetted drug crimes committed by Salvatore Michael Caruana” from 1981 to 1983. Patriarca, 807 F.Supp. at 170. Because on remand the court will decide if the Caruana conspiracy is relevant conduct for RICO sentencing purposes, its utilization as a basis for upward departure need not be considered here, and is vacated. For similar reasons, we must reject Patr-iarcal challenge to his cost-of-imprisonment fine. As part of his sentence, the district court ordered Patriarca to pay a fine of $50,000 pursuant to U.S.S.G. § 5E1.2(e), plus $122,344 for the cost of his imprisonment, and $3,954 for the cost of his supervision. Patriarca, 807 F.Supp. at 210. The later portion of the fine was assessed pursuant to U.S.S.G. § 5E1.2(i), which states the following: Notwithstanding the provisions of subsection (c) [the minimum-maximum fine table] of this section, but subject to the provisions of subsection (f) [the defendant’s ability to pay] herein, the court shall impose an additional fine amount that is at least sufficient to pay the costs to the government of any imprisonment, probation, or supervised release ordered. U.S.S.G. § 5E1.2(i). Patriarca contends that the Sentencing Reform Act, 18 U.S.C. § 3553(a), does not authorize imposition of a fine to recompense the government for the cost of incarceration or supervised release, and § 5E1.2(i) is therefore invalid. The few circuit courts to have addressed this question agree that the Sentencing Reform Act does not authorize the assessment of a fine solely to pay for the costs of a defendant’s imprisonment. United States v. Spiropoulos, 976 F.2d 155, 165-69 (3d Cir.1992); United States v. Hagmann, 950 F.2d 175, 187 n. 29 (5th Cir.1991), cert. denied, — U.S. -, 113 S.Ct. 108, 121 L.Ed.2d 66 (1992). They disagree, however, as to whether § 5E1.2(i) can be justified on other grounds. Compare United States v. Turner, 998 F.2d 534 (7th Cir.1993) (costs of confinement reflect seriousness of the crime and increase deterrence) and Hagmann, 950 F.2d at 187 (“the uniform practice of fining *84 criminals on the basis of their individualistic terms of imprisonment — an indicator of the actual harm each has inflicted upon society— is a rational means to assist the victims of crime collectively”) with Spiropoulos, 976 F.2d at 168 (“The cost of imprisoning a defendant has little, if anything, to do with the amount that the defendant has harmed his or her victim(s), and is therefore questionable as an appropriate method of restitution.”); see United States v. Doyan, 909 F.2d 412, 416 (10th Cir.1990) (“Whether the purpose of the contested fine is to punish, deter, or to spare the taxpayers a substantial expense that has been generated by an intentional criminal act, we cannot say that Guideline § 5E1.2(i) as applied here bears no rational relation to the legitimate governmental interest in criminal justice.”). The government here argues that § 5E1.2(i) is merely a means of achieving the clearly authorized purpose of punishing a defendant based on the seriousness of his or her offense. We do not find it appropriate to answer this question at the present time. First, Patriarca did not object to his cost-of-imprisonment fine at the time of sentencing. Hence, the district court had no reason to focus on the issue, and we lack the benefit of its considered views. Absent plain error, we normally will not consider an issue raised for the first time on appeal. See United States v. Newman, 982 F.2d 665, 672 (1st Cir.1992), petition for cert. filed, 61 U.S.L.W. 3751 (U.S. April 22, 1993) (No. 92-1703); United States v. Haggert, 980 F.2d 8, 11 (1st Cir.1992); United States v. Mondello, 927 F.2d 1463, 1468 (9th Cir.1991) (refusing to consider argument not raised below that the fine provisions of the Guidelines are contrary to statutory authorization). Because the fine issue is one which has divided our sister circuits, we cannot see that the district court’s alleged error in assessing the § 5E1.2(i) fine was a “plain” one within the meaning of Fed.R.Crim.P. 52(b). See United States v. Olano, — U.S. -, -, 113 S.Ct. 1770, 1777, 123 L.Ed.2d 508 (1993). In addition, our decision that Patriarca must be resentenced taking into account uncharged relevant conduct requires that we also vacate the fine portion of Patriarca’s sentence. Should the district court on remand determine that Patriarca must be sentenced at a higher base offense level, his minimum and maximum fine range under § 5E1.2(c) will likewise increase. Moreover, Patriarca’s cost of imprisonment necessarily depends upon the length of his confinement. On remand, Patriarca can argue that a cost-of-imprisonment fine under § 5E1.2(i) is inconsistent with the Sentencing Reform Act. Should the district court reject the argument and Patriarca again appeal from the fine, that will be the appropriate time for this court to decide the question. Presently, however, because we must vacate the fine and because Patriarca did not raise the issue below — hence failing to bring the claimed error to the district court’s attention for focused consideration — we find it inappropriate to decide whether § 5E1.2(i) is valid. II. Carrozza’s Sentence DefendanVappellant Robert F. Carrozza appeals from a judgment of conviction and a 228-month sentence imposed by the district court, after Carrozza pleaded guilty to 49 counts of racketeering-related offenses, including violations of the RICO statute, extortion, kidnapping, loansharking, narcotics distribution, gambling, obstruction of justice, and intimidation of a witness. 807 F.Supp. 156. Carrozza argues that the district court’s decision to “assume” that Carrozza’s base offense level should be adjusted upwards for his role in the offense constituted plain error. After extensive plea negotiations, Carrozza and four of his codefendants entered into plea agreements with the government. Pursuant to Fed.R.Crim.P. 11(e)(1)(C), Carrozza and the government 'agreed that a specific sentence of 228 months was “the appropriate disposition of the ease,” constituting “a justifiable departure within the meaning of § 6B1.2(c)(2) of the United States Sentencing Guidelines.” Both parties agree on appeal that this “justifiable departure” was understood to have been a downward departure. Apart from the agreed sentence, Carrozza received two additional benefits in exchange for his plea of guilty. First, the government promised not to prosecute Carrozza for his *85 alleged involvement in the murder of William Grasso and the attempted murder of Francis Salemme, Sr. Second, the government promised that Carrozza would not be subpoenaed to testify in any federal grand jury investigation in the District of Massachusetts relating to the activities of the Patriarca Family occurring before the date of the plea agreement. The plea agreements for all five defendants were made expressly contingent upon the district court’s acceptance of the pleas of guilty from each defendant. According to the government, the interdependency of the plea agreements reflected the government’s major purpose in entering the agreements— eliminating the need for any trial, which the parties estimated would take from six months to a year or more. Because removing some but not all of the defendants would not significantly reduce the time necessary to try the case, the government bargained for, and obtained, the option to withdraw all five plea agreements if any of the defendants moved successfully to withdraw his plea. The district court conditionally accepted the guilty pleas pending consideration of the presentenee reports (“PSR”). The preliminary PSR for Carrozza was completed on April 3,1992. The preliminary PSR calculated his applicable guideline range, based on an offense level of 33 and a criminal history category of IV, to be 188-235 months, and therefore concluded that the agreed sentence of 228 months was consistent with the guidelines. The PSR determined that there were no factors warranting departure. On April 9, the government filed its objections to this PSR, complaining that some of the PSR’s calculations of Carrozza’s offense level were too high and some were too low. The key objection made by the government was that the PSR should have made an upward adjustment pursuant to U.S.S.G. § 3Bl.l(a) for Carrozza’s role as an organizer or leader in several of the offenses charged. The government calculated the applicable guideline range to be 235-293 months, based upon an offense level of 35 and a criminal history category of IV. Finally, the government argued that there were justifiable factors to support a downward departure. Carrozza filed several specific objections to the PSR on April 17. Although Carrozza did not discuss the particulars of his own calculation of the applicable guideline range, he did argue that “a downward departure to the agreed upon sentence” was warranted. Car-rozza did not dispute the government’s calculations as to his role in the offenses charged. The Addendum to the PSR was completed on April 23. The Addendum accepted some of the government’s objections calling for a downward revision in the offense level calculations, but rejected the government’s role in the offense objection because the government had not provided “sufficient information ... in the details of the particular episodes to delineate the individual roles of the defendant within those episodes.” The Addendum recalculated the total offense level to be 31, yielding a guideline range of 151-188 months. The Addendum noted that this range would require “an upward departure if the Court were to sentence the defendant to the amount of time designated in the plea agreement [228 months].” On the same day that the Addendum to the PSR was disclosed, the government and Carrozza filed separate sentencing memoran-da, each arguing to the court that the agreed upon sentence constituted a justifiable downward departure. The sentencing hearing was held on April 29, 1992. At the outset of the sentencing hearing, the court explained that under Rule 11(e)(1)(C), it could either accept the plea agreements and impose the agreed-upon sentence in each case, or reject the agreements and offer the defendants an opportunity to withdraw their pleas. The court clearly articulated the disparate guideline ranges calculated by the government and the probation office and then stated: I think the most sensible thing to do is to not resolve that dispute but to decide whether the 228-month sentence, which I think is about 19 years, if it is, indeed, an eight-month downward departure as the Government intends, is appropriate---- *86 Well, I am going to proceed ... in the following fashion: I am not deciding whether the Guidelines are 151 months to 188 months or [if] Probation’s calculation is followed, which would involve 40-month upward departure or whether, as the Government contends, that Mr. Carrozza has assumed up to now, [or] at least up to the time of his plea, the calculations might be 235 to 293 months. I am going to analyze this in the context of the question being whether — if the Guidelines are 235 [to] 293 months, the seven-month downward departure to 228 months [as] called for by the plea agreement is justifiable. The court explained that it was not resolving the dispute because Fed.R.Crim.P. 32 does not require resolution of issues that will not be material to the sentence to be imposed. When asked if anyone objected to this procedure, the parties responded “no.” Consistent with their prehearing positions, both Carrozza and the government argued that the agreed upon sentence represented a justifiable downward departure from the applicable guideline range. The court thereafter sentenced Carrozza to 228 months imprisonment, to be followed by 60 months supervised release. On the same day, the district court entered an order relating to the presentenee reports. In this order, the court stated that one of the justifications for its downward departures for several of the defendants was that the departures “eliminated the need for both a lengthy trial (which it was estimated would take six months to a year) and for protracted sentencing hearings to resolve disputes relevant to the term of incarceration to be imposed on each defendant.” On April 30, 1992, the court entered its “Second Order Relating to Presentence Report” in Carrozza’s case. In that order, the court expressly relied on Fed.R.Crim.P. 32(c)(3)(D)(ii) as its justification for failing to calculate the sentencing guideline range applicable to Carrozza: With regard to the government’s objections to the PSR, the court, pursuant to Fed.R.Crim.P. 32(c)(3)(D)(ii), did not decide whether the applicable Sentencing Guidelines were 235 to 293 months as asserted by the government or 151 to 188 months as recommended by the Probation Officer. Rather, the court' assumed the Sentencing Guidelines were a minimum of 235 months and agreed with the government and the defendant that if the binding plea agreement, pursuant to Fed. R.Crim.P. 11(c)(1)(C) [sic], calling for a sentence of 228 months represented a departure, there were justifiable reasons for it. Thus, the agreed-upon 228 month sentence was imposed. Judgment was entered on May 1, 1992. Notwithstanding the court’s previous assertions that it was merely “assuming” that the government was correct, the judgment indicates that the court found the guideline range to be 235-293 months and imposed a downward departure for justifiable reasons. In a May 7, 1992 memorandum explaining its sentence, the court once again stated its basic assumption: In the Presentence Report, the Probation Department calculated Carrozza’s Sentencing Guidelines to 151 to 188 months. The government, however, contended that the proper calculation of Carrozza’s Sentencing Guidelines was 235 to 293 months. Carrozza’s plea agreement specified a sentence of 228 months, or 19 years, in prison. The court analyzed his plea agreement on the assumption that the required sentence represented a seven month downward departure. On appeal, Carrozza contends that the sentencing procedure employed by the district court was patently unlawful because the court failed to determine the applicable guideline range. A. Plain Error Standard 12 Carrozza concedes that because he failed to object to the district court’s course of *87 conduct during the sentencing hearing, his sentence can be reversed only upon a showing of plain error. See Fed.R.Crim.P. 52(b). Carrozza has failed to make such a showing here. The Supreme Court recently interpreted the plain error rule in United States v. Olano, — U.S. -, 113 S.Ct. 1770, 123 L.Ed.2d 508 (1993). In Olano, the Court reiterated the three limitations on appellate authority to recognize errors under Fed. R.Crim.P. 52(b): (1) there must be an “error,” (2) the error must be “plain,” and (3) the error must “affee[t] substantial rights.” Id. — U.S. at----•, 113 S.Ct. at 1777-78. Even if a defendant can establish all three criteria, an appellate court has discretion not to review the error because Rule 52(b) is written in permissive, not mandatory, terms. Id. — U.S. -, 113 S.Ct. at 1778. The standard to guide that discretion was stated in United States v. Atkinson, 297 U.S. 157, 160, 56 S.Ct. 391, 392, 80 L.Ed. 555 (1936): appellate courts should correct plain forfeited errors affecting substantial rights if the errors “seriously affect[] the fairness, integrity or public reputation of judicial proceedings.” Olano, — U.S. at -, 113 S.Ct. at 1779. We agree with Carrozza that the district court committed error when it “assumed” that Carrozza’s guideline range was 235-293 months prior to its “downward” departure to 228 months. Before accepting a plea agreement that contains a specific sentence under Fed.R.Crim.P. 11(e)(1)(C), a sentencing court is required to satisfy itself either that: “(1) the agreed sentence is within the applicable guideline range; or (2) the agreed sentence departs from the applicable guideline range for justifiable reasons.” U.S.S.G. § 6B1.2(e). To determine whether the sentence departs from the applicable guideline range for justifiable reasons, the court must first determine what the applicable guideline range is and then analyze whether a departure is authorized by 18 U.S.C. § 3553(b) and the general departure rules in Chapter 1, Part A (4)(b) of the Guidelines. See U.S.S.G. § 6B1.3, Commentary. In effect, § 6B1.2(c) instructs courts to apply general guideline principles when determining whether to accept a plea under Fed.R.Crim.P. 11(e)(1)(C). See U.S.S.G. § 1B1.1 (general instructions on applying the guidelines). 13 In sentencing Carrozza, the district court mistakenly believed that Fed.R.Crim.P. 32(c)(3)(D) authorized its decision not to determine an actual guideline range for Carroz-za’s offenses. As we have stated, this was error. Rule 32(c)(3)(D) 14 apparently relates *88 to factual inaccuracies in a presentence report, not to mixed questions of law and fact that a defendant does not dispute. See United States v. Hand, 913 F.2d 854, 857 (10th Cir.1990) (defendant’s disagreement over PSR’s legal conclusion that defendant was not a minor participant does not allege factual inaccuracies in the PSR and does not implicate Rule 32(c)(3)(D)). But see United States v. Rosado-Ubiera, 947 F.2d 644, 646 (2d Cir.1991) (Rule 32(c)(3)(D) was violated when court failed to resolve the defendant’s precise role in the offense). To be sure, a district court has inherent power not to decide disputes that are immaterial or irrelevant to the ultimate sentence. For example, a sentencing court need not determine whether prior convictions should be added to a defendant’s criminal history-score if the addition will not affect the defendant’s criminal history category. See United States v. Lopez, 923 F.2d 47, 51 (5th Cir.), cert. denied, — U.S. -, 111 S.Ct. 2032, 114 L.Ed.2d 117 (1991). We have also held that a sentencing court need not choose between two overlapping guideline ranges when the same sentence would have been imposed under either range. United States v. Ortiz, 966 F.2d 707, 718 (1st Cir.1992), cert. denied , — U.S. -, 113 S.Ct. 1005, 122 L.Ed.2d 154 (1993); United States v. Concemi, 957 F.2d 942, 953 (1st Cir.1992); United States v. Bermingham, 855 F.2d 925, 934 (2d Cir.1988). Here however, the two ranges did not overlap, nor was the sentencing factor immaterial to Carrozza’s guideline range. Instead, if the disputed factor was decided in the government’s favor, Carrozza’s sentencing range would be 235-293 months, but if the issue was decided as the probation office recommended, Carrozza’s guideline range would be 151-188 months. We have also intimated in the past that if a sentencing court intends to depart, an error in applying the guidelines may prove to be harmless if the court makes clear that it would have departed to the same sentence regardless of the exact guideline range. United States v. Plaza-Garcia, 914 F.2d 345, 347 (1st Cir.1990). We have never, however, suggested that a sentencing court need not decide a sentencing factor when one decision will result in an upward departure and another in a downward departure. Such a rule would be inimical to the very principle behind guideline departures. United States v. McCall, 915 F.2d 811, 814 n. 3 (2d Cir.1990) (rejecting government’s argument that incorrect application of guidelines, resulting in range of 151-188 months instead of 87-108 months, was irrelevant to court’s ultimate sentence because court indicated an intent to depart down for substantial cooperation — an indication that could not be reconciled with court’s 108-month sentence, which was at the high end of the correct guideline range). The district court, therefore, erred when it simply assumed that Carrozza’s guideline range was 235-293 months, and ignored the probation office’s suggestion that the range should be 151-188 months. And while we can understand and sympathize with the district court’s desire to avoid an obtuse decision that may have seemed academic, we think the error was “plain” in the sense that it was both “clear” and “obvious.” See Olano, — U.S. at -, 113 S.Ct. at 1777. That is enough to pass the second hurdle to appellate authority under Rule 52(b). We now turn to the third and often deciding factor in our plain error analysis— whether the error affects substantial rights. In most cases, “although perhaps not in every case, the defendant must make a specific showing of prejudice to satisfy the ‘affecting substantial rights’ prong of Rule 52(b).” Olano, — U.S. at -, 113 S.Ct. at 1778. For several reasons, we think Carrozza has not made such a showing here. First, Carrozza has not argued, and points to no evidence suggesting, that an actual adjustment for his role in the offense would have been improper. Instead, Carrozza merely assumes that the district court’s “assumption” was in fact wrong, and argues that he was prejudiced because his actual guide *89 line range might have been years shorter than that assumed by the district court. However, analysis of the record reveals that, in all likelihood, the district court’s assumption that Carrozza’s base offense level was 35 was correct. The difference between the government’s and the probation office’s calculations of Car-rozza’s base offense level resulted entirely from whether or not an upward adjustment should have been given for Carrozza’s role in the drug conspiracy. The probation office assigned a level 26 to Carrozza’s drug trafficking activity. The government argued that this level should be increased by four levels to 30 because Carrozza was an organizer/leader of this activity and because the activity involved more than five participants. See U.S.S.G. § 3Bl.l(a). Because Carrozza’s narcotics activities yielded the highest offense level among Carrozza’s various offenses and, therefore, served as a starting point for the multiple count analysis under U.S.S.G. § 3D1.4, the difference was crucial. Apply-' ing the multiple count analysis, both the probation office and the government added five levels — the probation office arriving at a total offense level 31 and a guideline range of 151— 188 months, and the government calculating a total offense level 35 and a guideline range of 235-293 months. The probation office rejected the government’s request for a role in the offense adjustment for any of Carrozza’s offenses, contending that there was insufficient evidence as to Carrozza’s role in- the individual offenses to make such a determination. While the specifics with regard to Carrozza’s role in his bookmaking and extortion offenses are rather sketchy, the government provided more than sufficient evidence that Carrozza directed the narcotics activities of five or more participants. The government’s 104-page factual submission to the probation office is replete with evidence that Carrozza directed and organized the drug trafficking conspiracy. Given this evidence, it is understandable why Carrozza completely neglected to argue the propriety of a role in the offense adjustment in his appellate brief and below. Since it is Carrozza’s burden to establish that the district court’s error affected substantial rights, his failure to argue that a four-level role in the offense adjustment would have been improper, combined with the fact that an adjustment would have been appropriate at least with respect to the crucial narcotics conspiracy, undermines Carrozza’s claim of prejudice. To be sure, a role in the offense adjustment is a mixed question of law and fact. In most instances, an appellate court will not examine such questions in the first instance. We make the analysis only to indicate the unlikelihood that Carrozza was prejudiced by the district court’s failure to decide the issue. A further reason for finding no error affecting substantial rights is the significant benefits Carrozza received in exchange for his plea of guilty. In exchange for Carroz-za’s agreement to a 228-month sentence, the government promised not to prosecute Car-rozza for his alleged involvement in the murder of William Grasso — an offense that could carry a sentence of life imprisonment — and the attempted murder of Frank Salemme, Sr. In addition, the government promised not to subpoena Carrozza to testify in any federal grand jury investigation in the District of Massachusetts relating to the activities of the Patriarca Family. This later promise was probably significant to Carroz-za, who, as a made member of the Patriarca Family, had taken the oath of “omerta” to protect the secrets of the Patriarca Family of La Cosa Nostra to his grave. In light of the significant benefits Carrozza received from the plea agreement, it is difficult to see how he was prejudiced by the district court’s acceptance of his plea and sentencing him to a term of imprisonment upon which he had specifically agreed. See United States v. Ybabez, 919 F.2d 508, 510 (8th Cir.1990) (“We do not discern a miscarriage of justice when a defendant receives the sentence he bargained for in a plea agreement.”), cert. denied, 499 U.S. 940, 111 S.Ct. 1398, 113 L.Ed.2d 454 (1991). Finally, even if Carrozza were able to establish some form of prejudice from the district court’s failure to address his role in the offense and thereby pass the third and final hurdle of appellate authority under Fed. R.Crim.P. 52(b), the ease would be an inap *90 propriate one for us to exercise our discretion to recognize plain forfeited errors. The Supreme Court has made clear on numerous occasions that courts of appeals should correct plain forfeited errors affecting substantial rights only if the errors “‘seriously affect[ ] the fairness, integrity or public reputation of judicial proceedings.’ ” Olano, — U.S. at -, 113 S.Ct. at 1779 (quoting Atkinson, 297 U.S. at 160, 56 S.Ct. at 392). We see no such serious effect here. Carrozza failed to object in circumstances strongly indicative that he wished to accept the compromise sentence because of the benefits it conferred. The attendant circumstances do not reflect discreditably upon the fairness, integrity or public reputation of the proceeding. We vacate Patriarca’s sentence and remand for resentencing in accordance with this opinion. Carrozza’s sentence is affirmed. Footnotes [1] The district court's extensive sentencing memorandum is published at 807 F.Supp. 165 (D.Mass.1992). [2] The relevant conduct guideline, in pertinent part, provides the following: Unless otherwise specified, (i) the base offense level where the guideline specifies more than one base offense level, (ii) specific offense characteristics, (iii) cross references in Chapter Two, and (iv) adjustments in Chapter Three, shall be determined on the basis of the following: (1)(A) all acts and omissions committed, aided, abetted, counseled, commanded, induced, procured, or willfully caused by the defendant; and (B) in the case of a jointly undertaken criminal activity (a criminal plan, scheme, endeavor, or enterprise undertaken by the defendant in concert with others, whether or not charged as a conspiracy), all reasonably foreseeable acts and omissions of others in furtherance of the jointly undertaken criminal activity, *75 that occurred during the commission of the offense of conviction, in preparation for that offense, or in the course of attempting to avoid detection or responsibility for that offense.... U.S.S.G. § IB 1.3(a)(1). The relevant conduct guideline quoted here is the amended version of § IB 1.3, which became effective on November 1, 1992, after Patriarca was sentenced. Because the 1992 amendments to § IB 1.3 only clarify the previous version of the guideline, we may refer to the later version. See IB 1.11(b)(2) ("the court shall consider subsequent amendments, to the extent that such amendments are clarifying rather than substantive changes”). [3] The RICO guideline provides the following: § 2E1.1. Unlawful Conduct Relating to Racketeer Influenced and Corrupt Organizations (a) Base Offense Level (Apply the greater): (1) 19; or (2) the offense level applicable to the underlying racketeering activity. U.S.S.G. § 2E1.1. [4] Aside from its departure from the relevant conduct guideline, the district court’s interpretation could raise other problems. For example, in some circuits the government need not allege specific predicate acts when it charges a defendant with RICO conspiracy. See United States v. Glecier, 923 F.2d 496, 501 (7th Cir.), cert. denied, - U.S. -, 112 S.Ct. 54, 116 L.Ed.2d 31 (1991); United States v. Phillips, 874 F.2d 123, 127-28 (3d Cir.1989). A court sentencing a defendant in such a case would be put in a difficult position if forced to apply literally the district court’s analysis. Because such cases do not identify and charge the “underlying racketeering activity,” a court following the district court’s approach might be limited to the base offense level of 19 as specified in § 2E 1.1 (a)(1), even though the real offense conduct underlying the conspiracy is considerably more serious than other level 19 offenses. [5] Section 1961(1) defines “racketeering activity” to include, inter alia, "any act or threat involving murder, kidnaping, gambling, arson, robbery, bribery, extortion, dealing in obscene matter, or dealing in narcotic or other dangerous drugs, which is chargeable under State law and punishable by imprisonment for more than one year; ... [and] any act which is indictable under ... title 18 ... section 1512 (relating to tampering with a witness, victim, or an informant)...." [6] The training staff at the Sentencing Commission apparently agrees. In the latest issue of the Sentencing Commission's Most Frequently Asked Questions About the Sentencing Guidelines, Vol. VI, Dec. 1, 1992, Question 30 asks: "The defendant was convicted of RICO (18 U.S.C. § 1962). How is the alternative base offense level at § 2E1.1(a)(2) determined?” The answer provides the following: Application note 1 to § 2E1.1 instructs that where there is more than one underlying offense (i.e., predicate act), each underlying offense should be treated as if contained in a separate count of conviction for the purposes of subsection (a)(2). (See § lB1.2(d) and Application Note 5.) Each of the underlying offenses, whether or not charged in substantive counts of conviction, are treated as if they were substantive counts of conviction, or "pseudo counts.” Id. The training staff’s informational booklet states that "[t]he information does not necessarily represent the official position of the Commission, should not be considered definitive, and is not binding upon the Commission, the court, or the parties in any case." Because! IB 1.2(d), by its own terms, is not applicable to RICO convictions, we do not follow the training staff's suggestion. [7] Official comments made by the Sentencing Commission at the time that it enacted § IB 1.2(d) clarify the purpose of Application note 5: [Application note 5] is provided to address cases in which the jury’s verdict does not specify how many or which offenses were the object of the conspiracy of which the defendant was convicted. Compare U.S. v. Johnson, 713 F.2d 633, 645-46 (11th Cir.1983) (conviction stands if there is sufficient proof with respect to any one of the objectives), with U.S. v. Tamopol, 561 F.2d 466 (3d Cir.1977) (failure of proof with respect to any one of the objectives renders the conspiracy conviction invalid). In order to maintain consistency with other § IB 1.2(a) determinations, this decision should be governed by a reasonable doubt standard. A higher standard of proof should govern the creation of what is, in effect, a new count of conviction for the purposes of Chapter Three, Part D (Multiple Counts). Because the guidelines do not explicitly establish standards of proof, the proposed new application note calls upon the court to determine which offense(s) was the object of the conspiracy as if it were "sitting as a trier of fact.” The foregoing determination is not required, however, in the case of offenses that are grouped together under § 3D1.2(d) (e.g., fraud and theft) because § IB 1.3(a)(2) governs consideration of the defendant’s conduct. U.S.S.G.App. C„ para. 75, p.39 (Nov. 1, 1989). [8] Rather than merely requiring a defendant to agree to commit a substantive RICO offense, this circuit follows the minority rule, which requires that a defendant agreed to commit, or in fact committed, two or more specified predicate crimes as part of the defendant's participation in the affairs of the enterprise in order to convict the defendant for a RICO conspiracy. United States v. Boylan, 898 F.2d 230, 241 (1st Cir.), cert. denied, 498 U.S. 849, 111 S.Ct. 139, 112 L.Ed.2d 106 (1990); United States v. Winter, 663 F.2d 1120, 1136 (1st Cir.1981), cert. denied, 460 U.S. 1011, 103 S.Ct. 1249, 75 L.Ed.2d 479 (1983). This minority rule, however, does not affect the general premise that a RICO conspiracy is a single object conspiracy. [9] At oral argument, Patriarca contended that if the district court determines that Patriarca’s base offense level on the RICO counts is 43 (i.e., if the court decides that Patriarca is responsible for a murder and that a minimal role adjustment would be improper), then the court in applying U.S.S.G. § 5G1.2(d) should impose consecutive sentences, which could total up to 65 years (20 years for each of the two RICO counts and 5 years for each of the five Travel Act counts). See United States v. Masters, 978 F.2d 281, 284 (7th Cir.1992) (affirming district court sentencing RICO defendant to consecutive maximum sentences of 20 years on each count, for a total of 40 years, in order to come as close as possible to life imprisonment prescribed for level 43 offenses), cert. denied, - U.S. -, 113 S.Ct. 2333, 124 L.Ed.2d 245 (1993). At least one member of the panel believes that serious constitutional concerns may arise if the defendant ultimately receives the equivalent of a life sentence on the-ground of his connection with a murder for which he was never indicted, tried or convicted by a jury. However, the district court may yet ultimately sentence Patriarca to considerably less than 65 years. Because the district court has not yet sentenced Patriarca under the relevant conduct guideline as we interpret it today, any decision as to the constitutional implications, if any, "of a 65-year sentence would be premature. [10] We recognize that determining uncharged relevant conduct could sometimes impose tremendous additional burdens on a court. Relief may be afforded, however, in some instances by the fact that district courts need not make findings as to acts proffered as relevant conduct if the findings will not reflect the offense level. See TJ.S.S.G. § 3D1.4 (instructions on determining the combined offense level). [11] As to the remaining two proffered acts of relevant conduct — Caruana's marijuana importation and fugitive status — , Patriarca contends that the court's finding that such conduct warranted a criminal history upward departure and the fact that the indictment mentions drug trafficking in general, rather than particularly alleging marijuana importation, preclude consideration of this conduct as relevant conduct. Neither contention has merit. The court treated the Caruana allegations under the upward departure guideline, § 4A1.3, only because the court thoúght this uncharged conduct could not qualify as relevant conduct under § 1B1.3. Moreover, we think the indictment’s generic allegation of narcotics trafficking is sufficient to permit the court to consider marijuana importation as relevant conduct. [12] In its jurisdictional statement, the government questions whether this court has jurisdiction over the instant appeal. The government notes that a defendant may only appeal a sentence pursuant to a Rule 11(e)(1)(C) plea agreement on the grounds that the sentence was imposed in viola *87 tion of law or as a result of an incorrect application of the guidelines. 18 U.S.C. § 3742(c). In his jurisdictional statement, Carrozza asserts only that the sentence was in violation of the law. Because Carrozza provides no authority for the proposition that a claim such as he raises of procedural error in determining a sentence may rise to the level of a claim that the resulting sentence was imposed in violation of the law, the government argues that this court is without jurisdiction to consider the appeal. Regardless whether the district court's error rises to the level of a violation of law, Carrozza clearly argues in his brief that the district court failed to apply the guidelines correctly when it "assumed” a role in the offense adjustment. That is sufficient to give this court jurisdiction to decide this appeal. See United States v. Smith, 918 F.2d 664, 668-69 (6th Cir.1990) (upholding the right of a defendant to file a similar appeal under 18 U.S.C. 3742(a)(1) or (a)(2)), cert. denied, 498 U.S. 1125, 111 S.Ct. 1088, 112 L.Ed.2d 1192 (1991). [13] The government argues that Carrozza has waived his right to have the district court determine an actual guideline range by expressly agreeing to the district court's decision to assume a guideline range. A deviation from a legal rule is not considered an "error” if that legal rule has been waived, as opposed to merely forfeited. See Olano, - U.S. at -, 113 S.Ct. at 1777 ("Whereas forfeiture is the failure to make the timely assertion of a right, waiver is the intentional relinquishment or abandonment of a known right.”) (internal quotations omitted). We doubt that the sentencing guidelines can be waived. For example, we suspect that an agreement between the government and a defendant not to apply the guidelines would be ineffective. Because of doubts that the sentencing guidelines are waivable, we rest our decision today on Car-rozza's failure to establish that the district court’s error affects substantial rights, and on our discretion not to recognize plain errors even when they do affect substantial rights. [14] Fed.R.Crim.P. 32(c)(3)(D) provides in pertinent part the following: If the comments of the defendant and the defendant's counsel or testimony or other information introduced by them allege any factual inaccuracy in the presentence investigation *88 report or the summary of the report or part thereof, the court shall, as to each matter controverted, make (i) a finding as to the allegation, or (ii) a determination that no such finding is necessary because the matter controverted will not be taken into account in sentencing. Fed.R.Crim.P. 32(c)(3)(D) (emphasis added).