finding cocaine as relevant conduct where only heroin chargedSummary of this case from U.S. v. Sumner
ARGUED FEBRUARY 15, 1996
DECIDED MAY 29, 1996
Rodney Cubbie, Steven Ingraham, argued, Office of the United States Attorney, Milwaukee, WI, for Plaintiff-Appellee.
Frank M. Tuerkheimer, argued, University of Wisconsin Law School, Madison, WI, for Defendant-Appellant.
Appeal from the United States District Court for the Eastern District of Wisconsin.
No. 95 CR 1
Thomas J. Curran, Judge.
Before KANNE, ROVNER, and DIANE P. WOOD, Circuit Judges.
Following a bench trial, Salvador Acosta was convicted of one count of possessing heroin with intent to distribute it in violation of 21 U.S.C. § 841(a)(1) and 18 U.S.C. § 2. Acosta was sentenced to 151 months in prison followed by three years of supervised release, and was fined $2,500. In calculating Acosta's base offense level, the district judge determined that, in addition to the 31.826 grams of heroin confiscated from Acosta's car and residence on the day of his arrest, Acosta was also responsible for participating in a series of transactions during which he delivered a total of approximately 7.5 kilograms of cocaine to a government informant. Acosta challenges his sentence on the grounds that the district court did not make an express finding that the cocaine transactions were relevant conduct under U.S.S.G. sec. 1B1.3(a)(2), that the court failed to explain why it rejected a lower estimate of the amount of cocaine Acosta had allegedly delivered to the informant, and that in any event the informant's testimony concerning the amounts of cocaine involved was internally inconsistent and therefore unreliable. Although we find no error in the district court's decision to treat the cocaine transactions as relevant conduct for purposes of sentencing, we nevertheless vacate Acosta's sentence and remand for reconsideration of the amount of cocaine to be attributed to him.
We note that at certain points in the record on appeal, Acosta's first name is spelled "Salvator."
On the evening of December 17, 1994, members of the Milwaukee County Sheriff's Department stopped Acosta's car pursuant to a tip by government informant Carlos Johnson, who was Acosta's passenger at the time. A search of the car uncovered 1,449 packets of heroin containing a total of 29.586 grams of the drug. Each of the heroin packets was labeled "New York Express." On January 4, 1995, a federal grand jury returned a one-count indictment charging Acosta with possessing with intent to distribute heroin. Acosta waived his right to a jury trial, and a two-day bench trial resulted in his conviction. The district court then ordered the preparation of a presentence report and scheduled a sentencing hearing. The presentence report initially recommended that Acosta be held responsible for a total of 31.826 grams of heroin and 20 kilograms of cocaine, resulting in a base offense level of 34. Prior to the sentencing hearing, Acosta filed several objections to the presentence report, disputing both the inclusion of his alleged cocaine trafficking activities as relevant conduct under U.S.S.G. sec. 1B1.3(a)(2), and the amounts of cocaine attributed to him in those transactions. The government replied to Acosta's objections in a letter to the court, suggesting that only 7.5 kilograms of cocaine be attributed to Acosta, as this amount was a conservative estimate of the total quantity of cocaine Acosta had delivered to Johnson. The government's position as to the amount of cocaine attributable to Acosta was noted in a June 14, 1995 addendum to the presentence report. The addendum also contained the response of U.S. Probation Officer Bernadette M. Chesak, in which Chesak stated that she had no additional information to offer the court concerning the cocaine, and that the court would therefore need to resolve the issue of relevant conduct at sentencing.
A sentencing hearing was held on June 15, 1995. Turning to the question of relevant conduct, the district court considered Acosta's contention that the trial testimony of Johnson was "very ambiguous, very vague," that it failed to specify dates of transactions or total quantity of cocaine, and that the evidence as a whole failed to provide a sufficient basis for holding Acosta responsible for any amount of cocaine. (Sentencing Tr. at 6-7.) The government submitted that the fact of Acosta's cocaine dealing was amply corroborated by the statements of Carmen Andujar, Acosta's wife, who was questioned by police in conjunction with the search of Acosta's Milwaukee residence on the day of his arrest. (Sentencing Tr. at 8.) In those statements, Andujar admitted that she disposed of a quantity of cocaine when she heard that police were at her door, that she had seen Acosta with cocaine on a number of occasions, and that Acosta had told her that he was selling both heroin and cocaine. (Presentence Report at paras. 21-23.) Andujar also stated that Acosta obtained both drugs from a source in New York City, that she went to New York several times with Acosta to pick up heroin and cocaine, and that she wired money at Acosta's behest to his drug source in New York on at least two occasions. (Presentence Report at paras. 22-24.) In their search of Acosta's residence, police confiscated six Western Union money transfer receipts showing that Acosta had sent substantial sums of money from Milwaukee to New York. Police also seized a notebook that appeared to be a drug ledger, which Johnson subsequently identified as reflecting the money he owed Acosta for a number of packages of cocaine Acosta had provided to him for resale. (Presentence Report at paras. 17-18.) A dust mask with cocaine residue, a digital scale, and 9.106 grams of cocaine, as well as a total of fifty-two packets of heroin, each bearing the stamp "New York Express," were also recovered from the residence. (Presentence Report at paras. 14 16.)
In support of its position that a total quantity of 7.5 kilograms of cocaine should be attributed to Acosta for sentencing purposes, the government relied on the trial testimony of Johnson. Johnson testified that he began purchasing cocaine from Acosta in the early summer of 1994 and continued to do so until his arrest in November 1994, when he began cooperating with law enforcement authorities. Johnson stated that he had obtained cocaine from Acosta at least once per week during that period, and at times as frequently as every other day. He further testified that he had made between thirty and fifty purchases of cocaine from Acosta in 1994. To determine the total quantity involved, the district judge considered the following portion of the trial transcript of Johnson's testimony on direct examination, which the judge read into the record at the sentencing hearing:
Question: Okay. What is the greatest amount of cocaine that you obtained from Salvator Acosta?
Answer: Anywhere from I would say the max would be two and a half kilos.
The court: Are you talking about at one time?
Mr. Cubbie: I will clarify that, Judge.
Question: My question is at one time. I should have asked that way. What's the most you ever picked up from Salvator Acosta?
Answer: About two and a half kilos at the very most, max.
Question: What's the smallest amount of cocaine that you obtained from Salvator Acosta at one time?
Answer: Anywhere from a half ounce all the way up to a kilo of cocaine.
Question: Let me ask you a different way. What is the smallest amount — the smallest amount that you got from him at one time?
Answer: I would say nine ounces.
(Sentencing Tr. at 16.)
The district judge then determined that Johnson's testimony was generally credible, if lacking in precision, and that it constituted sufficient evidence to hold Acosta responsible for thirty deliveries of at least nine ounces of cocaine to Johnson, totaling approximately 7.5 kilograms of the drug. (Sentencing Tr. at 18-20.) With respect to whether Acosta's cocaine sales to Johnson were "relevant conduct" for purposes of applying U.S.S.G. sec. 1B1.3(a)(2), the judge noted that Johnson's testimony concerning the transactions was sufficiently corroborated by other evidence in the record to ensure its probable accuracy. He then adopted the factual findings in the presentence report pertaining to that evidence. (Sentencing Tr. at 18-20.) The judge did not, however, expressly state that Acosta's cocaine sales to Johnson were part of "the same course of conduct or common scheme or plan" (see U.S.S.G. sec. 1B1.3(a)(2)) as Acosta's heroin trafficking offense. Aggregating the amounts of drugs involved in Acosta's offense of conviction and the uncharged cocaine sales, the judge determined that a total drug quantity of 31.826 grams of heroin and 7.5 kilograms of cocaine was attributable to Acosta, corresponding to a base offense level of 32 under the Guidelines. The judge then added a two-level enhancement for obstruction of justice on the ground that Acosta had been untruthful with the court concerning his ability to speak and understand English. Acosta's total offense level was thus set at 34, with a criminal history category of I, yielding a sentencing range of 151 to 188 months. The judge imposed a sentence of 151 months in prison followed by three years of supervised release, and a fine of $2,500.
Acosta does not appeal the obstruction of justice enhancement.
In calculating a defendant's base offense level under the Sentencing Guidelines, a district court is required to take into consideration not only the types and quantities of drugs specified in the offense of conviction, but also any drugs that were "part of the same course of conduct or common scheme or plan as the offense of conviction." U.S.S.G. sec. 1B1.3(a)(2); see United States v. Howard, 80 F.3d 1194, 1202 (7th Cir. 1996); United States v. Beler, 20 F.3d 1428, 1431 (7th Cir. 1994); United States v. Duarte, 950 F.2d 1255, 1263 (7th Cir. 1991), cert. denied, 506 U.S. 859 (1992). When a defendant in a drug distribution case is sentenced on the basis of uncharged conduct pursuant to section 1B1.3(a)(2), the government's burden of proof is twofold: it must first prove by a preponderance of the evidence that the uncharged conduct bore the necessary relation to the offense of conviction, see Duarte, 950 F.2d at 1263-64; it must then establish the quantity of drugs involved in that conduct, also by a preponderance of the evidence, see Beler, 20 F.3d at 1431. Given the factual nature of this inquiry, a district court's determination that uncharged offenses were part of the same course of conduct as the offense of conviction is reviewed only for clear error. United States v. Townsend, 73 F.3d 747, 751 (7th Cir. 1996); United States v. Sykes, 7 F.3d 1331, 1335 (7th Cir. 1993). Similarly, the court's calculation of the drug amount attributable to the defendant is a finding of fact that will not be disturbed absent clear error. Townsend, 73 F.3d at 751; Beler, 20 F.3d at 1431.
Acosta first challenges the district court's implicit determination that the series of cocaine sales to Johnson occurring in the summer and fall of 1994 were relevant conduct. Acosta contends that, contrary to the requirements of Beler and Duarte, the district court never made an express finding that the cocaine sales were "part of the same course of conduct or common scheme or plan" as the later heroin offense, and that the facts alleged in the presentence report and at the sentencing hearing do not support this finding. Acosta further asserts that the court never resolved his challenge to the factual basis of the uncharged activities, and instead relied on disputed facts, such as Carmen Andujar's alleged statements to police, in treating the cocaine transactions as relevant conduct under sec. 1B1.3(a)(2). Although conceding that the district judge never expressly stated that the cocaine sales to Johnson were relevant conduct, the government maintains that the facts established in the presentence report, which the judge adopted prior to sentencing Acosta, provide a sound basis for concluding that the cocaine sales bore the necessary relationship to Acosta's offense of conviction.
This court has held that when a district court aggregates drug quantities stemming from uncharged or unconvicted relevant conduct for the purpose of calculating a defendant's base offense level, the court should "explicitly state and support, either at the sentencing hearing or (preferably) in a written statement of reasons, its finding that the unconvicted activities bore the necessary relation to the convicted offense." Duarte, 950 F.2d at 1263. Yet where it is clear from the record that the district court considered and adopted the facts recited in the presentence report, as well as the government's reasoning concerning the significance of those facts in establishing the defendant's responsibility for uncharged conduct, we have upheld the court's decision to treat the uncharged activities as relevant conduct despite the lack of an express finding that the activities were part of "the same course of conduct" or "common scheme or plan" as the convicted offense. See United States v. Thomas, 969 F.2d 352, 355 (7th Cir.), cert. denied, 506 U.S. 896 (1992); see also United States v. Taylor, 72 F.3d 533, 548-49 (7th Cir. 1995).
The transcript of Acosta's sentencing hearing reveals that the district judge reviewed the presentence report containing the statements Andujar made to police on the day of Acosta's arrest, and that he carefully considered whether the information in the report obtained from her statements was reliable. The judge also considered whether Johnson's testimony was corroborated by the information developed by police in their investigation of Acosta, also as reflected in the presentence report. He then credited Johnson's testimony, adopted the findings of the presentence report, and calculated the total quantity of drugs attributable to Acosta by including the cocaine transactions. It was thus clear from the proceedings that the judge considered those transactions to be relevant conduct under the applicable Guideline provision. As in Thomas and Taylor, we see no reason to set aside the district court's implicit determination that the cocaine sales to Johnson constituted relevant conduct solely because the judge did not expressly find that those transactions were part of the same course of conduct as Acosta's heroin offense. See Thomas, 969 F.2d at 355-56; Taylor, 72 F.3d at 549. We will therefore uphold the district court's decision to treat the cocaine transactions as relevant conduct under sec. 1B1.3(a)(2) provided the court had a factual basis supporting that decision.
In examining the court's factual basis for applying sec. 1B1.3(a)(2), we observe that the sentencing transcript belies Acosta's claim that the district judge never resolved the factual dispute surrounding Andujar's statements to police. Indeed, it is clear from the transcript that the judge regarded Andujar's denial that she ever made incriminating statements to police as highly suspect. (Sentencing Tr. at 18-19.) Although Acosta has maintained that his wife made no statement to police on the night of his arrest, the district judge was, of course, entitled to credit the police officers' version of events as described in the presentence report. See United States v. Salinas, 62 F.3d 855, 859 (7th Cir. 1995). Moreover, the judge considered the other evidence in the presentence report that pointed to Acosta's involvement in cocaine trafficking and found it to be reliable, also noting that the evidence corroborated in a general way Johnson's trial testimony concerning the cocaine he had obtained from Acosta. Although mindful of the fact that Johnson's testimony was at times quite vague, the judge nevertheless expressly credited it. (Sentencing Tr. at 18.) It is well established that a sentencing court's credibility determinations are accorded exceptional deference. United States v. Anaya, 32 F.3d 308, 314 (7th Cir. 1994); United States v. Kozinski, 16 F.3d 795, 820 (7th Cir. 1994). Acosta has failed, moreover, to point to any evidence in the record that would cause us to question the district court's reliance on Johnson's testimony for the purpose of establishing that Johnson purchased some quantity of cocaine from Acosta during the latter part of 1994. We thus discern no error in the district court's determination that Acosta trafficked in cocaine in the months immediately preceding his arrest on the heroin charge.
We next consider whether the evidence was sufficient to establish that these uncharged cocaine sales were part of "the same course of conduct or common scheme or plan" as Acosta's heroin offense. As the commentary accompanying the Guideline explains, "same course of conduct" and "common scheme or plan" are closely related concepts. U.S.S.G. sec. 1B1.3(a)(2), comment. (n. 9). Two or more offenses are part of a common scheme or plan when they are "substantially connected to each other by at least one common factor, such as common victims, common accomplices, common purpose, or similar modus operandi." Id., comment. (n. 9(A)). Offenses that are not part of a common scheme or plan may nevertheless be part of the same course of conduct if there is a sufficient relationship between them to support the conclusion that they are "part of a single episode, spree, or ongoing series of offenses." Id., comment. (n. 9(B)). In assessing whether two or more offenses are part of the same course of conduct, we consider factors that point to a strong relationship between the uncharged conduct and the convicted offense, focusing on whether the government has demonstrated a significant "similarity, regularity, and temporal proximity [between] the uncharged acts and the offense of conviction." Sykes, 7 F.3d at 1336; see United States v. Cedano-Rojas, 999 F.2d 1175, 1180 (7th Cir. 1993); Duarte, 950 F.2d at 1264-65; U.S.S.G. sec. 1B1.3(a)(2), comment. (n. 9(B)). When one of the factors, such as similarity, is relatively weak or absent, a stronger showing of regularity and temporal proximity will support a finding of relevant conduct. See U.S.S.G. sec. 1B1.3(a)(2), comment. (n. 9(B)) ("When one of the above factors is absent, a stronger presence of at least one of the other factors is required."); see also Sykes, 7 F.3d at 1336.
We agree with the government that the evidence relied on by the district court in calculating Acosta's sentence provides ample support for its implicit conclusion that the cocaine sales constituted relevant conduct in this case. Johnson's testimony and Andujar's statements to police support the court's finding that Acosta obtained both heroin and cocaine from a source in New York City, and that he purchased both drugs for resale at the same time. Close geographic and temporal proximity in activities that a defendant undertakes to advance his drug trafficking enterprise are factors that indicate the activities are part of the same course of conduct. See Taylor, 72 F.3d at 548-49. Moreover, although Acosta was not in possession of cocaine when he was arrested, the packets of heroin that were confiscated from him at the time of his arrest were labeled as having originated from New York City. This fact corroborated the testimonial evidence and the presentence report that Acosta regularly obtained all of his drugs from a source in New York. In addition, the physical evidence found at Acosta's residence, including "tools of the trade" such as a dust mask with cocaine residue and a digital scale, provided further support for the conclusion that Acosta was dealing in cocaine at the time of the heroin offense. Finally, the district court credited Johnson's testimony that he had obtained cocaine from Acosta in a series of transactions that took place at least once per week over a period of six or seven months prior to Acosta's arrest, thus indicating that the sales were part of an ongoing pattern of cocaine trafficking that Acosta engaged in at the time of his heroin offense. See United States v. Santiago, 906 F.2d 867, 872 (2d Cir. 1990) (nature, number and frequency of acts considered in determining whether pattern of behavior exists). In light of the government's strong showing of regularity, and the temporal and geographic proximity between the uncharged cocaine sales and Acosta's heroin offense, the mere fact that two different types of narcotics were involved is of little or no importance. The district court's implicit conclusion that Acosta's cocaine sales were relevant conduct for purposes of sentencing thus easily survives our deferential review.
Acosta's final challenge is to the court's calculation of the amount of cocaine attributable to him in setting his base offense level. Acosta maintains that the district judge did not base that calculation on evidence having sufficient indicia of reliability as required by due process and the Guidelines themselves. As this court and others have emphasized, the quantity of drugs attributed to a defendant in a narcotics trafficking case is frequently the single most important determinant of the length of the defendant's sentence under the Guidelines. See Beler, 20 F.3d at 1432; Duarte, 950 F.2d at 1262-63; United States v. Brothers, 75 F.3d 845, 849 (3d Cir. 1996); United States v. Sepulveda, 15 F.3d 1161, 1196-97, 1198 (1st Cir. 1993), cert. denied, 114 S.Ct. 2714 (1994); U.S.S.G. sec. 2D1.1(a)(3). It is therefore crucial that the information the district court relies on in calculating the defendant's base offense level have "sufficient indicia of reliability to support its probable accuracy." U.S.S.G. sec. 6A1.3(a); see Beler, 20 F.3d at 1432; Brothers, 75 F.3d at 848. Given the contradictory nature of the testimony the court relied on to attribute 7.5 kilograms of cocaine to Acosta, and the fact that there was neither an adequate explanation of why the court chose that particular amount, nor any other evidence in the record that specifically corroborated it, we are unpersuaded that Acosta's sentence was based on reliable information.
The district judge arrived at the figure of 7.5 kilograms by multiplying Johnson's estimate of the minimum number of cocaine purchases he had made from Acosta, with the quantity the judge apparently believed was the smallest amount of cocaine Johnson had obtained in a single purchase. (Sentencing Tr. at 19-20.) The judge's general approach in arriving at this conservative estimate of the total amount of cocaine attributable to Acosta is one that courts have considered acceptable. See United States v. Henderson, 58 F.3d 1145, 1151 (7th Cir. 1995); United States v. Clay, 37 F.3d 338, 344 (7th Cir. 1994); Beler, 20 F.3d at 1434; see also United States v. Hill, No. 95-5246, 1996 WL 137507, at *11 (6th Cir. Mar. 28, 1996) (when choosing between plausible estimates of drug amount, court should "err on the side of caution"); Sepulveda, 15 F.3d at 1198 (same). As Acosta points out, however, Johnson offered plainly inconsistent estimates concerning the size of his smallest purchase of cocaine. On direct examination, Johnson first testified that the smallest amount of cocaine he obtained from Acosta in a single purchase ranged from one-half ounce to a kilogram. Immediately thereafter, in response to the same question posed in a slightly different manner, Johnson reconsidered his estimate and stated that the smallest amount was nine ounces. (Sentencing Tr. at 16.) The judge then used the nine ounce figure as the minimum quantity of cocaine and multiplied it by thirty, that being the minimum number of purchases Johnson made from Acosta between the early summer and late fall of 1994. On the basis of this calculation, to which the court added the 31.826 grams of heroin that Acosta was convicted of possessing, the judge set Acosta's base offense level at 32. (Sentencing Tr. at 20.) There was, however, no explanation of why the judge rejected the one-half ounce figure as the smallest amount purchased, a quantity that would have resulted in a significantly lower base offense level of 26, prior to adjustment for obstruction of justice. See U.S.S.G. sec. 2D1.1(c) comment. (n. 10). Yet as we observed in Duarte, "when the court clearly relies upon one of two contradictory statements offered by a single witness, it should directly address the contradiction and explain why it credits one statement rather than the other." 950 F.2d at 1266.
Although Acosta also challenged the estimate of the minimum number of times Johnson obtained cocaine from him during the summer and fall of 1994 because Johnson gave a wide-ranging estimate of thirty to fifty purchases, we note that the minimum of thirty purchases is fully consistent with Johnson's testimony concerning the frequency of purchases he made during that time frame. (Sentencing Tr. at 17-18.) As such, it was not clear error for the judge to consider that estimate sufficiently reliable to use as a basis for calculating the total drug amount.
Nor does the sentencing transcript reveal that the judge considered any corroborating evidence, such as that obtained in the search of Acosta's home, or that could be inferred from Andujar's description of Acosta's cocaine trafficking activities, that would serve to establish the greater plausibility of the nine ounce estimate, as opposed to the one-half ounce estimate. Although, as explained above, this evidence generally corroborated the fact of Acosta's cocaine sales to Johnson, it did not provide a basis for arriving at a particular quantity of cocaine for purposes of applying U.S.S.G. sec. 2D1.1(a)(3). See Beler, 20 F.3d at 1436-37 (fact that defendant made numerous drug sales does not, by itself, corroborate specific estimates of drug quantity); United States v. Miele, 989 F.2d 659, 668 (3d Cir. 1993) (same); see also Henderson, 58 F.3d at 1152 (court may not choose random number to establish drug quantity simply because it believed more drugs were involved than that number indicates). In the absence of an explanation or other evidence justifying the judge's choice of the nine ounce amount, we conclude that the calculation of Acosta's base offense level rests on an inadequate evidentiary basis, and as such is clearly erroneous.
As was pointed out at oral argument, nine grams (rather than ounces) is equal to approximately one-third of an ounce, which is closer to Johnson's low estimate of one-half ounce as the smallest amount of cocaine he ever obtained from Acosta in a single purchase. The district court may wish to explore this possibility on remand.
Relying on United States v. Mustread, 42 F.3d 1097 (7th Cir. 1994), and United States v. Isirov, 986 F.2d 183 (7th Cir. 1993), the government maintains that we should affirm Acosta's sentence because Acosta did not meet his burden of producing some evidence to show that the presentence report was unreliable or inaccurate, and instead stood on his bare denial of the facts contained in the presentence report. See Mustread, 42 F.3d at 1101-02; Isirov, 986 F.2d at 185-86. The government's reliance on this line of cases is misplaced. As is evident from the June 14, 1994 addendum to the presentence report, the probation office took no position as to the quantity of cocaine attributable to Acosta, and U.S. Probation Officer Chesak specifically noted that the district court would be required to make that determination at sentencing. Moreover, Acosta's challenge to the court's calculation does not rest on a bare denial of any facts contained in the presentence report, but on an inherent contradiction in the testimony that the sentencing court used to determine Acosta's base offense level. As this court recently explained in United States v. Lanterman, 76 F.3d 158 (7th Cir. 1996), because a defendant has a due process right to be sentenced on the basis of reliable information, "unreliable allegations must not be considered." Id. at 160-61 (emphasis in original). Thus, it is only when the facts alleged in a presentence report are shown to bear "sufficient indicia of reliability to support [their] probable accuracy" (U.S.S.G. sec. 6A1.3(a)) that the burden of producing some evidence to demonstrate that those facts are nevertheless incorrect or otherwise unreliable is shifted to the defendant. See id.; see also Townsend, 73 F.3d at 751-52; Salinas, 62 F.3d at 859. Yet evidence that is contradictory on its face is perhaps the prototype of unreliable evidence, and has been repeatedly recognized by this court as providing an inadequate basis for calculating a defendant's base offense level under the Guidelines. See, e.g., Beler, 20 F.3d at 1436; Duarte, 950 F.2d at 1266. Because the testimony relied on by the sentencing court was inherently inconsistent, and Acosta apprised the court that the testimony contained ambiguities, it is untenable to suggest that Acosta had a further obligation to produce material evidence to show that the estimate used to calculate the total amount of cocaine he delivered to Johnson was inaccurate. We remind the government that on remand, it bears the responsibility of producing some evidence from which the district court may make a reliable estimate of the quantity of cocaine that should be attributed to Acosta for purposes of determining his sentence. See Howard, 80 F.3d at 1205; Beler, 20 F.3d at 1431.
In this respect, Lanterman is, of course, fully consistent with Mustread and Isirov. These cases provide no support whatsoever for weakening or dispensing with the requirement that a defendant be sentenced only on the basis of reliable information. See Mustread, 42 F.3d at 1102 (district court's close examination of facts recited in presentence report established that they were not "naked or unsupported charge[s]"); Isirov, 986 F.2d at 186 (district court determined that facts in presentence report were reliable before rejecting defendant's bare denial of those facts).
For the foregoing reasons, we VACATE Acosta's sentence and REMAND for resentencing.