United States of America, Plaintiff-appellee, v. Octavio Carreon, and Armando Melendez, Defendants-appellants, 11 F.3d 1225 (5th Cir. 1994)

The central legal issue in this appeal is whether "relevant conduct" as defined in Sec. 1B1.3(a) (1) (B) of the Sentencing Guidelines includes conduct occurring before a criminal defendant joins a conspiracy.

Defendant-Appellants Armando Melendez and Octavio Carreon were convicted of conspiring to import more than 100 kilograms of marihuana in violation of 21 U.S.C. §§ 952, 960, and 963. Carreon was also convicted of conspiring to possess with intent to distribute more than 100 kilograms of marihuana in violation of 21 U.S.C. §§ 841 and 846, and of bribing a public official in violation of 18 U.S.C. § 201. Carreon questions whether he can be sentenced under U.S.S.G. 1B1.3(a) (1) (B) for conduct occurring before he joined the conspiracy; Melendez disputes whether the district court's failure to make explicit findings warrants remand for resentencing. Melendez further challenges the district court's failure to provide him with exculpatory or impeachment information contained in the Presentence Investigation Reports ("PSRs") of coconspirators turned government witnesses.

We hold today that the "reasonable foreseeability" requirement contained in U.S.S.G. 1B1.3(a) (1) (B) is prospective only, and consequently cannot include conduct occurring before the defendant joined the conspiracy. We also conclude that the district court's failure to make explicit findings for either Melendez or Carreon requires us to "second guess" the basis of the district court's sentencing as to both of these defendants. Consequently, we must reverse and remand both sentences for findings and resentencing. Finally, we conclude that, in light of United States v. Jackson, admittedly rendered after the instant trial and sentencing, the district court's failure to review the Presentence Investigation Reports ("PSRs") of government witnesses requires remand of Melendez's conviction to determine whether those PSRs contained any material exculpatory or impeachment information and, if they did contain such information, whether failure to provide it was harmless error.I

FACTS AND PROCEEDINGS

A procession of coconspirators turned government witnesses¹  testified that Armando Melendez and his father, Jesus "Chuy" Melendez, were involved in an extensive marihuana trafficking conspiracy between 1985 and 1992. These witnesses testified to purchasing and transporting extensive quantities of marihuana acquired from Armando Melendez and his family; through the years this marihuana was transported in everything from vans to trucks to planes to avoid detection. Indeed, several of these witnesses were caught during this period while attempting to smuggle marihuana for the Melendezes.² 

In 1989 Chuy Melendez, the patriarch of the clan, was murdered. By the end of 1989 a change had occurred in this drug trafficking scheme. Defendant-Appellant Octavio Carreon joined with Armando Melendez and others to bribe U.S. Border Patrol Agent Patrick Maynes--who was a childhood friend of Carreon--to provide Melendez and Carreon with information on law enforcement activity in the area where they smuggled.³  Maynes was working undercover, however, and spent the next year and one-half gathering evidence during meetings with Melendez, Carreon, and other members of the conspiracy.

The Melendez-Carreon drug smuggling business terminated in April 1992. During April, Maynes met several times with various members of the conspiracy and agreed to transport marihuana. Maynes eventually transported 327 pounds of marihuana to Albuquerque on April 12, where it was unloaded, moved to a residence, and then seized by police. On April 16, government agents moved in and seized corroborating evidence from the residences of Melendez and Carreon.⁴ 

Both Carreon and Melendez were indicted, along with thirteen others, in May 1992 for conspiring to import and conspiring to possess more than 100 kilograms of marihuana.⁵  The indictment alleged that this conspiracy operated from the beginning of January 1985 to the beginning of May 1992. Carreon and Melendez were also indicted for bribing a public official.⁶ 

In pretrial motions, Carreon filed for and was granted his request to be severed and tried only with Melendez. Melendez filed a pretrial motion to adopt all pretrial motions filed by his codefendants, which included a motion by codefendant Christopher Peter Bush requesting access to the PSRs of the government witnesses. The district court denied Bush's motion. On the day of trial, Melendez made his own separate motion requesting access to the PSRs of all government witnesses. The district court dismissed this request during trial as "being moot."

The jury found both Melendez and Carreon guilty of the conspiracy to import more than 100 kilograms of marihuana, but found only Carreon guilty of the conspiracy to possess. Carreon was also the only one found guilty of the bribery charge.

In sentencing the defendants, the district court accepted the drug quantity findings of 131,358 kilograms contained in the PSRs of Melendez and Carreon. These findings attributed all marihuana discussed at trial during the whole course of the conspiracy--from 1985 to 1992--to both Melendez and to Carreon. The 131,358 kilograms established a base offense level of 40 for both defendants; the district court adjusted this score upward two points as to Melendez because he had a supervisory role in the conspiracy, and downward two points as to Carreon because he was a minor participant.⁷ 

Both defendants were sentenced within the ranges established by the Sentencing Guidelines. Carreon was sentenced to 235 months imprisonment on the conspiracy counts and a concurrent 180-month term on the bribery count. Melendez was sentenced to 360 months imprisonment on the conspiracy count. Both were sentenced to five years of supervised release. In addition, Carreon received a $25,000 fine while Melendez received a $100,000 one. Both defendants timely appealed.

II

ANALYSIS

A. Relevant Conduct Under the Sentencing Guidelines

Under Sec. 2D1.1(a) (3) of the Sentencing Guidelines, the offense level of a defendant convicted of a drug trafficking offense is determined by the quantity of drugs involved in the offense. This quantity includes both drugs with which the defendant was directly involved, and drugs that can be attributed to the defendant in a conspiracy as part of his "relevant conduct" under Sec. 1B1.3(a) (1) (B) of the Guidelines. Relevant conduct for conspiratorial activity is defined in Sec. 1B1.3(a) (1) (B) as "all reasonably foreseeable acts and omissions of others in furtherance of jointly undertaken criminal activity."⁸  We made clear in United States v. Evbuomwan⁹  that for conspiratorial conduct to be attributed under Sec. 1B1.3(a) (1) (B), that conduct must be both "reasonably foreseeable" to the defendant and within the scope of the defendant's agreement.

Both Melendez and Carreon challenge the district court's calculation of their sentence under Sec. 1B1.3, albeit on different grounds. How the Guidelines are to be interpreted is a question of law, which we review de novo.¹⁰  Factual findings under the Guidelines are reviewed only for clear error.¹¹ 

Melendez argues that the district court erred in failing to make explicit findings as to how it calculated the drug quantities attributed to him under Sec. 1B1.3. Specifically, Melendez contends that the district court's failure to find 1) that the drugs attributed to him in the PSR were within the scope of and in furtherance of a conspiracy that he joined, and 2) that all of the marihuana quantities included in the PSR were reasonably foreseeable by him, warrants remand for findings and resentencing. In the instant case, the district court simply adopted the PSR report, including the quantity attributed from the conspiracy, as "well-justified." The government contends that this adoption constitutes an implicit finding sufficient to establish both that Melendez was a member of the conspiracy, and that the quantities at issue were reasonably foreseeable.

In United States v. Puma¹²  we remanded for resentencing because the district court had failed to make an express finding that the conspiratorial activity at issue was reasonably foreseeable as required by former Sec. 2D1.4 of the Guidelines--the same type of finding as required here by Sec. 1B1.3(a) (1) (B). Moreover, Rule 32 of the Federal Rules of Criminal Procedure mandates that the sentencing court make findings regarding any controverted facts in the PSR, or state that those facts will not be taken into account in sentencing.¹³  We have nevertheless rejected the proposition that a court must make a "catechismic regurgitation of each fact determined"; instead, we have allowed the district court to make implicit findings by adopting the PSR.¹⁴  This adoption will operate to satisfy the mandates of Rule 32 when the findings in the PSR are so clear that the reviewing court is not left to "second-guess" the basis for the sentencing decision.¹⁵ 

Here, the PSR simply accepts virtually all of the drug quantities discussed in the testimony at trial, and attributes those quantities to Melendez. In response to Melendez' voluminous objections to those quantities--including specific objections to the reasonable foreseeability of many of the transactions and events involving those quantities--the probation officer preparing the report stated that he was submitting the matter "for the Court's consideration."

The problem here is that we are unable to determine how the district court resolved these issues. For example, the government's theory supporting the foreseeability finding--that Melendez was the "key man" in the smuggling organization, and hence was aware of all the transactions engaged in by that organization--was at least partially rejected by the court at the sentencing hearing when it rejected the PSR finding that Melendez was the organizer or leader of the organization.¹⁶  In rejecting this finding, the court stated:

As to the role in the offense, I think it's somewhat exaggerated to characterize [Melendez] as the mastermind or as the organizer or leader. He was a key member, of course, and occupied some supervisory role, but I think a two-level adjustment rather than four would be more appropriate under the evidence I heard in the case.

Although this statement appears to resolve the "jointly agreed to" issue by concluding that Melendez was a member of the conspiracy in some capacity, it appears to cloud the foreseeability issue under the government's theory of the case. We thus are left to second guess the basis for the district court's calculation here and must consequently remand for findings.¹⁷ 

2. Reasonable Foreseeability and Prior Conduct

Carreon contends that he cannot be sentenced under Sec. 1B1.3(a) (1) (B) for conduct occurring before he joined the conspiracy. Simply put, Carreon insists that the requirement of "reasonable foreseeability" contained in Sec. 1B1.3(a) (1) (B) cannot be backward looking to include prior conduct. Before we address this contention, we pause to note that Carreon's sentence suffers from the deficiency that infected the sentencing of Melendez. The district court followed the same approach as it had in sentencing Melendez by concluding that the drug quantities listed in the PSR of Carreon--which are virtually identical to the ones listed in the PSR of Melendez--were "well justified."

On appeal, Carreon attempts to limit the scope of his agreement (and hence his sentencing accountability) by contending that the instant case contains at least two conspiracies: the bribery-importation conspiracy involving Maynes, and an earlier importation conspiracy centered around Melendez and the government witness, Charles Aragon. Carreon concedes he was a member of the Maynes conspiracy but argues that he had no involvement in the earlier Melendez-Aragon conspiracy. The government counters that there was only one conspiracy here--with Melendez as the "key man" tying all the disparate transactions and actors together--and that, consequently, the scope of Carreon's agreement could extend to all activity engaged in by that conspiracy. The simple adoption of Carreon's PSR by the district court unfortunately does not disclose what the court found regarding this issue.

Carreon and the government also hotly contest whether Carreon entered the conspiracy (or conspiracies) in 1987 or in 1989, but, given the absence of findings, we cannot discern what the district court found regarding this issue either. The parties contest this issue for different reasons though: as noted above, Carreon contends that relevant conduct as defined in Sec. 1B1.3(a) (1) (B) cannot include conduct occurring before he joined the conspiracy. The drugs attributed to Carreon by the PSR obviously include quantities arising out of transactions that occurred before he joined the conspiracy, as they include transactions that occurred from 1985 to 1992. Accordingly, Carreon's sentence is presently premised on the notion that relevant conduct can include conduct occurring before he joined any conspiracy. As such, the "relevant conduct-prior conduct" issue is a necessary part of the sentence challenged and is thus properly reviewable.¹⁸  As this issue is res nova in this circuit,¹⁹  we begin by turning to the applicable provisions in the Sentencing Guidelines.

a. The Sentencing Guidelines

The appropriate starting point is the plain language of the Guidelines themselves.²⁰  Both the Chairman and the General Counsel of the Sentencing Commission (the "Commission") have referred to the relevant conduct section as the "cornerstone" of the Guidelines.²¹  Including relevant conduct in sentencing allows the defendant to be sentenced based, at least in part, on the seriousness of his actual behavior.²²  Tying the sentence to the seriousness of the actual offense in turn helps effectuate the congressional policy of proportionality--that the "system impose [ ] appropriately different sentences for criminal conduct of differing severity."²³

Relevant conduct under Sec. 1B1.3 includes acts engaged in by coconspirators--a particularly important provision in drug conspiracy prosecutions given that the sentences in such prosecutions are determined based on drug quantities attributable to the defendant,²⁴ and that in many instances the defendant will not be directly involved in many of the transactions giving rise to those quantities. Nonetheless, the Guidelines have retained their emphasis on sentencing the defendant as an individual: The Commentary to Sec. 1B1.3 states that "the focus is on the specific acts and omissions for which the defendant is to be held accountable in determining the applicable guideline range, rather than on whether the defendant is criminally liable for an offense as a principal, accomplice, or conspirator."²⁵

Determining the appropriate degree of sentencing accountability for conspiratorial activity--while still serving the need for proportionality in sentencing and for sentencing the defendant as an individual--has proven difficult for the Commission. A review of the changes effected by the Commission sheds some light on the Commission's intent relative to these issues.

The original Guidelines contained only a cryptic reference to accountability for conspiratorial activity.²⁶  Perhaps recognizing the weaknesses of this approach, the Commission promulgated an emergency amendment that became effective January 15, 1988.²⁷  This amendment referred to conspiratorial accountability as conduct "for which the defendant would be otherwise accountable," and defined such conduct to include "conduct in furtherance of the conspiracy that was known to or was reasonably foreseeable by the defendant."²⁸

In an amendment effective November 1, 1989, the Commission dropped the "known to" language, thus including conspiratorial conduct only if that conduct was "reasonably foreseeable" to the defendant.²⁹ This amendment also attempted to distinguish the individual from the conspiracy as a whole by stating:

Because a count may be broadly worded and include the conduct of many participants over a substantial period of time, the scope of the jointly-undertaken criminal activity and hence relevant conduct, is not necessarily the same for every participant. Where it is established that the conduct was neither within the scope of the defendant's agreement, nor was reasonably foreseeable in connection with the criminal activity the defendant agreed to jointly undertake, such conduct is not included in establishing the defendant's offense level under this guideline.³⁰ 

In its most recent amendment, which became effective November 1, 1992 (the 1992 Amendment), the Commission once again attempted to clarify and to delineate the extent of liability for conspiratorial activity. The Commission first moved the "reasonable foreseeability" element into the text of the guideline.³¹ Next, the Commission rewrote its commentary to emphasize that the "scope of the agreement" and "reasonable foreseeability" are independent and necessary elements. The second note to the new Commentary now provides in pertinent part:

In the case of 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.³² 

Finally, the Commission emphatically rejected the notion that criminal liability and sentencing accountability are coextensive. The first note of the Commission's new Commentary now states:

The principles and limits of sentencing accountability under this guideline are not always the same as the principles and limits of criminal liability. Under [Sec. 1B1.3] the focus is on the specific acts and omissions for which the defendant is to be held accountable in determining the applicable guideline range, rather than on whether the defendant is criminally liable for an offense as a principal, accomplice, or conspirator.³³ 

In sum, changes wrought by the Commission in Sec. 1B1.3 indicate that the Commission has developed "reasonable foreseeability" as a means to limit the sentencing accountability of a defendant. The Commission has apparently developed this limit to ensure that the congressional goal of proportionality is served--that the sentence of the defendant is indeed tied to the "specific acts and omissions for which the defendant is to be held accountable."³⁴ 

b. Circuit Law and "Reasonable Foreseeability"

Like the Commission, those circuits which have addressed this issue have struggled to ascertain the appropriate limits of sentencing accountability for conspiratorial activity. Specifically, other circuits have disagreed on whether "reasonable foreseeability" under Sec. 1B1.3 may include conduct occurring before a defendant joins a conspiracy. In two opinions predating the 1992 Amendments, United States v. Miranda-Ortiz³⁵  by the Second Circuit and United States v. Edwards³⁶  by the Seventh, both courts concluded that "reasonable foreseeability" may include prior conduct. In contrast, in two opinions following³⁷  that amendment, United States v. O'Campo³⁸  by the First Circuit and United States v. Petty³⁹  by the Ninth, those courts flatly rejected the proposition that reasonable foreseeability can be backward-looking to encompass prior conduct.

In Miranda-Ortiz the Second Circuit simply assumed without discussion that prior conduct could be reasonably foreseeable under the Guidelines, although the court remanded that case for resentencing to determine whether the defendant actually knew, or should have known, of the drug quantities involved in the conspiracy before he joined.⁴⁰  In Edwards the Seventh Circuit concluded that past conduct can be "reasonably foreseeable" by analogizing to cases in which defendants were held liable for large quantities of drugs based on their degree of participation in the conspiracy.⁴¹  These cases, however, either did not address the prior conduct issue or involved drug transactions that occurred after the defendant joined the conspiracy.⁴²  The Edwards court justified these analogies by postulating that reasonable foreseeability incorporated the limiting principle contained in substantive conspiracy law, namely, that the scope of liability is determined by the scope of the agreement.⁴³ 

The approach used by the Seventh Circuit in Edwards is troubling. That approach--which treats the scope of the agreement as defining the extent of what is reasonably foreseeable--contradicts our requirement in Evbuomwan (and the Guidelines requirement in the 1992 Amendment) that the sentencing court must find both that the defendant agreed to the joint criminal activity, and that the agreed-to activity must also be reasonably foreseeable. The Edwards approach is also inconsistent with the clear statement in the Guidelines' Commentary that sentencing accountability and criminal liability are not necessarily coextensive.⁴⁴

In contrast to the Edwards approach, the First Circuit in O'Campo treated reasonable foreseeability as a separate element and then concluded that it would be oxymoronic to conclude that "prior conduct" could be "reasonably foreseeable."⁴⁵  The Ninth Circuit in Petty agreed with the O'Campo court's rejection of Edwards by observing first that the concept of relevant conduct contained in the Guidelines repudiates "the general [ly] harsh law of conspiracy that a conspirator is 'bound by all that has gone on before in the conspiracy,' "⁴⁶  and second that courts do not have a warrant to turn the notion of foreseeability 180 degrees to apply it to prior conduct.⁴⁷ 

c. Resolution

We conclude that the approach used by the First and Ninth Circuits in O'Campo and Petty is the better reasoned one, as it more closely follows the dictates of the Guidelines.⁴⁸  Applying foreseeability only prospectively comports with the plain meaning of the term "foreseeable."⁴⁹  It also effects the individualized sentencing polices underlying the Guidelines "relevant conduct" provision: that sentencing is focused on "the specific acts and omissions for which the defendant is to be held accountable."⁵⁰ Such an approach ensures that the congressional policy of proportionality is served, as the sentence of the defendant is tied to the severity of the defendant's conduct.⁵¹  Finally, the approach used in O'Campo and Petty, unlike the one in Edwards, is consistent with the requirement we imposed in Evbuomwan and the Commission imposed in their 1992 Amendments--that the "scope of the agreement" and "reasonable foreseeability" are independent and necessary elements of relevant conduct under Sec. 1B1.3(a) (1) (B). We thus hold today that "relevant conduct" as defined in Sec. 1B1.3(a) (1) (B) is prospective only, and consequently cannot include conduct occurring before a defendant joins a conspiracy.

Without attempting to pretermit the district court's findings on remand, we observe that the instant case aptly illustrates the dangers of failing to limit temporally the sentencing accountability of a defendant. Both Carreon and Melendez were held responsible for the same quantity of drugs--quantities that arose from transactions occurring from 1985 to 1992. Yet according to the government's own admission, Carreon--unlike Melendez--was a late comer to the Melendez drug smuggling organization.⁵²  Basing Carreon's sentence on the same quantity of drugs as Melendez' would thus countermand the congressional goal of proportionality in sentencing--an important goal that, as the Sentencing Commission instructs, is met by focusing the sentencing of a defendant on his individual acts and omissions, not his criminal liability under the substantive law of conspiracy.⁵³ 

As Carreon's presently constituted sentence is predicated on attribution of conduct occurring before he joined the Melendez conspiracy, we are constrained to remand for resentencing. We require that in resentencing Carreon the district court determine: 1) when Carreon joined the conspiracy (or conspiracies), 2) what drug quantities were within the scope of Carreon's conspiratorial agreement (or agreements),⁵⁴  and 3) of these drug quantities, which were reasonably foreseeable--prospectively only--by Carreon. We observe that evidence of prior conduct is not wholly irrelevant to these inquiries; rather, any knowledge that Carreon may have had regarding prior conduct will provide useful evidence of what Carreon agreed to and what he reasonably foresaw when he joined the conspiracy. The O'Campo court accurately explains when, and to what extent, such evidence of prior conduct is still relevant:

We have emphasized that proving a "defendant knew what was going on" is not sufficient to establish co-conspirator criminal liability.... We similarly believe that the base offense level at sentencing cannot be based on mere knowledge of historic facts ... [T]he relevant inquiry is to determine the foreseeable object to which the individual conspirator agreed. The past performance of the conspiracy, known to the late-joining co-conspirator, may be relevant to a careful analysis directed to the understanding such a co-conspirator has of the anticipated quantity of the drug distributions the conspiracy intends to undertake after he joins....

[T]he base offense level of a co-conspirator at sentencing should reflect only the quantity of drugs he reasonably foresees it is the object of the conspiracy to distribute after he joins the conspiracy. In making the judgment what a co-conspirator can "reasonably foresee," the earlier transactions of the conspiracy before he joins but of which he is aware will be useful evidence. However, a new entrant cannot have his base offense level enhanced at sentencing for drug distributions made prior to his entrance merely because he knew they took place. Thus for example, if a defendant joins a conspiracy he knew previously engaged in distribution of extraordinarily amounts of drugs, his knowledge of those prior acts will inform the judgment about what he reasonably could have foreseen the conspiracy would entail in the future. He might, for example, then be found reasonably to have foreseen future distributions of similarly large quantities.⁵⁵ 

The O'Campo court's explanation adequately describes our holding today.

Finally, in order to delineate properly the extent of our holding today, we address a valid concern raised by the government. Perhaps because of the 1992 Amendments to the Guidelines, the government is no longer advocating the inclusion of pre-joinder occurrences as relevant conduct as it was under Edwards. Nonetheless, the government remains concerned that, through temporal fortuity or manipulation, a defendant may avoid sentencing accountability for his conduct. The government poses the following hypothetical:

Several persons agree to import a large quantity of drugs into the United States. After the drugs have been imported and distributed, the defendant--with full knowledge of the illegal objectives of the conspiracy--agrees to join the conspiracy to collect the money due for those drugs. The conspiracy ends--either because the members were caught or because they decided to disband--after completion of this transaction.

The government posits that a "bright line" prospective-only rule for relevant conduct would allow this hypothetical defendant to escape sentencing accountability for his conduct: all importation and distribution predated his joinder, and no drug quantities, post-joinder, are attributable to the defendant's collection function.

Although the government has indeed raised a valid concern, we observe that the government would be reading our holding today too broadly if it believes that we have foreclosed all sentencing accountability for its hypothetical defendant under Sec. 1B1.3. Section 1B1.3 includes two primary grounds on which to hold a defendant accountable for conduct by others: subsection (a) (1) (B) for jointly undertaken criminal activity,⁵⁶ and subsection (a) (1) (A) for "all acts and omissions committed, aided, abetted, counseled, commanded, induced, procured, or willfully caused by the defendant."⁵⁷ Subsection (a) (1) (B) includes a "reasonable foreseeability" limitation; subsection (a) (1) (A) does not.⁵⁸ Furthermore, subsections (a) (1) (A) and (a) (1) (B) are separate and independent grounds for imposing sentencing accountability.⁵⁹

Consequently, nothing we hold today regarding the "reasonable foreseeability" limit to subsection (a) (1) (B) would foreclose the government from holding its hypothetical defendant accountable under subsection (a) (1) (A). Clearly, such a defendant has "aided or abetted" a drug smuggling transaction when his agreed-to function is so causally and functionally intertwined with the transaction that it could not be completed absent his participation.⁶⁰  As such, that hypothetical defendant could be held accountable for that transaction under subsection (a) (1) (A).

B. Review of PSRs of Coconspirators-Witnesses

Most of the government witnesses testifying against Melendez were coconspirators in the Melendez organization who had themselves been subjected to various criminal charges. Melendez requested access to the PSRs of these witnesses in order to acquire any exculpatory or impeachment information under Brady and Giglio. Working from the established premise that information contained in a PSR is confidential, the district court denied this request.

Subsequent to the trial in the instant case, we decided United States v. Jackson,⁶¹  in which we reconciled the need to protect the confidentiality of PSRs with the right of the defendant to have access to exculpatory and impeachment evidence under Brady and Giglio. We accomplished this reconciliation in Jackson by holding that the trial court should examine the PSR in camera and release any exculpatory or impeachment evidence to the defendant, while protecting the confidentiality of the rest of the PSR.⁶² 

Particularly in light of the absence of the coconspirators' PSRs from the record on appeal, the government concedes that we must remand here in light of Jackson. But the government contends that we must remand for "harmless error" analysis by the district court. Melendez, not surprisingly, argues that the district court's failure to anticipate Jackson warrants not just vacatur of his sentence, but also reversal of conviction and remand for a new trial. We agree with the government.

The remand-and-review approach is consistent with Rule 52's mandate that " [a]ny error [s] ... not affect [ing] substantial rights shall be disregarded"⁶³  and our requirement that the defendant must show that the requested evidence was "material" before he can claim any infringement of a substantial right under Brady or Giglio.⁶⁴  In short, until the district court inspects the PSRs and compares them against the information Melendez had at trial, it is impossible to determine whether Melendez was in fact denied access to material Brady or Giglio information, and, if so, whether he suffered prejudice as a result of this denial.

Consequently, we remand to the district court to 1) conduct an in camera inspection and make appropriate findings as to whether the PSRs of the government witnesses contained any material Brady or Giglio information, and 2) compare those findings against the evidence Melendez had at trial to determine whether the failure to provide this information was harmless error. So that these findings and conclusions are reviewable on appeal, we require that the district court ensure that these PSRs are made a part of the record, albeit under seal if need be.

The defendants have raised a host of other issues of varying merit. We briefly address them in turn.

1. Sufficiency of the Evidence as to Carreon

Carreon contends that there was insufficient evidence to support his convictions on the conspiracy to possess and to import marihuana because the evidence at trial indicated that there were multiple conspiracies, instead of one overall conspiracy on which he was indicted. According to Carreon, the instant case involves at least two conspiracies: the bribery-importation conspiracy involving Maynes (of which Carreon concedes he was a member), and an earlier importation conspiracy centered around Melendez and the government witness, Charles Aragon. Even if we were to assume arguendo that this case involves multiple conspiracies, Carreon's argument fails for lack of prejudice to a substantial right.⁶⁵ 

We have long held that "when the indictment alleges ... a single conspiracy, but the 'government proves multiple conspiracies and a defendant's involvement in at least one of them, then there is no variance affecting that defendant's substantial rights.' "⁶⁶  This holding is subject to the caveat that substantial rights are affected when the defendant is subjected to transference of quilt, that is, the danger that the defendant may be convicted because of his association with, or conspiracy for unrelated purposes with, codefendants who were members of the charged conspiracy.⁶⁷ 

Here, the evidence adequately demonstrates that Carreon joined with Melendez to engage in activity that was part of the charged conduct--on more than one occasion Carreon was taped discussing the importation and distribution of marihuana with Maynes. This discussion eventually led to the actual importation of more than the minimum charged quantity of marihuana. Carreon's argument is thus reduced to asserting that he did a little bribing, a little smuggling, and a little distributing, just not as much as Melendez and the others. Such an argument negates any claim of an infringement on a substantial right.

Carreon contends that the district court erred in denying his motion for a bill of particulars, and that this error caused him substantial prejudice. According to Carreon's theory, the bill of particulars would have put him on notice that the majority of the government's evidence concerned a conspiracy separate from the one he joined. Carreon complains that the evidence from this other conspiracy was improperly used against him to attribute "guilt by association." In sum, Carreon insists that, had he had the bill of particulars, he would have been able to sever his trial from Melendez.

We conclude that Carreon's argument fails for lack of demonstrated prejudice. Even if we were to assume that the district court's denial of the bill of particulars was an abuse of discretion--a highly doubtful assumption⁶⁸ --Carreon has failed to show that his rights were substantially prejudiced by the ultimate harm flowing from denial, namely, the failure to sever. To establish harm from failure to sever, Carreon would have had to show that the joint trial "compromise [d] a specific right [of his], or prevente [d] the jury from making a reliable judgment about guilt or innocence."⁶⁹  In contrast, Carreon claims no specific harm from the failure to sever other than to assert--baldly and without supporting factsthat such failure violated his right to a fair trial under the Due Process Clause. But, as noted above, this assertion is negated by the evidence at trial incriminating Carreon--which was direct, straightforward, and substantial.

In pretrial motions, Carreon requested and received severance of his and Melendez's trial from that of the other coconspirators. Carreon admitted that he was properly joined with Melendez because of their purported participation in a scheme to bribe Maynes. On appeal, Carreon now argues that he should also have been severed from Melendez.

Because Carreon failed to object to joinder with Melendez--indeed, he requested it--we review the district court's failure to sever here only for plain error. Plain error is "error so obvious and substantial that failure to notice it would affect the fairness, integrity, or public reputation of (the) judicial proceedings and would result in manifest injustice."⁷⁰  As noted above, substantial evidence indicates that Carreon was involved in a conspiracy to possess and to import marihuana. The district court's failure to sever plainly did not cause any "manifest injustice" here.

4. Heightened Standard of Proof for Certain Sentencing Facts

Carreon contends that the district court's findings of drug quantities used to sentence him are subject to a "beyond a reasonable doubt" burden of proof.⁷¹  Carreon reasons that this higher standard is required here because the "sentencing fact is a 'tail that wags the dog of the substantive offense.' "⁷²  Specifically, Carreon was indicted and eventually found guilty of conspiring to possess and to import 100 or more kilograms of marihuana, but was sentenced based on a finding that the quantity involved was around 130,000 kilograms. The guideline range for the minimum quantity found by the jury is 63 to 78 months;⁷³  Carreon was sentenced to 235 months based on the sentencing findings.

Our recent opinion in United States v. Mergerson⁷⁴  disposes of Carreon's claim. In Mergerson the relevant sentencing fact--that the defendant conspired to possess more than one kilogram of heroin--increased his sentence from thirty years to life under the Guidelines to mandatory life in prison without the possibility of release under 21 U.S.C. § 841(b) (1) (A) (i).⁷⁵  We concluded that the increase in the minimum sentence from thirty years to life "did not have the dramatic effect upon sentencing necessary to require the reasonable doubt standard to be considered."⁷⁶  The difference here--between approximately six and almost twenty years--likewise does not constitute such a dramatic effect that it would justify considering, much less imposing, the higher burden of proof.

5. Sufficiency of the Evidence as to Melendez

Melendez was found guilty of conspiring to import 100 or more kilograms of marihuana, but was acquitted of conspiring to possess that amount and of bribing a public official. Although substantial evidence exists to support Melendez's importation conviction if the testimony of the government's witnesses is accepted as credible,⁷⁷  we pretermit further discussion on this issue as we must remand Melendez's conviction for review of the PSRs of the coconspirators-witnesses.

6. Sentencing Accountability for Acquitted Conduct

Melendez also argues that the district court violated the Due Process Clause when it attributed marihuana to him that was part of the possession conspiracy for which he was acquitted. Melendez's contention--that a district court cannot consider quantities related to a conviction for which he was acquitted--is meritless. A district court may base a defendant's sentence on conduct for which the defendant was acquitted because the government need only establish sentencing facts (unlike the elements of the crime) by a preponderance of the evidence.⁷⁸ 

III

CONCLUSION

We hold today that relevant conduct as defined in Sec. 1B1.3(a) (1) (B) is prospective only, and consequently relevant conduct under Sec. 1B1.3(a) (1) (B) cannot include conduct occurring before the defendant joins a conspiracy. As Carreon's sentence includes quantities arising out of drug transactions occurring before he joined any conspiracy, we must vacate Carreon's sentence and remand for resentencing. We also conclude that the absence of sentencing findings as to both Carreon and Melendez warrants remand for findings and for resentencing both defendants.

Finally, we must remand Melendez's conviction for findings as to whether the PSRs of the coconspirators-government witnesses contained material Brady or Giglio information, and, if they do contain such information, whether failure to disclose such information was harmless error.

For the foregoing reasons, the sentences of Carreon and of Melendez are VACATED and their cases REMANDED for resentencing; and the conviction of Melendez is REMANDED (but not reversed) for additional findings by the sentencing court.