United States v. Soto-Beniquez ( 2003 )


Menu:
  •               United States Court of Appeals
    For the First Circuit
    Vol. II of II
    No. 01-1619
    UNITED STATES OF AMERICA,
    Appellee,
    v.
    WILLIAM SOTO-BENÍQUEZ,
    Defendant, Appellant.
    No. 01-1674
    UNITED STATES OF AMERICA,
    Appellee,
    v.
    JUAN SOTO-RAMÍREZ,
    Defendant, Appellant.
    No. 00-1547
    UNITED STATES OF AMERICA,
    Appellee,
    v.
    EDUARDO ALICEA-TORRES,
    Defendant, Appellant.
    No. 01-1620
    UNITED STATES OF AMERICA,
    Appellee,
    v.
    RAMON FERNÁNDEZ-MALAVÉ,
    Defendant, Appellant.
    No. 00-1464
    UNITED STATES OF AMERICA,
    Appellee,
    v.
    CARMELO VEGA-PACHECO,
    Defendant, Appellant.
    No. 00-1488
    UNITED STATES OF AMERICA,
    Appellee,
    v.
    ARMANDO GARCÍA-GARCÍA,
    Defendant, Appellant.
    No. 00-1470
    UNITED STATES OF AMERICA,
    Appellee,
    v.
    JOSE LUIS DE LEÓN MAYSONET,
    Defendant, Appellant.
    No. 00-1362
    UNITED STATES OF AMERICA,
    Appellee,
    v.
    RENE GONZALEZ-AYALA,
    Defendant, Appellant.
    No. 00-1543
    UNITED STATES OF AMERICA,
    Appellee,
    v.
    JUAN ENRIQUE CINTRÓN-CARABALLO,
    Defendant, Appellant.
    No. 00-1361
    UNITED STATES OF AMERICA,
    Appellee,
    v.
    MIGUEL VEGA-COLÓN,
    Defendant, Appellant.
    No. 00-1456
    UNITED STATES OF AMERICA,
    Appellee,
    v.
    MIGUEL VEGA-COSME,
    Defendant, Appellant.
    APPEALS FROM THE UNITED STATES DISTRICT COURT
    FOR THE DISTRICT OF PUERTO RICO
    [Hon. Daniel R. Domínguez, U.S. District Judge]
    Before
    Selya, Circuit Judge,
    Coffin, Senior Circuit Judge,
    and Lynch, Circuit Judge.
    Marlene Apontes-Cabrera for appellant Soto-Beníquez.
    Miriam Ramos-Grateroles for appellant Soto-Ramírez.
    Raymond Rivera Esteves for appellant Alicea-Torres.
    Luz M. Rios-Rosario for appellant Fernández-Malavé.
    Javier Morales-Ramos for appellant Vega-Pacheco.
    Rachel Brill for appellant García-García.
    Roberto Roldan-Burgos for appellant de León Maysonet.
    Victor Miranda-Corrada, for appellant Gonzalez-Ayala.
    Rafael Anglada-Lopez for appellant Cintrón-Caraballo.
    Marcia G. Shein for appellants Vega-Cosme and Vega-Colón.
    Jacabed Rodriguez-Coss and Michelle Morales, Assistant United
    States Attorneys, with whom H.S. Garcia, United States Attorney,
    and Sonia I. Torres-Pabon, Assistant United States Attorney, were
    on brief, for appellee.
    November 20, 2003
    f)    Improper Admission of Rule 702 Expert Testimony As
    Lay Testimony Under Rule 701
    (Cintrón-Caraballo)
    Cintrón-Caraballo     argues    that   the   court    should   have
    excluded the testimony of eleven witnesses because they provided
    expert testimony but, he says, were not disclosed as experts under
    Rule 702.8       These witnesses included eight forensic examiners
    (Ruben Diaz-De Leon, Alfredo Roman-Rodriguez, Virginia Cortes, Luis
    Batista-Maldonado, Nelson Morales-Huerta, Luis Mercedes-Rodriguez,
    Francisco    Ramos-Seda,    and     Cesar     W.    Ostolaza-Perez),        two
    pathologists (Dr. Yocasta Brougal-Mena and Dr. Francisco Cortes),
    and a firearms examiner (Juan B. Maldonado). This was prejudicial,
    Cintrón-Caraballo argues, because the defendants would have been
    entitled to summaries of the witnesses' testimony if they had been
    designated as experts. See Fed. R. Crim. P. 16(a)(1)(G)(defendants
    are entitled to summaries of all expert testimony, which must
    include "the witness's opinions, the bases and reasons for those
    opinions, and the witness's qualifications").
    The district court correctly determined that none of the
    eight forensic examiners provided expert testimony.              Witnesses who
    testify only about their perceptions of an event, or about lay
    opinions arising out of those perceptions, see Fed R. Evid. 701,
    8
    The trial in this case took place in 1999, before the
    December 1, 2000 effective date of the amendments to Rules 701 and
    702. Accordingly, we apply the pre-amendment Rules and case law.
    -63-
    are not experts under Rule 702 regardless of any specialized
    training or experience they may possess.                            See United States v.
    Paiva,    
    892 F.2d 148
    ,       157    (1st     Cir.   1989)     ("[T]he      individual
    experience and knowledge of a lay witness may establish his or her
    competence, without qualification as an expert, to express an
    opinion on       a     particular           subject    outside      the    realm    of    common
    knowledge."); see also United States v. Rivera-Santiago, 
    107 F.3d 960
    , 968 (1st Cir. 1997).                  That rule is dispositive here: the court
    permitted       each      of    the    witnesses       to    testify      only    about     their
    observations         at        the     various      crime     scenes       they     personally
    investigated.          Indeed, the court consistently reminded both the
    witnesses and the lawyers that if any of these witnesses' testimony
    "sound[ed] like a 702 [opinion] . . . [he would] not admit it."
    Although at points the district court faced difficult decisions
    about the side of the Rule 701 / Rule 702 divide on which a
    witness's opinion fell, there was no abuse of discretion in the
    court's    resolution            of    these     issues.      See    Kumho       Tire    Co.   v.
    Carmichael, 
    526 U.S. 137
    , 152 (1999) (review of a district court's
    decision to admit or exclude expert testimony is for abuse of
    discretion).
    Nor did the district court abuse its discretion in
    allowing the expert testimony of the two pathologists, Drs. Cortes
    and   Brugel-Mena.                   The    district     court      found,        despite      the
    government's failure to label the witnesses' testimony and reports
    -64-
    as   Rule   702    material          in     its    pre-trial       disclosure,   that   the
    government had effectively complied with the applicable disclosure
    requirements.            In     particular,          the    government     had    informed
    defendants before trial that both pathologists would be testifying
    about several autopsies and provided the defendants with copies of
    all of these autopsy reports.                 Although Dr. Cortes testified about
    one autopsy report that he did not personally prepare, the district
    court permitted this substitution because the pathologist who had
    prepared that report was unavailable to testify due to serious
    illness.     There is no generalized prohibition on allowing experts
    to testify about autopsy reports that they did not personally
    prepare.     See Manocchio v. Moran, 
    919 F.2d 770
    , 780 (1st Cir.
    1990).
    The government failed to formally designate the last
    witness, Juan Maldonado, as an "expert", but it did inform the
    defendants that Maldonado would be testifying about ballistics and
    provided    the    defense           with    all    of     Maldonado's    notes    on   his
    testimony.         And        once    again,       the     district     court    permitted
    Maldonado's testimony due to the lack of prejudice to the defense.
    Here,    though,    the        court      compensated        the    defendants    for   the
    government's failure to adhere to the technical requirements of
    Fed. R. Crim. P. 16 by certifying the witness only as a ballistics
    expert, and refusing to also certify him as a "firearms expert."
    This decision was an appropriate sanction against the government
    -65-
    and undercuts the defendant's prejudice argument.
    g)   Rule 404(b) "Bad Act" Evidence
    (Cintrón-Caraballo)
    Cintrón-Caraballo   argues    that   the   court   erred    in
    admitting evidence of his March 8, 1994 arrest by Puerto Rico
    police, and of the contemporaneous seizure of a gun that he was
    carrying.    He argues that this evidence was impermissible bad act
    evidence under Fed. R. Evid. 404(b) because it was not relevant to
    demonstrating his participation in the conspiracy, which, according
    to the indictment, had ended one day earlier, on March 7.       Cintrón-
    Caraballo also argues that the firearms evidence should have been
    excluded as unreliable because the firearm had been destroyed by
    Puerto Rico authorities.
    These arguments are unavailing.      Evidence of Cintrón-
    Caraballo's arrest was admissible under Rule 404(b) because the
    arrest was for activities evidencing his participation in the
    conspiracy charged in the indictment.        The arrest took place on
    Street B, where the drug point that the government alleged Cintrón-
    Caraballo supervised was located.        This evidence demonstrated an
    overt act in furtherance of the alleged conspiracy and Rule 404(b)
    explicitly provides that evidence of bad acts is admissible for
    purposes other than showing actions in conformity with those acts.
    See Fed. R. Evid. 404(b).
    The fact that the indictment charged the conspiracy with
    ending "on or about" March 7 does not change this conclusion.          The
    -66-
    "on or about" language left the district court leeway to conclude
    that the arrest fit within this time frame, and thus that the
    arrest was evidence of an act directly in furtherance of the
    conspiracy.      Cf. Portela, 167 F.3d at 704 (indictment charging
    defendant with possession of cocaine "on or about" March 1995
    provided "perfectly adequate" notice to the defendant for acts
    charged in April 1995).
    We also reject Cintrón-Caraballo's related objection that
    the photograph of the gun should have been excluded because there
    was no reliable evidence that it was the gun actually seized from
    him.   The district court found that there were sufficient indicia
    of reliability that the photograph was what it purported to be
    because specific markings on the gun in the photograph matched the
    description in the police report.            The arresting officer also
    testified that the photograph depicted the weapon seized from
    Cintrón-Caraballo.      Under these circumstances, the photograph was
    properly authenticated.      See Fed. R. Evid. 901(a).
    h)     Admission of Evidence on Rebuttal
    (Soto-Ramírez)
    In   the   government's      case-in-chief,    Negrón-Maldonado
    testified   that    Soto-Ramírez   was    involved   in   the   murder   of   a
    government informant, Ana Luz Dones-Arroyo. Soto-Ramírez countered
    this testimony by suggesting that the government did not have
    sufficient evidence to indict him for Dones-Arroyo's murder because
    it had accepted his guilty plea to the charge of accessory after
    -67-
    the fact.   In rebuttal, the government called Juan Maldonado, who
    had previously testified as a ballistics expert, but whom the court
    had refused to further qualify as a firearms expert.                Maldonado
    testified that the same weapon that was used to kill Robles-
    Rodríguez -- a murder to which Soto-Ramírez had pled guilty -- was
    also used in the murder of Dones-Arroyo.
    Defendant   argues   that   Maldonado's     testimony    was    not
    admissible as rebuttal evidence because Soto-Ramírez's argument
    that he only pled guilty to the accessory after the fact charge was
    not "a sweeping denial" of his involvement in the Dones-Arroyo
    murder.   We review the admission of rebuttal evidence for abuse of
    discretion. See United States v. Leon-Delfis, 
    203 F.3d 103
    , 113
    (1st Cir. 2000); Faigin v. Kelly, 
    184 F.3d 67
    , 85 (1st Cir. 1999).
    The district court did not abuse its discretion in
    rejecting    Soto-Ramírez's      argument   and   admitting    Maldonado's
    testimony   as   rebuttal   evidence.       "Rebuttal    evidence     may   be
    introduced to explain, repel, contradict or disprove an adversary's
    proof." United States v. Laboy, 
    909 F.2d 581
    , 588 (1st Cir. 1990).
    That is exactly what the government did here.           The defense opened
    the door to Maldonado's testimony when it attempted to demonstrate
    that Soto-Ramírez was only an accessory after the fact because that
    claim implied that Soto-Ramírez was not guilty of the underlying
    murder.
    Soto-Ramírez also argues that Maldonado's testimony in
    -68-
    rebuttal should have been excluded because it was expert firearms
    testimony that the district court had specifically excluded during
    the government's case-in-chief.     The court rejected this argument,
    concluding that the defendants had sufficient notice that Maldonado
    would testify to this issue and thus that its previous holding
    limiting Maldonado's testimony to ballistics was not applicable.
    This conclusion was sound.       The defendants were provided with a
    report before trial that Maldonado would testify that the same
    weapons were used in the Robles-Rodríguez and Dones-Arroyo murders.
    2.   Brady and Giglio Claims
    (Soto-Beníquez, Soto-Ramírez, Alicea-Torres)
    Defendants contend that there were multiple Brady and
    Giglio violations.    See Brady v. Maryland, 
    373 U.S. 83
    , 87 (1963);
    Giglio v. United States, 
    405 U.S. 150
    , 153-54 (1972).                First,
    several defendants challenge the government's failure to reveal the
    apparent inconsistencies in Rodríguez-López's story when it first
    became aware of them in the summer of 1998.        At that time, all the
    prosecution knew was that Rodríguez-López had lied to the FBI (not
    the grand jury) about being present at Rivera-González's murder.
    He had not testified to the grand jury about that murder.          Although
    the prosecution must reveal material information that is favorable
    to the accused, the fact that Rodríguez-López may not have been
    present at the Rivera-González murder is not exculpatory evidence.
    Admittedly,   the    analysis   might    have   been   different    if   the
    government had ultimately called Rodríguez-López as a witness at
    -69-
    trial: his earlier lies to the government would certainly have
    constituted a basis for impeaching him.             See Giglio, 
    405 U.S. at 153-54
    ; Moreno-Morales v. United States, 
    334 F.3d 140
    , 145 (1st
    Cir. 2003).        But the government did not call Rodríguez-López.
    Furthermore, as the trial court noted, the defendants knew a month
    before trial that Rodríguez-López had lied, so they had sufficient
    time to interview him and have him testify if they so desired.
    Several defendants also argue that the government failed
    to   fully   disclose   the   extent   of   Negrón-Maldonado's        plea   and
    cooperation agreement with the government.                After testifying on
    direct examination about ten murders, Negrón-Maldonado admitted
    during redirect examination that the government had promised him
    favorable treatment in his related state court proceedings in
    exchange for his testimony.        Counsel for Soto-Beníquez and Soto-
    Ramírez immediately moved for a mistrial, telling the court that
    the prosecution had never disclosed its intervention in the Puerto
    Rico courts on behalf of the witness.                   (The prosecution had
    disclosed the existence of a plea arrangement between itself and
    Negrón-Maldonado).      At side-bar, the        government explained that
    while the     Commonwealth    of   Puerto   Rico    had    made    certain   oral
    assurances to the witness at the prosecution's behest, no agreement
    had been reduced to writing and thus there was no document that
    could have been produced to inform the defense of the agreement.
    The   government's    obligation      to   disclose    impeachment
    -70-
    evidence is not, as suggested by the prosecution, dependent on
    whether that evidence has been reduced to written form.                          See
    Giglio, 
    405 U.S. 152
    , 154-55 (reversing conviction where an oral
    agreement between a prosecutor and key witness was not disclosed to
    the defense).         Here, the government failed to disclose the full
    extent    of    its   agreement   with    the    witness       until   the   defense
    uncovered the details of the arrangement during cross-examination.
    Nonetheless, the defendants were not prejudiced by the
    government's delay in revealing this information and are not
    entitled to reversal on appeal. See United States v. Lemmerer, 
    277 F.3d 579
    , 588 (1st Cir. 2002) (a                      defendant must show that
    "learning the information altered the subsequent defense strategy,"
    and that given timely disclosure, "a more effective strategy would
    likely have resulted" (quoting United States v. Devin, 
    918 F.2d 280
    , 290 (1st Cir. 1990)); United States v. Ingraldi, 
    793 F.2d 408
    ,
    411-12 (1st Cir. 1986) (same).           Negrón-Maldonado admitted the full
    extent    of    his   arrangement      with    the    government   during     cross-
    examination. Moreover, the defendants' strategy in cross-examining
    Negrón-Maldonado        was   surely    not    impacted    by   the    government's
    delayed    disclosure.        Even     without       knowing   about   the   federal
    prosecution's intervention in state court, the defense's cross-
    examination of Negrón-Maldonado was intended to suggest that the
    witness was fabricating his testimony in order to receive favorable
    treatment.      There has been no showing that having a larger quantum
    -71-
    of evidence than originally supposed would have altered the way in
    which the defense cross-examined the witness, and the court granted
    additional time to defense counsel to prepare and investigate the
    new information before cross-examination of the witness resumed.
    Again, we do not approve of the prosecution's conduct; we hold only
    that it does not provide a basis for reversal.
    Soto-Beníquez     and   Soto-Ramírez   also   allege   that   the
    prosecution failed to disclose that it had granted immunity to
    Janet   Garcia-Diaz,   the   girlfriend    of   Torrens-Alicea,    another
    cooperating witness.    This argument is without merit.       Garcia-Diaz
    was told that she would not be prosecuted after she specifically
    inquired of the government whether she needed a lawyer, on the same
    day that she was called as a defense witness by Vega-Colón to
    impeach Torrens-Alicea's testimony for the prosecution.           There was
    a window of, at most, several hours between the government's
    statement to Garcia-Diaz and the defendants' discovery of this
    supposed grant of immunity. Even assuming that the defendants were
    entitled to this information under Giglio, they became aware of the
    so-called "grant of immunity" on the same day that it was extended.
    No prejudice has been shown.
    Finally, Soto-Beníquez and Soto-Ramírez suggest that the
    prosecution did not disclose the fact that cooperating witnesses
    were allowed to make unmonitored phone calls, visit with their
    spouses, and take pictures of themselves "half-naked" in government
    -72-
    offices.     Defendants, though, were informed by discovery letter
    about several visits by family members to cooperating witnesses.
    In any case, these benefits pale in comparison with the deals
    negotiated in the plea bargains. Defendants were well aware of the
    agreements with cooperating witnesses (absent that of Negron-
    Maldonado,     discussed   above)   and   used   them   well   in   cross-
    examination.    There was no prejudice to the defendants.
    3.    Closing Arguments
    (de León Maysonet, Gonzalez-Ayala, García-García)
    Several defendants urge that the prosecution's closing
    argument led to reversible error.
    In its rebuttal in closing the prosecutor argued:
    And one point that I want to make clear as to Ramon
    Fernandez Malave, as to Carmelo Vega Pacheco, as to Rene
    Gonzalez Ayala, as to Jose Luis de Leon Maisonet and
    anyone else who argues here before you that they are here
    before you pleading not guilty, pleading their innocence.
    Well, let me tell you something, ladies and gentlemen of
    the jury, a plea of not guilty is not, not a declaration
    of innocence.     A plea of not guilty simply means,
    government, prove your case. But a plea of not guilty is
    not a declaration of innocence.
    (emphasis added).    Defense counsel objected:
    We would like to interpose an objection, hinges on the
    constitution right to the presumption of innocence.
    The court replied, in the presence of the jury:
    There is a presumption of innocence going on.         Fine.
    (emphasis added).    The prosecutor echoed that:
    There is a presumption of innocence.          They are to be
    -73-
    presumed innocent, that is not what I'm arguing against,
    Your Honor. And I understand the jurisprudence from the
    First Circuit supports my argument.
    The court then said:
    Keep on going.
    Later, the prosecution made a similar statement:
    Carmelo Vega Pacheco again comes before you and says my
    client is pleading not guilty.     Again, a plea of not
    guilty is not a declaration of innocence. It simply puts
    the government to its proof.     And he argues, yes, he
    participated in [the] Quintana massacre but that was in
    furtherance of a different conspiracy.
    (emphasis   added).     The    defense   again   objected    to    this   later
    statement, but not on the ground asserted on appeal -- namely, that
    the prosecutor's comments undercut the presumption of innocence.
    Because a contemporaneous objection was made by defense
    counsel to the earlier statement, we review de novo the question of
    whether    the   argument    was   improper   and   review   for       abuse   of
    discretion the court's ruling on whether the misconduct, if any,
    warrants a new trial.       United States v. Hernandez, 
    218 F.3d 58
    , 68
    (1st Cir. 2000).    We conclude that error occurred but that it does
    not warrant a new trial.
    On appeal, the prosecution argues that these statements
    were an accurate description of the law. It also contends that its
    comments    were   invited    by   the   improper   argument      of    several
    defendants that their pleas of not guilty in this case were
    reliable indications of their innocence because if they were guilty
    they would have admitted it in this case, as they did in the state
    -74-
    court.
    The prosecution is wrong on both points.          First, the
    prosecutor's comments did undermine the presumption of innocence.
    By saying that a plea of not guilty is "not a declaration of
    innocence" but simply means "government, prove your case," the
    prosecutor undercut the axiomatic principle that a defendant is
    presumed innocent until proven guilty and need not declare or prove
    that he is innocent.          Regardless of the complex relationship
    between the presumption of innocence and the prosecution's duty to
    convince the jury beyond a reasonable doubt, see, e.g., Taylor v.
    Kentucky, 
    436 U.S. 478
    , 483-85 (1978) (noting the scholarly debate
    concerning whether the presumption of innocence is analytically
    distinct from the requirement that the government prove guilt
    beyond a reasonable doubt); McCormick on Evidence § 346 (5th ed.
    1999)    (suggesting   that    the   presumption   of   innocence   is   "a
    convenient introduction to the statement of the burdens upon the
    prosecution"), due process requires that both of these principles
    guide the jury in reaching its verdict.        Taylor, 
    436 U.S. at
    483-
    86; Coffin v. United States, 
    156 U.S. 432
    , 453, 461 (1895).              To
    undercut one, even if the other remains standing, is improper.           It
    is for precisely this reason that a district court's failure to
    instruct the jury on the presumption of innocence may violate due
    process even when the jury has been properly informed of the
    prosecution's burden of proving guilt beyond a reasonable doubt.
    -75-
    Taylor, 
    436 U.S. at 488-89
    .
    The prosecution's contention that the statements were a
    justified response to the argument of defense counsel is also
    incorrect.     Although it is true that, in certain circumstances, a
    prosecutor's otherwise impermissible statements during closing
    argument may be allowable because they were "invited" by defense
    counsel, United States v. Henderson, 
    320 F.3d 92
    , 107 (1st Cir.
    2003), this was clearly not such a case.                     Defense counsel's
    argument that the defendants' pleas of not guilty in federal court
    were particularly trustworthy because the defendants had formerly
    pled guilty in state court was not improper and did not justify the
    prosecutor's response.
    Not every prosecutorial error in making closing argument
    justifies     a   new   trial,        even    when   that    error     undermines
    constitutional     rights.       No    reversible    error    occurs    when   the
    reviewing court determines beyond a reasonable doubt that the
    constitutional error was harmless. Wihbey, 75 F.3d at 772 n.6; see
    also United States v. Hasting, 
    461 U.S. 499
    , 510-11 (1983).                As the
    Supreme Court has clarified, the relevant question "is not what
    effect the constitutional error might generally be expected to have
    upon a reasonable jury, but rather what effect it had upon the
    guilty verdict in the case at hand."             Sullivan v. Louisiana, 
    508 U.S. 275
    , 279 (1993); see United States v. Rivera-Santiago, 
    107 F.3d 960
    , 967 (1st Cir. 1997).
    -76-
    We   conclude     beyond      a    reasonable      doubt    that      the
    prosecutor's       improper    closing     argument      did   not    prejudice    the
    defendants in this case. The court gave curative instructions that
    established the presumption of innocence immediately after the
    prosecutor's first improper statement (the only time the defense
    made    the    appropriate      objection).         It    then       reinforced    the
    presumption in its general instructions to the jury, noting that it
    "is a cardinal principle of our system that every person accused of
    a crime is presumed to be innocent unless and until his/her guilt
    is     established    beyond     a    reasonable      doubt."          Given      those
    instructions and the strong evidence of guilt, we conclude beyond
    a reasonable doubt that the statements did not affect the ultimate
    outcome of the case, especially when they occupied only several
    seconds in a six-month long trial.
    4.     Cumulative Effect of Errors
    (Soto-Beníquez, Soto-Ramírez)
    A series of errors, each one of which is individually
    "harmless," may have a cumulative effect that requires a new trial.
    United States v. Sepulveda, 
    15 F.3d 1161
    , 1195-96 (1st Cir. 1993).
    Defendants      rely   on   this    proposition,        arguing     that,
    considered as a whole, the prosecution's missteps warrant a new
    trial. To this point, we have concluded that the prosecution erred
    in repeatedly failing to meet discovery deadlines, in neglecting to
    disclose the extent of its plea arrangement with Negrón-Maldonado,
    and in making inappropriate remarks during closing arguments. This
    -77-
    conduct is blameworthy and the government should take steps to see
    that it does not recur.
    Still, the government's bad behavior does not require
    that the jury's verdict of guilt be set aside.                 At a minimum, to
    overturn a verdict, the prosecution's bad behavior must have
    prejudiced the defendants. See, e.g., United States v. Joyner, 
    191 F.3d 47
    ,   53   (1st    Cir.   1999)    (in     evaluating    allegations   of
    prosecutorial misconduct, "the unavoidable bottom line" is "whether
    we deem it likely, or not, that any prejudice affected the outcome
    of the case").      Although the frustrations of defense counsel are
    understandable, that test is not met here.
    The defense was not demonstrably prejudiced by any of the
    government's violations, and sometimes even gained an advantage
    from them.        The defendants ultimately received the necessary
    discovery and were provided with compensation such as additional
    discovery and the exclusion of otherwise admissible evidence. When
    the defendants learned of the federal prosecution's intervention in
    state court on behalf of Negrón-Maldonado, the court offered them
    additional time to cross-examine the witness.                    And while the
    court's   curative       instruction     during    the   prosecutor's   closing
    arguments was concise, it was sufficient in the context of the
    overall instructions to assure that the jury was properly appraised
    of the import of the presumption of innocence.                 The totality of
    errors argument is unsuccessful.
    -78-
    D.   Post-Trial
    1.     Sufficiency of Evidence as to CCE Count
    (Soto-Beníquez, Soto-Ramírez)
    Soto-Beníquez     and      Soto-Ramírez    argue    that    there   is
    insufficient evidence to support their convictions.                 They offer no
    further explanation, except to cite to their filings before the
    district court.        Their argument as to sufficiency of evidence has
    been waived.     See Grella, 42 F.3d at 36. "If counsel desires our
    consideration of a particular argument, the argument must appear
    within    the   four    corners   of   the    brief   filed    in   this   court."
    Executive Leasing Corp. v. Banco Popular de P.R., 
    48 F.3d 66
    , 67-68
    (1st Cir. 1995).        Attorneys cannot circumvent this requirement by
    referencing their district court filings.              
    Id. at 68
    .
    2.     Sufficiency of Evidence as to Conspiracy Count
    (Cintrón-Caraballo, Vega-Cosme, Vega-Colón)
    Vega-Cosme and Vega-Colón argue that the evidence is
    insufficient to tie them to the charged conspiracy. We reject this
    claim.
    The government presented overwhelming evidence of Vega-
    Cosme's participation in the conspiracy.                Government witnesses
    testified that Vega-Cosme had a series of agreements with other
    members of the conspiracy to maximize drug revenue.                 He negotiated
    with the Chacho gang on behalf of the drug points to end the
    warfare that was interfering with drug sales, met with Negrón-
    Maldonado to coordinate the colors of crack capsule caps to avoid
    -79-
    competition between the points, and arranged the opening of his own
    drug point on Laguna Street with Soto-Ramírez. Although Vega-Cosme
    correctly notes that there is no evidence that Soto-Ramírez acted
    as his supplier, the government did present testimony that Vega-
    Cosme supplied narcotics to Soto-Ramírez, along with ammunition
    used in shootings of rival gang members in 1992 and 1993.
    The    record    also    shows    sufficient       evidence    of    the
    participation of Vega-Cosme's son, Miguel Vega-Colón.                Vega-Colón
    packaged crack cocaine, heroin, and marijuana for his father's
    point. He also stood armed guard at the Callejón Nueve drug point,
    which was owned by Rodríguez-López.            Both of those activities were
    in furtherance of the conspiracy.            Moreover, based on Vega-Colón's
    presence    at    meetings    between   Vega-Cosme       and    Soto-Ramírez,      a
    reasonable jury could have concluded that he joined the conspiracy
    knowingly and voluntarily.
    Vega-Cosme       and    Vega-Colón    also     argue    that        their
    convictions are based on unreliable testimony from co-conspirators
    who "had clear incentives to testify untruthfully."                 In assessing
    the sufficiency of evidence, credibility determinations must be
    resolved in favor of the verdict.            United States v. Guerra-Garcia,
    
    336 F.3d 19
    , 22 (1st Cir. 2003).              Credibility judgments are the
    province of the jury, not of this court.
    Cintrón-Caraballo also makes an insufficiency of evidence
    argument.    He argues that the district court erred in admitting
    -80-
    certain expert testimony and Rule 404(b) "bad act" evidence, and
    contends that without this evidence, the only evidence tying him to
    the conspiracy is the uncorroborated testimony of co-conspirators.
    This argument fails. The contested evidence was properly admitted.
    See Part III.B(f)-(g).       Moreover, even if Cintrón-Caraballo's
    conviction rested only on co-conspirator testimony, the jury was
    entitled to credit such testimony and convict him on that basis.
    See United States v. Torres-Galindo, 
    206 F.3d 136
    , 140 (1st Cir.
    2000).
    3.   Special Verdict and Jury Instructions for CCE Count
    (Soto-Beníquez, Soto-Ramírez)
    Soto-Beníquez and Soto-Ramírez claim that the district
    court committed reversible error when it failed to instruct the
    jury to determine the quantity and type of drugs.        First, they
    argue that the drug amount is an element of the CCE offense and
    that the jury was not otherwise instructed to find a minimum drug
    amount.   Second, they argue that Apprendi v. New Jersey, 
    530 U.S. 466
     (2000), requires that the drug amount be proven to the jury
    beyond a reasonable doubt. The standard of review for alleged jury
    instruction errors involving the interpretation of the elements of
    a statutory offense is de novo.     United States v. Shea, 
    150 F.3d 44
    , 49-50 (1st Cir. 1998).
    We reject both arguments.     As to the claim that the drug
    amount is an element of the CCE offense, the CCE statute plainly
    does not require a minimum drug amount for a conviction.
    -81-
    As to Soto-Ramírez and Soto-Beníquez's Apprendi argument,
    Apprendi   requires   that    "[o]ther    than   the   fact   of   a   prior
    conviction, any fact that increases the penalty for a crime beyond
    the prescribed statutory maximum must be submitted to a jury, and
    proved beyond a reasonable doubt."        
    530 U.S. at 490
    .    Here, absent
    a finding of drug quantity, the statutory maximum for CCE is
    already life imprisonment: the statute authorizes a sentence of
    twenty years to life imprisonment regardless of drug amount.              
    21 U.S.C. § 848
    (a)-(c).     A drug amount above a certain level can
    result in a mandatory life sentence, § 848(b), but does not change
    the statutory maximum.       Hence, no Apprendi violation has occurred
    with regard to the CCE convictions.9
    4.    Special Verdict for Conspiracy Count
    (Alicea-Torres, Fernández-Malavé)
    Alicea-Torres and Fernández-Malavé argue that the trial
    court erred in not providing the jury with a special verdict form
    requiring it to determine the quantity and type of drugs as to each
    defendant. They argue that a special verdict form was necessary to
    9
    Soto-Beníquez and Soto-Ramírez also argue reversible
    error based on the district court's denial of their request for a
    special verdict sheet requiring the jury to find the type and
    amount of drugs as to each defendant.     In criminal cases, the
    failure to use a special verdict form is reviewed for abuse of
    discretion. United States v. Ellis, 
    168 F.3d 558
    , 562 (1st Cir.
    1999).   For the reasons discussed above, we find no such abuse
    here. It was not necessary for the jury to determine the quantity
    or type of drugs to convict Soto-Ramírez and Soto-Beníquez on the
    CCE count.
    -82-
    ensure that the jury was unanimous as to drug type and quantity.
    This argument is based on Apprendi and is discussed later.
    E.   Sentencing
    1.    Apprendi and Related Issues
    a)     Apprendi Error
    (Alicea-Torres,   Fernández-Malavé,             Vega-Pacheco,
    Vega-Cosme, Vega-Colón)
    Five    of    the    non-CCE      defendants   --    Alicea-Torres,
    Fernández-Malavé,       Vega-Pacheco,     Vega-Cosme,     and     Vega-Colón    --
    assert that their sentences violated the rule of Apprendi.                   They
    argue that the amount of drugs distributed by the conspiracy was
    not proven beyond a reasonable doubt to a jury, that the drug
    amount raised the statutory maximum, and that their sentences must
    be vacated as a result.        We conclude that Apprendi error did occur,
    but that the error was harmless.
    The jury instructions in this case did not make direct
    reference to drug amount or quantity.               Instead, the jury was
    instructed that, to find the defendants guilty of the conspiracy
    count, it had to find that the government proved the conspiracy
    charged in the indictment beyond a reasonable doubt.               The jury was
    provided   with   a     copy   of   the   indictment,     which    charged     the
    defendants with knowingly and intentionally distributing more than
    five kilograms of heroin, more than five kilograms of cocaine, more
    than five kilograms of crack cocaine, and more than 100 kilograms
    of marijuana.      However, the jury was also instructed that the
    -83-
    actual amount of drugs need not be proven, and that the government
    need only prove that defendants distributed or possessed with
    intent to distribute a "measurable amount" of drugs.
    The   latter   part   of    these   instructions   resulted    in
    Apprendi error.     This case presents an even stronger case of
    Apprendi error than United States v. Nelson-Rodriguez, 
    319 F.3d 12
    (1st Cir. 2003).    In that case, as here, the jury was given a copy
    of the indictment and instructed that to find the defendants guilty
    on the conspiracy count, it had to find them guilty of the
    conspiracy in the indictment.          
    Id. at 45
    .    The indictment in
    Nelson-Rodriguez, as here, specified drug types and quantities
    sufficient to support the defendants' sentences. 
    Id.
     We concluded
    that   this   instruction   was      insufficient   to   elicit   a     jury
    determination of the threshold drug amount and quantity.          
    Id.
       The
    same analysis applies here with greater force because the jury was
    specifically instructed that it need only find a "measurable
    amount" of drugs.
    All five defendants who raise the Apprendi issue were
    sentenced above the default statutory maximum.            Absent a jury
    determination of drug amount or type, the default statutory maximum
    is based on the distribution of unspecified amounts of marijuana,
    which results in a maximum sentence of five years for first-time
    felony drug convictions and ten years if a prior such conviction
    exists. 
    21 U.S.C. §§ 841
    (b)(1)(D), 846. Alicea-Torres, Fernández-
    -84-
    Malavé,    Vega-Pacheco,   and   Vega-Cosme   were   sentenced   to   life
    imprisonment, and Vega-Colón was sentenced to 292 months, or about
    twenty-five years, of imprisonment.
    The existence of an Apprendi error, however, does not end
    the inquiry.     If the defendants failed to preserve their Apprendi
    objection below, their sentences are vacated only if we find plain
    error.    United States v. Cotton, 
    535 U.S. 625
    , 631 (2002).     If they
    did preserve their objection, their sentences are vacated only if
    we find that the error was not harmless beyond a reasonable doubt.
    Nelson-Rodriguez, 
    319 F.3d at 49
    .
    Fernández-Malavé preserved his Apprendi objection, and we
    assume without deciding that the remaining defendants did the same,
    as it makes no difference to the outcome.       Defendants' trial took
    place from December 1998 to June 1999, before Apprendi was decided.
    At the time, several defendants requested a special verdict form
    requiring the jury to determine the drug amount and type as to each
    defendant.     But only Fernández-Malavé, who was the sole non-CCE
    defendant sentenced after Apprendi was decided, challenged his
    sentence before the district court on this basis.          The question
    whether the remaining defendants' special verdict request was
    sufficient to preserve an Apprendi objection, absent a separate
    objection at sentencing, is a complex one.       Nelson-Rodriguez, 
    319 F.3d at 48
    .      Here, as in Nelson-Rodriguez, we prefer to assume
    without deciding that the objection was preserved and the harmless
    -85-
    error standard applies.         
    Id.
    The Apprendi error in this case was harmless beyond a
    reasonable doubt. An Apprendi error is harmless where the evidence
    overwhelmingly establishes the minimum drug quantity needed to
    justify the statutory maximum under which the defendants were
    sentenced.        Martinez-Medina,    279    F.3d   at   121-22.      Here,   the
    government    produced   overwhelming        evidence    that   the   conspiracy
    involved at least five kilograms of cocaine, which triggers a
    maximum sentence of life imprisonment for all co-conspirators under
    
    21 U.S.C. § 841
    (b)(1)(A) and § 846.           Government witnesses Negrón-
    Maldonado and Torrens-Alicea both testified that Rodríguez-López
    and Gonzalez-Ayala stole a 200 kilogram shipment of cocaine in
    Fajardo, which was then brought back to Bitumul for distribution.
    Negrón-Maldonado further testified that in 1991 he purchased a
    kilogram of cocaine per week from Soto-Ramírez.                 After May 1992,
    Negrón-Maldonado      stated,    he   and    another     co-conspirator    named
    Manolín, who managed Soto-Ramírez's point while Soto-Ramírez was in
    prison, each purchased one kilogram of cocaine per week from Soto-
    Beníquez to be sold at their respective drug points.                     Negrón-
    Maldonado also testified that after January 1993, he continued to
    purchase from Soto-Beníquez three-eighths of a kilogram of cocaine
    each week for his drug point.
    In   addition,     the   government    presented      overwhelming
    evidence that the conspiracy distributed more than the 50 grams of
    -86-
    crack cocaine necessary to trigger a life sentence under 
    21 U.S.C. § 841
    (b)(1)(A).    According to Negrón-Maldonado, 125 grams of crack
    cocaine would yield approximately 800 to 850 crack capsules using
    the conspiracy's packaging techniques.     Negrón-Maldonado testified
    that Cintrón-Caraballo received 600 capsules of crack cocaine (90
    grams) per week for distribution at his drug point in 1990, 800 to
    850 capsules (125 grams) per week in 1991, and 1000 capsules (150
    grams) per week in early 1993.          Furthermore, Negrón-Maldonado
    testified that at the beginning of 1992, he would "cook" 500 grams
    to a kilogram of cocaine into crack cocaine two to three times per
    week for Soto-Ramírez's points; this alone amounts to one to three
    kilograms per week.     At around the same time, Negrón-Maldonado
    himself was also selling 500 to 800 crack capsules, or between 75
    to 125 grams of crack, per week.        When he left for the United
    States, he sold an additional 2000 crack capsules, or over 300
    grams of crack. All told, Negrón-Maldonado estimated that a single
    drug point would distribute at least one kilogram of crack cocaine
    per month -- more than twenty times the amount necessary to trigger
    a life sentence.
    In the face of this overwhelming evidence, defendants
    argue that the testimony of co-conspirators alone is never a
    sufficient basis to find an Apprendi error harmless beyond a
    reasonable doubt.    That is not so.    See United States v. Stewart,
    
    306 F.3d 295
    , 324-25 (6th Cir. 2002) (finding Apprendi error
    -87-
    harmless beyond a reasonable doubt based on the testimony of co-
    conspirators).    Defendants offered no evidence contradicting the
    conspiracy-wide drug quantities at trial, and they point to no such
    evidence on appeal, except their attack on the general credibility
    of the two witnesses.
    The jury in this case could not have convicted all eleven
    defendants of participation in the conspiracy without believing the
    testimony of Negrón-Maldonado and Torrens-Alicea regarding at least
    some of the transactions. Negrón-Maldonado and Torrens-Alicea also
    testified   regarding   the    quantity     of    drugs   involved   in   those
    transactions.    Defendants offer no explanation for why the jury
    would believe Negrón-Maldonado and Torrens-Alicea's account of each
    defendant's   activities      in   furtherance     of   the   conspiracy,   but
    discredit their testimony regarding the quantity or type of drugs
    involved in those activities.          In Nelson-Rodriguez, we found an
    Apprendi error harmless on very similar facts: the jury could not
    have convicted without crediting informant testimony, the same
    informant testified to the drug amount, and the defendant offered
    no reason to disbelieve the testimony except a general attack on
    the witness's credibility.         
    319 F.3d at 49-50
    .
    Defendants further protest that even if the conspiracy
    writ large involved the requisite quantities and types of drugs,
    the Apprendi error is not harmless.              They argue that it was not
    reasonably foreseeable to each of them individually, from their
    -88-
    limited   involvement,        that    such    quantities       of    drugs   would   be
    involved.         Under      the     Sentencing      Guidelines,        a    narcotics
    conspirator's sentence is based on the amount of drugs he actually
    handled, negotiated, or saw, as well as the amount of drugs that he
    reasonably could have foreseen to be embraced by the conspiracy he
    joined.     Rodriguez, 162 F.3d at 149; U.S.S.G. § 1B1.3 & cmt. 2.
    Defendants argue that unless this court is certain that the jury
    would   find   the    drug    quantity       reasonably    foreseeable        to   each
    defendant, the Apprendi error cannot be harmless.
    We reject this argument.              Apprendi does not require that
    the jury determine beyond a reasonable doubt the quantity of drugs
    foreseeable to each defendant.             Apprendi requires only that juries
    determine facts necessary to increase the statutory maximum.                         
    530 U.S. at 490
    .      Here, the conspiracy-wide drug quantity determines
    the   statutory      maximum.        See    
    21 U.S.C. § 846
        (holding     each
    conspirator responsible for the quantity of drugs distributed by
    the conspiracy).          As long as the sentence falls within this
    statutory maximum, the district court may determine the quantity of
    drugs reasonably foreseeable to each defendant by a preponderance
    of the evidence and sentence each defendant accordingly. Derman v.
    United States, 
    298 F.3d 34
    , 42-43 (1st Cir. 2002).                    In determining
    whether an Apprendi error is harmless, the determinative question
    is whether the evidence overwhelmingly establishes the amount of
    drugs distributed by the conspiracy as a whole.                     It does here.
    -89-
    b)   Multi-Object Conspiracy
    (Vega-Cosme)
    Vega-Cosme raises a related argument that the defendants
    were charged with a multi-object conspiracy.          In a multi-object
    conspiracy charge, a jury convicts the defendants of distributing
    one type of drug or another type of drug.      See, e.g., United States
    v. Dale, 
    178 F.3d 429
    , 431 (6th Cir. 1999) (jury instructed to
    convict if conspiracy distributed crack cocaine or marijuana).
    Although the First Circuit has not ruled on this issue, other
    circuits have held that when a defendant is charged with a multi-
    object conspiracy, and the jury returns a general verdict, the
    statutory maximum should be based on the object carrying the lowest
    maximum   penalty.   See,   e.g.,   
    id. at 432
    ;   United   States   v.
    Orozco-Prada, 
    732 F.2d 1076
    , 1083-84 (2d Cir. 1984).           Vega-Cosme
    argues that the jury in this case returned a general verdict on a
    multi-object conspiracy charge -- namely, that the defendants
    distributed more than five kilograms of heroin, cocaine, or cocaine
    base, or more than 100 kilograms of marijuana.        He argues that the
    statutory maximum should therefore have been based on the penalty
    for conspiring to distribute 100 kilograms of marijuana, which is
    up to forty years imprisonment under 
    21 U.S.C. § 841
    (b)(1)(B).          If
    this were indeed the case, Vega-Cosme's life sentence would be
    problematic.
    Vega-Cosme's argument, though, is without merit.             The
    defendants were not charged with a multi-object conspiracy.             The
    -90-
    indictment charged them with distributing more than five kilograms
    each of heroin, cocaine, and crack cocaine, and more than 100
    kilograms of marijuana.10      Because those drug quantities and types
    were    joined   by   the   conjunctive   term    "and"   rather     than   the
    disjunctive term "or," there was no ambiguity about the crime
    charged.    See United States v. Neuhausser, 
    241 F.3d 460
    , 469-70
    (6th Cir. 2001) (no Apprendi error in sentencing defendant to
    higher statutory maximum for cocaine conspiracy, when defendant was
    charged with conspiracy to distribute both cocaine and marijuana);
    United States v. Banks, 
    78 F.3d 1190
    , 1203 (7th Cir. 1996) (no
    ambiguity where indictment was phrased in conjunctive rather than
    disjunctive); United States v. Watts, 
    950 F.2d 508
    , 515 (8th Cir.
    1991) (same).
    Vega-Cosme argues that, regardless of the indictment, the
    jury    instructions    transformed   Count      Two   into   a   multi-object
    conspiracy charge.      The jury instructions contain a definition of
    "possession with intent to distribute" that required the government
    to "prove beyond a reasonable doubt that the defendant knew he was
    possessing a controlled substance" but not that "the defendant knew
    which particular controlled substance was involved."                Vega-Cosme
    contends that this instruction changed the conjunctive term "and"
    10
    The indictment did not need to specify the exact amount
    of drugs involved in the conspiracy, as long as it alleged the
    appropriate threshold amounts necessary to support the defendants'
    sentences. Cf. Derman v. United States, 
    298 F.3d 34
    , 42 n.4 (1st
    Cir. 2002).
    -91-
    in the indictment into the disjunctive term "or." This argument is
    meritless.       First, this definition should not have affected the
    jury's consideration of the conspiracy charge.               The conspiracy
    count of the indictment and the jury instructions regarding the
    elements of conspiracy require the jury to find that defendants
    conspired to "distribute" controlled substances to return a guilty
    verdict;     nowhere   does    the   term   "possession    with   intent   to
    distribute" appear.      Second, even if the term had appeared, the
    definition requires the government to prove that the defendants did
    in fact possess specific drugs, even if they did not know which
    drugs they possessed.         Thus, the government must still show that
    the defendant possessed cocaine and cocaine base and heroin and
    marijuana, even if the defendant himself did not know the specific
    drugs that he had in his possession.           Finally, even if a multi-
    object conspiracy were charged and an Apprendi error therefore
    occurred, Vega-Cosme admits that review would be for plain error
    because he did not preserve this argument below.            We have already
    determined that any Apprendi error as to drug amount or type would
    be harmless; a fortiori, no plain error occurred.
    c)     Failure to Reference § 841(b)(1)(B) in Indictment
    (Fernández-Malavé)
    Fernández-Malavé argues that his sentence must be vacated
    because the indictment was defective under Apprendi for failing to
    reference    specifically      §   841(b)(1)(B),    the   statutory   penalty
    subsection under which he was sentenced.           This claim is meritless.
    -92-
    The indictment included threshold drug quantities and types. There
    is no Apprendi requirement that the penalty subsection be included
    in the indictment once the drug quantity and type are alleged.                   See
    United States v. Eirby, 
    262 F.3d 31
    , 38 (1st Cir. 2001).
    2.     Sufficiency of Evidence As to Drug Quantities
    (Soto-Ramírez, Soto-Beníquez, Cintrón-Caraballo, Vega-
    Colón, Vega-Cosme)
    Five       defendants    argue,    separate   from    their   Apprendi
    claims, that the district court erred in determining the amount of
    drugs attributable to them.           They argue that the evidence does not
    establish   by     a    preponderance       the   quantity   or   type    of   drugs
    necessary to support the calculation of their base offense levels.
    We review factual determinations at sentencing for clear error.
    United States v. Damon, 
    127 F.3d 139
    , 141 (1st Cir. 1997).
    The district court's determinations of drug amounts were
    not clearly erroneous.              Under the Sentencing Guidelines, each
    defendant must be sentenced based on the amount of drugs that he
    handled, negotiated, saw, or could reasonably have foreseen to be
    embraced by the conspiracy.           Rodriguez, 162 F.3d at 149; U.S.S.G.
    § 1B1.3 & cmt. 2.           Applying this standard, the district court
    attributed at least 1.5 kilograms of crack cocaine to each of the
    five defendants, resulting in a base offense level of 38 for each.
    The    record     at    trial     and   sentencing    supports      this
    calculation. Negrón-Maldonado testified that Soto-Ramírez sold 300
    crack capsules to Cintrón-Caraballo on at least ten occasions in
    -93-
    1990, which amounts to 450 grams of crack cocaine.                   He also
    testified that at the beginning of 1990, Soto-Ramírez supplied him
    with 1000 to 1500 crack capsules, or about 150 to 200 grams, per
    week.   Assuming that these purchases continued for at least seven
    weeks, Soto-Ramírez would have distributed 1.5 kilograms of crack
    cocaine just from sales to Cintrón-Caraballo and Negrón-Maldonado
    in 1990.    Furthermore, according to Negrón-Maldonado's testimony,
    during early 1992, Negrón-Maldonado "cooked" one to three kilograms
    of cocaine into crack cocaine per week for Soto-Ramírez's points.
    In one and a half weeks, Negrón-Maldonado would have packaged at
    least 1.5 kilograms of crack cocaine for distribution at Soto-
    Ramírez's points.        Although Soto-Ramírez was in prison at this
    time, the drug quantity was reasonably foreseeable to him because
    he was still supervising his drug points by telephone.
    The government presented evidence that Soto-Beníquez
    supplied other members of the conspiracy with cocaine, which they
    converted    into   at   least   1.5    kilograms   of   crack   cocaine   and
    distributed at their respective drug points.               Negrón-Maldonado
    testified that each week in early 1992, he bought one to three
    kilograms of cocaine from Soto-Beníquez and converted it into crack
    cocaine.     Over two weeks, this would exceed the required 1.5
    kilograms.     Negrón-Maldonado also testified that Soto-Beníquez
    supplied Cintrón-Caraballo with 125 grams of cocaine per week and
    Negrón-Maldonado with 75 to 125 grams of cocaine per week, which
    -94-
    was converted into crack cocaine for sale at their respective drug
    points.     Over eight weeks, this would exceed the required 1.5
    kilograms.       Soto-Beníquez     protests    that   he   could    not    have
    reasonably foreseen that the cocaine he supplied would be converted
    into crack cocaine.       The district court had sufficient evidence to
    conclude     otherwise.      Negrón-Maldonado     testified       that    Soto-
    Beníquez's house was used, repeatedly, as a location for converting
    cocaine into crack cocaine.         He also testified that when Cosme-
    Sobrado was killed, Soto-Beníquez picked up the proceeds from Soto-
    Ramírez's    crack   cocaine    point   and   determined   that    money    was
    missing.      Bitumul is a small community; given Soto-Beníquez's
    involvement, it would be difficult for him to be wholly ignorant
    that his co-conspirators were producing crack cocaine.                    These
    activities and Soto-Beníquez's leadership role in the conspiracy
    are sufficient to support the conclusion that Soto-Beníquez knew
    Negrón-Maldonado and Cintrón-Caraballo sold crack cocaine at their
    points and that he could have reasonably foreseen that the cocaine
    he supplied to them would be converted into crack cocaine.
    Cintrón-Caraballo's crack cocaine point was in operation
    throughout the duration of the conspiracy, according to Negrón-
    Maldonado's testimony.         Negrón-Maldonado testified that Cintrón-
    Caraballo purchased over 90 grams of crack cocaine per week for
    distribution at his point in 1990, 125 grams per week in 1991, 125
    grams on a regular basis in 1992, and over 150 grams per week in
    -95-
    early 1993.    Assuming that Cintrón-Caraballo's drug point operated
    at least four weeks each year, the total amount of crack cocaine
    purchased for distribution would exceed 1.5 kilograms.
    Negrón-Maldonado testified that the drug point operated
    by Vega-Cosme and Vega-Colón sold crack cocaine in 1990, 1992, and
    1993.    Negrón-Maldonado estimated that one drug point distributing
    crack cocaine would usually sell at least one kilogram per month.
    That volume of sales alone is sufficient to exceed the required 1.5
    kilograms.       In    addition,    the     government     presented   evidence
    indicating that the quantity of crack cocaine sold by Negrón-
    Maldonado and Cintrón-Caraballo was reasonably foreseeable to Vega-
    Cosme.    Vega-Cosme met at least three times with the others to
    coordinate    the     color   of   the    caps   on    their   respective   crack
    capsules.     These meetings are evidence that Vega-Cosme had some
    awareness of his co-conspirators' crack cocaine sales and that,
    after color-coding was instituted, he had some way of tracking
    their activities.
    In response, all five defendants argue that this evidence
    is unreliable because it is based on the testimony of cooperating
    co-conspirators and was uncorroborated.               Vega-Colón and Vega-Cosme
    argue in their brief that absent a rule requiring corroboration of
    such evidence, "unsuspecting defendants would be entirely at the
    mercy of cooperating co-defendants, who have all the incentive in
    the world to testify in a manner [whether truthful or not] that
    -96-
    will assist the government in obtaining a larger sentence."                            That
    risk        is   real,    but    vacating        their    sentences       for   lack     of
    corroboration is not the answer.11                       Here, the cooperating co-
    defendants were vigorously cross-examined, and defense counsel had
    the    opportunity        to    present       evidence    of   the   witnesses's       plea
    agreements, grants of immunity, and receipt of government money.
    If the government's dilatory production of discovery materials had
    impeded the cross-examination, the situation might be different.
    But it did not.          The jury found the co-conspirators credible, and,
    for sentencing purposes, so did the trial court.                      These plausible
    credibility determinations cannot be disturbed on appeal.                               Cf.
    Torres-Galindo, 
    206 F.3d at 139-40
     ("Uncorroborated testimony of a
    cooperating accomplice may sustain a conviction so long as that
    testimony is not facially incredible . . . .").
    3.        Alleged Rule 32 Violation
    (Cintrón-Caraballo, Vega-Cosme)
    Cintrón-Caraballo argues that the district court erred
    (1) in permitting the government to introduce evidence in support
    of an upward adjustment for his role as a supervisor in the
    conspiracy        under    U.S.S.G.       §    3B1.1     and   (2)   in   granting     the
    adjustment.         Although the probation officer did not include the
    upward adjustment for a supervisory role in the PSR and the
    11
    If the government is mindful of its obligations,
    countervailing incentives, such as avoidance of perjury charges,
    can reduce the incentive to lie.
    -97-
    government did not object to this omission before sentencing, the
    government attempted to argue the upward adjustment to the court at
    the sentencing hearing.      We agree with Cintrón-Caraballo that this
    course of action violated Fed. R. Crim. P. 32, which requires that
    "[w]ithin 14 days after receiving the presentence report, the
    parties shall communicate in writing to the probation officer, and
    to each other, any objections" to it.         Fed. R. Crim. P. 32(b)(6)(B)
    (2000) (amended 2002).
    Any possible prejudice to Cintrón-Caraballo from the
    government's non-compliance was cured by the district court's grant
    of a two-week continuance to give defense counsel an adequate
    opportunity to respond to the government's late submission.                See
    United States v. Young, 
    140 F.3d 453
    , 457 (2d Cir. 1998) ("The
    sentencing   court    may   impose     sentencing   enhancements   belatedly
    suggested by the Government and not contained in the PSR, provided
    the     defendant     is    afforded     an   adequate       opportunity    to
    respond . . . ." (internal citation omitted)).
    Vega-Cosme raises a similar argument that the district
    court    erred   in   granting   a     two-level    upward   adjustment    for
    possession of a weapon.          Here again, the adjustment was not
    included in the PSR, and the prosecution did not object to its
    omission before sentencing but argued for the enhancement at
    sentencing, which was in violation of Rule 32.           Because Vega-Cosme
    did not object at sentencing, review is for plain error.              United
    -98-
    States v. Frisby, 
    258 F.3d 46
    , 47-48 (1st Cir. 2001).                           There was
    none. Vega-Cosme had the opportunity to respond to the evidence at
    the sentencing hearing. Contrast United States v. Curran, 
    926 F.2d 59
    , 62 (1st Cir. 1991), in which the court vacated a defendant's
    sentence when he had no opportunity to contradict letters that were
    not included in the PSR and that the court relied upon in reaching
    its decision.           Given the trial testimony regarding Vega-Cosme's
    role in obtaining ammunition for the conspiracy and the extensive
    murder       evidence    presented     at    trial,     the       government's    belated
    seeking of a firearms enhancement could not have come as such a
    surprise       to    Vega-Cosme   as    to       render       the   entire      sentencing
    proceeding a miscarriage of justice.
    4.      Denial of Downward Adjustment
    (Gonzalez-Ayala, de León Maysonet)
    Gonzalez-Ayala     and       de   León     Maysonet      argue     that   the
    district court committed an error of law when it refused to grant
    them     a    downward     adjustment       based       on    their     roles     as   minor
    participants in the conspiracy.              They argue in their brief that the
    district court "failed to realize that the guidelines permitted the
    sentencing court to decrease defendants' sentencing level" based on
    the fact that "the appellants' level of participation was below that
    of   the     other    defendants."          Mistakes         of   law   in   applying    the
    Sentencing Guidelines are reviewed de novo.                       United States v. Cali,
    
    87 F.3d 571
    , 575 (1st Cir. 1996).                     The district court made no
    -99-
    mistake of law.
    The     district   court    correctly    determined     that      these
    defendants were not entitled to a minor-role adjustment merely
    because they were the least culpable among those who were actually
    indicted.       See United States v. Daniel, 
    962 F.2d 100
    , 103 (1st Cir.
    1992).    The relevant inquiry is whether the defendant played a part
    that     made    him   substantially    less    culpable   than    the       average
    participant in similar crimes.          See U.S.S.G. § 3B1.2 cmt. 3; United
    States v. Brandon, 
    17 F.3d 409
    , 460 (1st Cir. 1994).
    Absent a mistake of law, we review the district court's
    fact-based       determination   that    a     defendant   was    not    a    minor
    participant for clear error.       United States v. Rosario-Peralta, 
    199 F.3d 552
    , 571 (1st Cir. 1999).           The court's determination was not
    clearly erroneous. The government presented testimony at trial that
    de León Maysonet stored weapons and narcotics for the conspiracy in
    1992, stood as an armed guard at drug points in 1993, and packaged
    and sold narcotics at the Callejón Nueve point in 1993.                         The
    government also presented evidence at trial that he participated in
    an unsuccessful mission to Fajardo to find and kill an individual
    named Vitito, who had been hired to kill those responsible for
    stealing the 200 kilograms of cocaine.            The evidence is sufficient
    to support the district court's finding that de León Maysonet was
    not a minor participant, based on his two-year involvement and his
    participation in a variety of criminal activities           in support of the
    -100-
    conspiracy.     As to Gonzalez-Ayala, government witnesses Negrón-
    Maldonado and Torrens-Alicea testified at trial that he participated
    in the planning and execution of the theft of 200 kilograms of
    cocaine in Fajardo and that he received profits from the sale of
    that cocaine.   They also testified that he helped package heroin and
    cocaine for distribution at the Callejón Nueve point.                   The record
    supports the district court's conclusion, based on the quantity of
    drugs he helped obtain for the conspiracy and his ongoing role in
    the packaging and sale of those drugs, that Gonzalez-Ayala was not
    a minor participant.
    5.    Grant of Upward Adjustment
    (Cintrón-Caraballo)
    Cintrón-Caraballo          argues     that       the   evidence      was
    insufficient to support the district court's grant of a three-level
    sentencing    enhancement   for   his     role    as    a    supervisor    in   the
    conspiracy    under   U.S.S.G.    §    3B1.1(b).        Under     the   Sentencing
    Guidelines, a three-level enhancement is permissible "[i]f the
    defendant was a manager or supervisor (but not an organizer or
    leader) and the criminal activity involved five or more participants
    or was otherwise extensive."          U.S.S.G. § 3B1.1(b).        Review of this
    determination is for clear error.         United States v. Brown, 
    298 F.3d 120
    , 122 (1st Cir. 2002).
    The district court correctly counted the eleven defendants
    convicted in the trial as meeting the "five or more participants"
    prong.    The more serious question is whether Cintrón-Caraballo was
    -101-
    a   manager      or   supervisor,    terms    not   defined   in    the   Sentencing
    Guidelines but described in U.S.S.G. § 3B1.1 cmt. 4 as involving:
    the exercise of decision making authority, the nature of
    participation in the commission of the offense, the recruitment
    of accomplices, the claimed right to a larger share of the
    fruits of the crime, the degree of participation in planning or
    organizing the offense, the nature and scope of the illegal
    activity, and the degree of control and authority exercised
    over others.
    Cintrón-Caraballo argues that there was no firm evidence that he was
    a supervisor; there was only rumor and innuendo.               But the government
    presented testimony at trial that Cintrón-Caraballo controlled a
    drug point at Street B that sold crack cocaine and that he had
    "Nanito, . . . Bennie's little brothers, and other persons" selling
    for him.      He also had Negrón-Maldonado cook cocaine into crack for
    his drug point.         The district court did not clearly err in finding
    that Cintrón-Caraballo acted as a supervisor in running his drug
    point.
    6.      Denial of Downward Departure
    (Soto-Ramírez)
    Soto-Ramírez challenges the district court's denial of a
    downward departure based on his upbringing.                 The record contained
    well-documented        evidence     that    Soto-Ramírez    had    suffered   severe
    neglect and sexual abuse as a child.                  A denial of a downward
    departure is generally non-reviewable unless the lower court's
    failure to depart stemmed from a misapprehension of its authority
    under      the    Sentencing      Guidelines.         See     United      States   v.
    -102-
    Rivera-Rodriguez, 
    318 F.3d 268
    , 275 (1st Cir. 2003).      This standard
    is unaffected by the PROTECT Act, which applies when the decision
    made is to grant a departure.      
    18 U.S.C. § 3742
    (e).    Here, Soto-
    Ramírez argues that the district court failed to recognize that it
    had the power to grant a downward departure based on abuse that
    Soto-Ramírez suffered as a child.        But the district court did
    acknowledge its power to depart.    It expressly stated, "I have [the]
    authority to depart because of an upbringing situation which may
    have affected the defendant."   Accordingly, we have no jurisdiction
    to review its decision on this issue.
    IV.
    This was a lengthy and complex case handled patiently and
    well by the trial court.      Despite missteps by the prosecution,
    defendants received a fair trial and sufficient evidence supported
    both the verdicts and the sentences for each defendant, which are
    affirmed.   So ordered.
    -103-