On July 25, 1978, Arnaldo Dario Rosado and Carlos Soto Arrivi were shot and killed by police on a mountain site in Puerto Rico known as Cerro Maravilla. The two men, both of whom were members of the Puerto Rico independence movement (independentistas), had gone to Cerro Maravilla apparently intending to blow up or otherwise sabotage a television tower located on the mountain. The police reported after the event that Rosado and Soto Arrivi met their death in a shootout while resisting arrest.

In part because of its political overtones, the Cerro Maravilla incident drew immediate and intensive scrutiny from politicians, the media, and the public.1 This scrutiny escalated as details emerged suggesting that the two independentistas may have been murdered, their deaths occurring after they had surrendered to the police. Although the attention given the shootings took many forms, the most visible were the hearings conducted by the Senate of Puerto Rico throughout much of 1983. The hearings, televised and widely followed throughout the island, reached their climax with the testimony of Jose Montanez Ortiz, Carmelo Cruz Arroyo, and Miguel Cartagena Flores. These men, members of the police force who were involved in the Cerro Maravilla operation, testified in substance that Rosado and Soto Arrivi were captured and fully subdued, only then to be beaten and ultimately killed by a shotgun blast fired at close range. Their testimony also indicated that the police went to Cerro Maravilla with orders to kill Rosado and Soto Arrivi, and that the officers agreed upon a false version of the events to present to the public. The media reported in great detail every revelation made at the hearings.2

As part of the examination of the shootings, two federal grand juries launched investigations into whether the police conduct had violated any federal statutes. From December 14, 1979 to November 18, 1983, appellants in this case, all of whom were members of the intelligence division of the Police of Puerto Rico and allegedly present at Cerro Maravilla when the shootings occurred, testified before the grand juries. During an overlapping period, October 18, 1979 through April 2, 1980, at least seven of the nine appellants were deposed in a civil action brought by the survivors of Rosado and Soto Arrivi in the United States District Court for the District of Puerto Rico alleging violations of the decedents' federally-protected rights.

On February 6, 1984, a federal grand jury for the District of Puerto Rico returned a 44 count indictment against appellants and Luis Reveron Martinez, a member of the police force whose appeal is not before us. Count 1 of the indictment charged appellants with violating 18 U.S.C. Sec. 371 by conspiring (1) to obstruct justice and a criminal investigation; (2) to give false testimony in depositions and before federal grand juries; and (3) to suborn perjury. The count alleged that appellants engaged in the conspiracy to "prevent the citizens of Puerto Rico and law enforcement authorities of Puerto Rico and the United States from learning that Arnaldo Dario Rosado and Carlos Soto Arrivi had been unlawfully brutalized and killed by officers of the Police of Puerto Rico...." The remaining counts of the indictment charged each appellant with two or more substantive counts of perjury, or suborning perjury, for testimony given before the federal grand juries or in depositions taken in the federal civil rights action, in violation of 18 U.S.C. Secs. 1621-1623.

In August of 1984, the district court granted appellants' motion for continuance, citing the extensive and prejudicial publicity that had surrounded the case. The court was concerned that in the ensuing months the Cerro Maravilla shootings would receive heightened attention because of the gubernatorial election scheduled for November of that year. Relying on evidence produced at the legislative hearings, some members of the Puerto Rico Senate, a body controlled by a political party hostile to incumbent governor Romero Barcelo, claimed that the government had participated in the cover-up of the shootings--a charge that propelled the Cerro Maravilla incident into the center of the 1984 gubernatorial campaign. The district court feared that until the elections were over, the degree of publicity would make it impossible for appellants to receive the fair and impartial trial guaranteed by the United States Constitution. The court set forth its reasons for these concerns in a comprehensive opinion. United States v. Perez-Casillas, 593 F. Supp. 794 (D.P.R.1984). Two months after the election, which resulted in the defeat of the incumbent governor, appellants moved for a second continuance, arguing that the prejudicial publicity had not abated, making a fair trial still impossible. The same judge who had granted the previous continuance denied, without opinion, any further one.

After an extensive voir dire beginning on February 5, 1985 and concluding on February 28, 1985, the trial began. On March 28, 1985, the jury returned verdicts of guilty on 36 of the 44 counts.3 Following sentencing, appellants filed a timely notice of appeal and now assign various errors. For the reasons hereinafter stated, we reverse each of Colon Berrios' convictions (Counts 1, 37, and 38), while we affirm as to the remainder.

All appellants contend that their due process rights were violated by the district court's refusal in January 1985 to continue the trial one more time. They argue that the Puerto Rico media had given so much attention to the investigation of the Cerro Maravilla shooting, and appellants' alleged cover-up, as to make it impossible for appellants to receive an impartial trial as guaranteed by the United States Constitution. We do not find reversible error.

We begin by agreeing with appellants and with our dissenting brother that public exposure in Puerto Rico to the details and personalities of the Cerro Maravilla affair was almost universal. When the district judge granted an initial continuance of the trial on August 22, 1984, she called "Cerro Maravilla the media event of the years 1983-1984." 593 F. Supp. 794, 798. The court went on to declare that the shootings were one of the most heavily debated issues of the local electoral campaign in 1980; that in 1983 the Senate of Puerto Rico had held widely followed, televised public hearings on the shootings in which a version of the facts emerged that contradicted the position adopted by the police officers; and, most importantly, that the press, radio, and television coverage of the story, including the senate investigation, was so extensive that "defendants, in effect, stood trial before an entire community and guilt was adjudicated without the constitutional protections afforded a criminal defendant in a court of law...." Id. at 805. Recognizing that the intensive publicity was unlikely to abate prior to the November 1984 gubernatorial election, the court declined to proceed with the trial in the summer of 1984, ordering a continuance until February 1985.

In January 1985, appellants moved for a second continuance, contending that because of the election, the public attention given the shootings had actually escalated. Not only had Cerro Maravilla been the central issue of the recently concluded campaign, the new governor, Hernandez Colon, who had charged his opponent with abetting the coverup allegedly carried on by defendants, was said in his inaugural address to have "proclaimed the 'entrapment and summary execution of two young independence supporters by police at Cerro Maravilla.' " Among the documents appellants presented to show that press coverage had not diminished significantly was a 17-page appendix listing by title and date the newspaper articles and television programs focusing on the shootings and related events that had appeared since the court's August 1984 order. Nevertheless, in a footnote order dated January 18, 1985, the district court denied appellants' motion and set the case for trial on February 5, 1985.

It would have been preferable for the district court, when denying the motion for a second continuance, to have stated its reasons, and, in particular, to have indicated in what respects the prejudicing conditions existing in August 1984 had abated. From the act of denial itself, however, we may assume that the judge, herself a resident of Puerto Rico and the same judge who had continued the case earlier after making extensive findings as to the atmosphere then, felt that the climate had sufficiently improved to permit a fair trial. By February 1985 the election, in which Cerro Maravilla played a major role, had ended. The trial, moreover, was preceded by a 17-day voir dire during which approximately 195 prospective jurors were questioned extensively as to his or her feelings concerning the Cerro Maravilla events and the defendants.4 During the voir dire the court had a further opportunity, by listening to the venire persons' responses and observing their demeanor, to determine whether an impartial jury could be obtained. Sixteen jurors, including alternates, were finally selected, all of whom indicated that they could render an impartial verdict. They were sequestered during the trial.

We believe, for reasons hereinafter discussed, that the trial itself and the court's refusal to grant a second continuance did not violate appellants' constitutional rights.

Analysis of the case law indicates that, for the publicity surrounding the Cerro Maravilla incident and the investigation to have infringed the protections afforded by the due process clause and the sixth amendment, appellants must demonstrate at least one of the following: (1) the trial itself was conducted in a "circus like" atmosphere; (2) the actual jurors who sat on the case possessed fixed opinions that prevented them from judging impartially the guilt or innocence of the defendants; or (3) the community was so saturated with inflammatory publicity as to call into question the jurors' assertions and require us to presume their partiality. We discuss in turn each of these theories as applied here.

The Supreme Court has indicated that it will vacate a conviction tied to a trial "entirely lacking in the solemnity and sobriety to which a defendant is entitled in a system that subscribes to any notion of fairness and rejects the verdict of a mob." Murphy v. Florida, 421 U.S. 794, 799, 95 S. Ct. 2031, 2036, 44 L. Ed. 2d 589 (1975); accord Sheppard v. Maxwell, 384 U.S. 333, 86 S. Ct. 1507, 16 L. Ed. 2d 600 (1966); Estes v. Texas, 381 U.S. 532, 85 S. Ct. 1628, 14 L. Ed. 2d 543 (1965). In both Estes and Sheppard, the impact of the media's presence and conduct in the courtroom was found to have created a circus-like atmosphere that deprived defendants of the "judicial serenity and calm" essential to a fair trial. The Court took note of the judge's failure, in each case, to shield the jurors from public pressure once the trial began. See Sheppard, 384 U.S. at 353, 86 S.Ct. at 1517 ("the judge's failure to insulate [the jurors] from reporters and photographers ... exposed them to expressions of opinion from both cranks and friends"); Estes, 381 U.S. at 545, 85 S.Ct. at 1634 (televised trial exposed jurors to "pressures of knowing that friends and neighbors have their eyes upon them," and the pressure of knowing that they must "return to neighbors who saw the trial themselves").

Nothing in appellants' trial resembled the conditions criticized in Estes and Maxwell. Insofar as the record can reveal, the proceedings were conducted with dignity and impartiality, containing none of the circus-like aspects the Supreme Court found troubling in Estes and Maxwell. By sequestering the jury throughout the trial, the court shielded the jurors during the period of the trial itself from exposure to media accounts and communications with and from outsiders. There is absolutely no reason to believe, therefore, that any untoward community passions spilled over into the courtroom so as to infect the trial process, or reached the jurors during the trial so as to influence them while they assessed the evidence.

The sixth amendment to the United States Constitution guarantees a criminal defendant the right to be tried "by a panel of impartial, 'indifferent' jurors." Irvin v. Dowd, 366 U.S. 717, 722, 81 S. Ct. 1639, 1642, 6 L. Ed. 2d 751 (1961). This standard, however, does not require that

the jurors be totally ignorant of the facts and issues involved. In these days of swift, widespread and diverse methods of communication, an important case can be expected to arouse the interest of the public in the vicinity, and scarcely any of those best qualified to serve as jurors will not have formed some impression or opinion as to the merits of the case. This is particularly true in criminal cases. To hold that the mere existence of any preconceived notion as to the guilt or innocence of an accused, without more, is sufficient to rebut the presumption of a prospective juror's impartiality would be to establish an impossible standard. It is sufficient if the juror can lay aside his impression or opinion and render a verdict based on the evidence presented in court.

Id. at 722-23, 81 S.Ct. at 1642-43; see also Patton v. Yount 467 U.S. 1025, 1037 n. 12, 104 S. Ct. 2885, 2891 n. 12, 81 L. Ed. 2d 847 (1984) (the Constitution requires only that the juror "can lay aside his opinion and render a verdict based on the evidence presented in court"); Dobbert v. Florida, 432 U.S. 282, 303, 97 S. Ct. 2290, 2303, 53 L. Ed. 2d 344 (1977) ("extensive knowledge in the community of either the crimes or the putative criminal is not sufficient by itself to render a trial constitutionally unfair"); id. ("one who is reasonably suspected" of committing a heinous crime "cannot expect to remain anonymous"); Murphy v. Florida, 421 U.S. 794, 800, 95 S. Ct. 2031, 2036, 44 L. Ed. 2d 589 (1975) (jurors need not "be totally ignorant of the facts and issues involved"). As Justice Clark recognized in Rideau v. Louisiana, 373 U.S. 723, 733, 83 S. Ct. 1417, 1423, 10 L. Ed. 2d 663 (1963) (Clark, J., dissenting), "it is an impossible standard to require [a] tribunal to be a laboratory, completely sterilized and freed from any external factors."

Here, the court conducted a voir dire of nearly 200 potential jurors which lasted 17 days and is recorded in over 3,000 transcript pages. (The court had excused another 55 venire persons upon personal hardship requests.) The voir dire was conducted by the judge, in the presence of the defendants and counsel; potential jurors were summonsed individually into the courtroom and, besides the usual questions, were examined extensively as to their knowledge of the Cerro Maravilla affair and their attitude toward those involved.5 A jury was finally selected that we must assume satisfied the district court as meeting the constitutional requirements of impartiality.6 The trial court's determination as to the impartiality of jurors may be set aside only for manifest error. See Patton, 467 U.S. at 1031, 104 S.Ct. at 2888; Irvin, 366 U.S. at 723-24, 81 S.Ct. at 1642-44; United States v. McNeill, 728 F.2d 5 (1st Cir.1984). The "special deference" given to a trial court's assessment of juror impartiality derives from the fundamental role "[d]emeanor plays ... not only in determining juror credibility but also in simply understanding what a potential juror is saying." Patton, 467 U.S. at 1038 & n. 14, 104 S. Ct. at 2892 & n. 14. See also Reynolds v. United States, 98 U.S. (8 Otto) 145, 156-57, 25 L. Ed. 244 (1879) ("the manner of the juror while testifying is often times more indicative of the real character of his opinions than his words.... Care should, therefore, be taken in the reviewing court not to reverse the ruling below ... except in a clear case").

Our review of the voir dire testimony of the 12 jurors who sat on the case reveals no manifest error. To be sure, virtually all of the empanelled jurors testified to having been exposed to news or telecasts of the Puerto Rico Senate's hearings on the shootings. None of their voir dire testimony, however, reveals significant bias against appellants. One juror appeared to have some prior suspicion of a cover-up by the police based on the televised hearings, despite his professions of impartiality.7 On the other hand, two jurors expressed a generalized kind of sympathy toward the defendants,8 and one other juror expressed a generalized doubt about the tactics of the defendants' victims.9 The voir dire testimony of the remaining jurors shows no evidence of any preconceived opinions.10 All 12 jurors testified repeatedly on voir dire that they would base their deliberations solely on the evidence adduced at trial, and that they could impartially reach a verdict. On this record, we do not find manifest error in the district court's determination to seat the jurors as impartial, nor does the record indicate in any way that they lacked impartiality.

Given the propriety and decorum with which the trial was conducted, the sequestration of the jurors, and the apparent impartiality of the selected jurors as revealed in their individual voir dire responses, the only remaining question is whether the publicity surrounding the Cerro Maravilla affair created an environment so irremediably tainted that we must hold, as a matter of law, that appellants were denied a fair trial. Our dissenting brother thinks this is so. He believes that even though appellants did not seek or wish a change of venue, they should have been granted a requested second continuance, and that denial of a continuance was a reversible error.11

As already stated, we readily accept that this case involves a publicity issue of unusual seriousness, presenting as it did charges of cold-blooded murder and police corruption that became causes celebres throughout Puerto Rico. Being a compact, insular community, Puerto Rico is highly susceptible to the impact of local media. Where, as here, an issue touches on fundamental political disagreements, involves a televised legislative investigation and charges of a government coverup, and becomes a focal point of an animated gubernatorial campaign, public attention is focused even more heavily on the island's various media.

The issue, however, is whether the effect of the publicity, widespread as it undoubtedly was, was such as to foreclose appellants' right to a fair trial. We see no convincing evidence of this. Were we to accept our dissenting brother's position, appellants, who did not seek such a change of venue and made it clear on appeal they did not want one, could be insulated from trial indefinitely.

In most cases where the Supreme Court or lower courts have reversed because of publicity surrounding a criminal trial, there has been specific proof of a tainted trial process or of prejudiced jurors. The Supreme Court has indicated, however, that in very extraordinary circumstances it may disregard the jurors' professions of impartiality and presume prejudice when the community has been overcome by widespread, highly inflammatory publicity. See, e.g., Dobbert v. Florida, 432 U.S. 282, 303, 97 S. Ct. 2290, 2303, 53 L. Ed. 2d 344 (1977) ("in the absence of a 'trial atmosphere ... utterly corrupted by press coverage' ", defendant must point to "specific portions of the record ... which would require a finding of constitutional unfairness") (citation omitted) (emphasis supplied). As an indirect means of determining whether community prejudice resulting from publicity may have unconsciously infected the jurors who were seated, the Court has sometimes noted how many non-seated members of the venire admitted to a disqualifying prejudice. As the Court stated in Murphy v. Florida, 421 U.S. 794, 803, 95 S. Ct. 2031, 2037, 44 L. Ed. 2d 589 (1975), "[i]n a community where most veniremen will admit to a disqualifying prejudice, the reliability of the others' protestations may be drawn into question; for it is then more probable that they are part of a community deeply hostile to the accused, and more likely that they may unwittingly have been influenced by it." Thus, in Irvin v. Dowd, 366 U.S. 717, 81 S. Ct. 1639, 6 L. Ed. 2d 751 (1961), the Court reversed a murder conviction after finding a "pattern of deep and bitter prejudice" in the community. As support for this conclusion, the Court noted not only that eight of the twelve jurors had, before trial, admitted to believing the defendant guilty, but that 90 percent of the venire members examined on the point admitted to having some opinion as to guilt, "ranging in intensity from mere suspicion to absolute certainty." Id. at 727, 81 S.Ct. at 1645.

In the present case, none of the seated jurors admitted at the voir dire to any such belief that defendants were guilty; and our review of the voir dire record relative to those not seated indicates that the percentage of prejudiced venire members was well below the threshold that would justify presuming bias on the part of seated jurors despite the latter's professions of impartiality. Our dissenting brother quotes impressively from the responses of some of the more prejudiced venire members, (who, of course, did not sit as jurors), but our review indicates that of those venire members whose responses can be adequately construed on whether they had an opinion as to appellants' guilt or innocence, approximately 25 percent admitted, in varying degrees, to believing that defendants were guilty. Roughly ten percent of the venire suggested some belief that appellants may be innocent.12 (Virtually all members of the venire admitted to some prior knowledge of the Cerro Maravilla incident and the subsequent investigation.)

Although these statistics are inherently imprecise, based as they are on the often subjective task of categorizing voir dire testimony, they provide a useful estimate of venire attitudes. In Murphy v. Florida, the Court found that 20 of the 78 venire members questioned--roughly 25 percent--were excused because they indicated an opinion as to petitioner's guilt. In affirming the conviction, the Supreme Court stated that "[t]his may indeed be 20 more than would occur in the trial of a totally obscure person, but it by no means suggests a community with sentiment so poisoned against petitioner as to impeach the indifference of jurors who displayed no animus of their own." 421 U.S. at 803, 95 S.Ct. at 2038. By the same token, we cannot conclude from the approximately one-quarter of the venire here who admitted to believing in appellants' guilt that the community as a whole was so prejudiced against defendants by inflammatory pretrial publicity as to call into question the seated jurors' assertions of impartiality.

Perhaps the strongest support for appellants' argument that we must presume prejudice from the publicity is to be found in one case, Rideau v. Louisiana, 373 U.S. 723, 83 S. Ct. 1417, 10 L. Ed. 2d 663 (1963). Rideau is the only case in which the Supreme Court has reversed a conviction based solely on the egregiousness of pretrial publicity and without specific proof that the jurors who sat--or the trial itself--were prejudiced thereby.13 Rideau involved a defendant, Wilbert Rideau, who robbed a bank in Lake Charles, Louisiana, kidnapped three of the bank's employees, and killed one of them. After defendant was apprehended, he confessed on film to having committed the crime. On the day he confessed and each of the next two days, a Lake Charles television station aired the film, reaching anywhere from one-third to two-thirds of the community's population. Following his arraignment, Rideau moved for a change of venue, which the state trial court denied. Rideau was subsequently convicted and sentenced to death.

The Supreme Court overturned the conviction, holding specifically "that it was a denial of due process of law to refuse the request for a change of venue." Id. at 726, 83 S.Ct. at 1419. The Court reasoned that given the tens of thousands of people who saw the detailed confession, "[a]ny subsequent court proceeding in a community so pervasively exposed to such a spectacle could be but a hollow formality." Id. The Court presumed prejudice, finding it unnecessary to review the transcript of the voir dire or to take any other step to ascertain the effect the televised confession had on community sentiment.

We do not believe Rideau controls the instant appeal. First, Rideau and the other publicity cases involved defendants and crimes which the community could only view with total abhorrence. Here, however, the charged conduct, violent and brutal though it was, would not cause appellants to be seen in black and white terms by all members of the community equally. The Cerro Maravilla victims were themselves alleged terrorists, who, in going to Cerro Maravilla to blow up a television tower, adopted both means and goals many people would find distasteful. Defendants were police of previously good repute, not members of the criminal element like Rideau. The case was highly politicized, undoubtedly causing some in the community to view the senate investigation as a politically motivated proceeding whose findings were inherently suspect. Moreover, the question of who was at fault--the defendant police officers; their supervisors; the incumbent governor, Romero Barcelo, who was accused of covering up the incident; or the terrorists--was likely a matter both of considerable doubt and difference of opinion. Thus, it would be wrong to assume that the Puerto Rico public responded uniformly to the publicity and, as one person, stood ready to convict.14 The jurors could well have seen the trial as a genuine opportunity to hear the evidence in a reliable setting and determine, at last, the truth.

Second, the publicity present in Rideau was, in the context of the trial, more prejudicial than the publicity at issue here. In Rideau, the community was exposed within a few weeks of trial to televised replays of defendant's own confession given soon after the kidnapping and murder. In the instant case, the public was exposed not to confessions by the defendants, but to incriminating testimony by witnesses who were granted immunity. The senate hearings that elicited this testimony occurred five years after the shootings at Cerro Maravilla and two years before the trial. These facts, while not diminishing the seriousness of the questions raised by the extensive and continuous publicity, suggest that the overwhelming lynch-mob atmosphere described in Rideau was not present. In the absence of such an atmosphere, and convinced that the trial itself was conducted fairly, we are reluctant to make a conclusive presumption of prejudice, against the evidence of the empanelled jurors' voir dire and the conclusion of the district court that the jury was not biased against defendants.

Finally, unlike Rideau, the question here is not whether it was constitutional error for a court to refuse a requested change of venue. No such change was sought, and appellants have made it clear on appeal that they did not desire one. A change of venue would have ensured a trial before jurors without previous knowledge of the Cerro Maravilla affair.15

It is true that this circuit earlier observed that "Puerto Rico is singularly unsuited to a change of venue...." In re San Juan Star Co., 662 F.2d 108, 117 (1st Cir.1981). But this observation, in the context made, was not a major pronouncement. At issue in San Juan Star was the validity of protective orders issued by the district court at an earlier stage in the civil rights case arising from this same incident. Because the orders limited the press's access to deposition testimony, the court's review included an inquiry as to whether measures less restrictive of first amendment rights could preserve the fairness of any upcoming criminal trial. In that context, the court made the quoted observation. In the same sentence, the court noted the unsuitability of granting a continuance: "postponement of a trial of such urgent proportions could seriously jeopardize important interests in its resolution." 662 F.2d at 117. Given the latter reference, it is clear the venue comment in San Juan Star falls short of supporting appellants' argument that, in Puerto Rico, a continuance is the preferred mechanism for shielding criminal trials from the effects of adverse publicity. While we recognize the problems associated with a change in venue, it remains a feasible option for a Puerto Rico accused confronted with publicity at home.

Courts have held that a motion for a change of venue is the preferred remedy where a community has been saturated with publicity adverse to the defendant. Finnegan v. United States, 204 F.2d 105, 110 (8th Cir.) ("[N]ewspaper publicity tending to excite public prejudice against a defendant is not usually considered as a sufficient reason for granting an application for continuance.... [T]here was no application for a change of venue, which is usually considered the correct remedy in such situations."), cert. denied, 346 U.S. 821, 74 S. Ct. 36, 98 L. Ed. 347 reh. denied, 346 U.S. 880, 74 S. Ct. 118, 98 L. Ed. 387 (1953). See also Dennis v. United States, 302 F.2d 5, 8 (10th Cir.1962) ("In ruling on Travis' pretrial motions for severance and continuance on the grounds of prejudicial publicity, the trial court indicated that the proper remedy was change of venue, and invited such motion."), rev'd on other grounds, 384 U.S. 855, 86 S. Ct. 1840, 16 L. Ed. 2d 973 (1966). Cf. United States v. Pfingst, 477 F.2d 177, 185-86 (2d Cir.) (fact that "[a]ppellant never moved for a change of venue" supports trial judge's denial of motion for new trial based on prejudicial publicity), cert. denied, 412 U.S. 941, 93 S. Ct. 2779, 37 L. Ed. 2d 400 (1973).

Application for a continuance rather than a change of venue is particularly disfavored where, as here, there is little reason to believe that the prejudicial publicity complained of will abate within a foreseeable period:

The denial of a motion for an indefinite or substantial continuance predicated upon widespread adverse pretrial publicity about a defendant is all the more warranted when, as here, there is sound reason to believe that the defendant will continue to be a controversial, publicity-invoking figure and, hence, that there is little assurance that the passage of time will result in an abatement or subsidence of critical publicity in the foreseeable future.

United States v. Hoffa, 156 F. Supp. 495, 500 (S.D.N.Y.1957). Cf. United States v. Marcello, 280 F. Supp. 510, 519 (E.D.La.1968) ("[D]ue to the continuing nature of the prejudicial publicity in this area, ... nothing is gained by a continuance under such circumstances, except to delay the inevitable, and it is improper to grant the continuance in such a situation because it does not obviate the difficulty.").16

We are aware of only one case in which the district court's refusal to grant a continuance led the circuit court to order a new trial, Delaney v. United States, 199 F.2d 107 (1st Cir.1952). Unlike the instant case, Delaney involved "massive pre-trial publicity, on a nationwide scale," id. at 113, stemming from public hearings conducted by the United States Congress about a series of similar crimes around the country. There, a change of venue would have provided no relief from the effects of adverse publicity, such that, "under the circumstances of this case we do not think that the defendant's appeal stands any worse for failure on his part to apply for a change of venue." Id. at 116.

In addition, the source of the court's order in Delaney lay not in "the rock-bottom requirements of the due process clause of the fourteenth amendment," but in the appellate courts' supervisory power for "establishing and maintaining civilized standards of procedure and evidence" in the federal courts. Id. at 113. In choosing to exercise its supervisory power, the court placed great emphasis on the fact that the hearings giving rise to the prejudicial publicity had been staged by a coordinate branch of the same government to which the trial court belonged:

If the United States, through its legislative department ... chooses to hold a public hearing inevitably resulting in such damaging publicity prejudicial to a person awaiting trial on a pending indictment, then the United States must accept the consequence that the judicial department, charged with the duty of assuring the defendant a fair trial before an impartial jury, may find it necessary to postpone the trial until by lapse of time the danger of the prejudice may reasonably be thought to have been substantially removed.

Id. at 114. By contrast, the legislative hearings on the Cerro Maravilla shootout were staged by the Senate of Puerto Rico, and they were initiated before any of the defendants had been indicted. The pattern of official conduct that led this court in Delaney to exercise its supervisory power is therefore less evident here.17

It is, of course, true that the sixth amendment entitles a criminal accused to a "speedy and public trial, by an impartial jury of the State and district wherein the crime shall have been committed." U.S. Const. amend. VI. One can argue that an accused is not required to yield his constitutional right to be tried in the vicinage in order to secure another constitutional right, trial before an impartial jury. But this argument is not dispositive in circumstances like these. See United States ex rel. Darcy v. Handy, 351 U.S. 454, 463, 76 S. Ct. 965, 970, 100 L. Ed. 1331 (1956) (although not dispositive, petitioner's failure to move for a change of venue as a means "to prevent the drawing of an unfair trial jury from a community allegedly infected with hysteria and prejudice" is significant); Stroble v. California, 343 U.S. 181, 193-94, 72 S. Ct. 599, 605-06, 96 L. Ed. 872 (1952). The sixth amendment also guarantees a speedy trial, a right to which the government, as well as the accused, has a claim. Government--society itself--cannot be utterly deprived of the right to prosecute and try someone for a crime within a reasonable time merely because of widespread community knowledge.

If an accused in a situation such as the present seeks a change of venue, judicial fairness may require that it be granted. As in Rideau, should the request be denied, this could be a denial of due process of law. But if an accused expressly declines to seek a venue change, we think he carries a significantly heavier burden to show that widespread community publicity concerning the crimes of which he is charged render his trial presumptively unfair--that, because of the publicity alone, the jurors drawn from the district where he insists he has a right to be tried are presumptively unfit to meet the "impartial jury" standard of the sixth amendment. Society cannot be placed in a position where it is unable to bring to trial the violators of its laws.

Here, although the evidence shows intense publicity, it is not, given the factors discussed above, so extreme as to warrant our presuming prejudice solely from the nature of the publicity.18

In reviewing the district court's refusal to grant a second continuance, we operate under an "abuse of discretion standard." See, e.g., United States v. Gullion, 575 F.2d 26, 28 (1st Cir.1978). After carefully reviewing the record, we conclude that (1) the trial was conducted in a proper and decorous manner; (2) the jurors' voir dire testimony showed no evidence of disabling bias; and (3) neither the voir dire answers of other venire members nor the nature of the publicity itself warrants our presuming prejudice. Accordingly, we hold that the district court did not abuse its discretion in denying appellants' motion for a second continuance.

On March 2, 1985, the day opening statements were made, Rios Polanco moved to exclude from evidence details pertaining to the beatings and actual killing of the two independentistas. The district court denied the motion. The court later refused a similar request made by Rios Polanco, Colon Berrios, Quiles Hernandez, Mateo Espada, and Perez Casillas. Appellants claim error, asserting that the details of the murder (1) were not relevant to a prosecution for perjury, and (2) even if relevant, were so prejudicial as to require being excluded. We disagree.19

The evidence at issue was clearly relevant. A number of the counts charged appellants with lying about specific aspects of the shootings. For example, Count 6 alleged that, induced by Perez Casillas, Miguel Marte Ruiz perjured himself before a grand jury in claiming to have heard only one set of shots. To prove this charge, the government had to demonstrate, among other things, that at least two rounds of shots were fired. In theory, this burden could have been satisfied by witnesses testifying merely that they heard a second volley. To make such testimony persuasive, however, the government needed to explain why two rounds were fired, which required going into the details of the killings. See United States v. Mills, 704 F.2d 1553, 1559 (11th Cir.1983) (district court did not abuse its discretion by admitting evidence of "illegal or otherwise improper acts which did not constitute elements of the crime charged in the indictment, but [which] pertained to a chain of events forming the context, motive, and set-up of the crime"), cert. denied, 467 U.S. 1243, 104 S. Ct. 3517, 82 L. Ed. 2d 825 (1984).

Count 19 alleged that Rafael Torres Marrero committed perjury in an March 12, 1980 deposition when he claimed, among other things, that Soto Arrivi, after being shot, fell down a steep ravine. To satisfy its burden of proving falsity, the government had to demonstrate where the body in fact lay, which required reference to the details of the shootings (e.g., that the victims were allegedly kneeling together, far from any ravine, when killed by a shotgun blast fired at close range).

Finally, Count 24 charged Luis Reveron Martinez with making false material declarations before the grand jury in testifying that the two independentistas "never stopped resisting. They were at all times shooting against us and we were defending ourselves." Obviously, it was relevant to proof of falsity for the government to show that, although the victims had been kneeling and begging for mercy, they were beaten and subsequently killed.20

Evidence may, of course, be relevant but unfairly prejudicial. Fed.R.Evid. 403. Whether this is so is a matter left, within broad limits, to the discretion of the trial court. See, e.g., United States v. Kadouh, 768 F.2d 20, 21 (1st Cir.1985); United States v. Kepreos, 759 F.2d 961, 964 (1st Cir.), cert. denied, --- U.S. ----, 106 S. Ct. 227, 88 L. Ed. 2d 227 (1985). The question is one of "unfair" prejudice--not of prejudice alone. See, e.g., Dollar v. Long Manufacturing, N.C., Inc., 561 F.2d 613, 618 (5th Cir.1977) ("virtually all evidence is prejudicial or it isn't material. The prejudice must be 'unfair' "), cert. denied, 435 U.S. 996, 98 S. Ct. 1648, 56 L. Ed. 2d 85 (1978). Here, appellants were accused of engaging in specific acts of perjury as part of a scheme intended to cover up the shootings at Cerro Maravilla. We cannot say that it was "unfair" to permit the jury to become acquainted with the government's evidence pertaining to the shootings--brutal as the details were--where they were so closely intertwined with the acts of perjury charged. Cf. Scales v. United States, 367 U.S. 203, 255-56, 81 S. Ct. 1469, 1499-1500, 6 L. Ed. 2d 782 (1961). Given the direct relevance of the evidence to the government's case, the district court did not abuse its discretion by denying appellants' motion.

While the jury was being selected, Colon Berrios, Rios Polanco, Bruno Gonzalez, Gonzalez Perez, Quiles Hernandez, Mateo Espada, Torres Marrero, and Perez Casillas each moved for severance. The events covered by the perjury indictments did not indicate that these particular defendants had engaged in the alleged beating and shooting of the two independentistas. Appellants argued that their right to a fair trial would be jeopardized by the "spillover effect" of being tried with the officers supposed to have engaged in such savage acts. The district court denied the motions, finding that (1) they were not timely filed and (2) even if timely, they lacked merit.

In support of its conclusion that these appellants had waived the right to request severance, the court noted that Fed.R.Crim.P. 12(b)(5) states explicitly that motions for severance must be raised prior to trial. The failure to satisfy that deadline "shall constitute waiver thereof, but the court for cause shown may grant relief from the waiver." Fed.R.Crim.P. 12(f). Here, appellants contend that the dangers of a joint trial did not become apparent until after the voir dire had commenced. The voir dire, they claim, revealed that the venire members were focusing on the more violent aspects of the case, thereby raising the specter of "guilt by association." The district court rejected this argument, holding that

Although [the argument] is relevant to a request for a special instruction such as that already included in the individualized voir dire, in a case where no one is charged with murder this is certainly no reason to order severance of the trials.

Thus, the court refused to "grant relief from waiver" as allowed under Fed.R.Crim.P. 12(f).

In reviewing the district court's finding of waiver we operate under an abuse of discretion standard. See United States v. Brooks, 567 F.2d 134, 138-39 (D.C.Cir.1977); see generally United States v. Greenleaf, 692 F.2d 182, 187 (1st Cir.1982) (whether to grant a severance is lodged in the trial court's discretion), cert. denied, 460 U.S. 1069, 103 S. Ct. 1522, 75 L. Ed. 2d 946 (1983); United States v. Perez, 648 F.2d 219, 224 (5th Cir.), cert. denied, 454 U.S. 970, 102 S. Ct. 516, 70 L. Ed. 2d 388 (1981) (same). Here, a number of reasons support the court's finding of waiver. Most importantly, appellants could not have been surprised at the type of evidence the government intended to produce. Bruno Gonzalez's motion admits that "the Indictment ... did foreshadow the confussion [sic] of the evidence and the clear prejudice" to him.

Nonetheless, we would hesitate to uphold the denial of severance on grounds of waiver if compelling interests of justice otherwise required. See Schaffer v. United States, 362 U.S. 511, 516, 80 S. Ct. 945, 948, 4 L. Ed. 2d 921 (1960) ("judge has a continuing duty at all stages of the trial to grant a severance if prejudice does appear"). In the case at bar, however, we do not believe that the court's refusal to sever "deprived defendants of a fair trial, resulting in a miscarriage of justice." United States v. Albert, 773 F.2d 386, 388 (1st Cir.1985). The district court found that:

Although none of the defendants are [sic] charged with murder they are accused of conspiring to conceal a murder. The government cannot meet its burden and rest its case merely by proving that defendants had combined and "conspired to obstruct justice." The underlying premise behind the charges described in the indictment requires proof that the government's version of how things happened is the true one. It must prove beyond a reasonable doubt that the events in Cerro Maravilla occurred as it alleges they did occur and that all defendants knowingly conspired to hide this from the authorities conducting investigations. Proof of such events would have to be presented against each defendant even if each one were to be tried separately.

Order Denying Motion for Severance, at 3.

The district court did not abuse its discretion in drawing this conclusion. In United States v. Ciampaglia, 628 F.2d 632 (1st Cir.), cert. denied, 449 U.S. 956, 101 S. Ct. 365, 66 L. Ed. 2d 221 (1980), we held that where multiple defendants are being tried on charges directly related to the same overall scheme and transaction, and "where separate trials would have necessarily involved repetitive use of most of the same evidence and same facts, we find no possibility of [an abuse of discretion] absent a clear showing of substantial prejudice." Id. at 643; see also United States v. Crooks, 766 F.2d 7, 10 (1st Cir.), cert. denied, --- U.S. ----, 106 S. Ct. 421, 88 L. Ed. 2d 362 (1985); United States v. Cleveland, 590 F.2d 24, 29 (1st Cir.1978) ("Defendants must show a very significant degree of prejudice to overcome society's interest in conducting joint trials....").

As the district court noted, the government was required to show that its version of the events was substantially true in order to support a conviction under Count 1, which charged all defendants with conspiracy to cover-up the events of Cerro Maravilla by committing perjury. The government could not prove the existence of such a coverup unless it first showed what actually occurred at Cerro Maravilla. Thus, even in a separate trial, the general details of the beatings and the murders, at a minimum, would have been admissible against all defendants.

To be sure, being tried with defendants who did the shooting meant that the jury learned of the greater criminality of those associates. But the mere possibility of this kind of spillover is not ordinarily thought a sufficiently real danger to require a severance. In United States v. Cleveland, 590 F.2d 24 (1st Cir.1978), three defendants were tried together for armed robbery. Because one of the defendants had made admissible confessions, the other defendants sought severance. The district court refused to sever and we affirmed, noting that the risk of spillover is seldom sufficient to warrant separate trials. Id. at 28-29; see United States v. Arruda, 715 F.2d 671, 679 (1st Cir.1983) (allegation of guilt by association inadequate to render denial of severance an abuse of discretion). In drawing this conclusion, we cited with approval a Second Circuit decision that upheld a lower court's refusal to sever even though, conceivably, defendants might have had a better chance of acquittal in a separate trial. Cleveland, 590 F.2d at 28; see also United States v. Martinez, 479 F.2d 824, 828 (1st Cir.) ("prejudice means more than just a better chance of acquittal at a separate trial").

In addition to the general admissibility of the evidence against all appellants and the strong presumption in favor of joint trials, two other factors lead us to uphold the district court's refusal to sever. First, the court provided a clear instruction warning the jury to consider each count separately. See United States v. Greenleaf, 692 F.2d 182, 187 (1st Cir.1982), cert. denied, 460 U.S. 1069, 103 S. Ct. 1522, 75 L. Ed. 2d 946 (1983); United States v. Escalante, 637 F.2d 1197, 1201-02 (9th Cir.), cert. denied, 449 U.S. 856, 101 S. Ct. 154, 66 L. Ed. 2d 71 (1980). Second, the jury, by returning acquittals on 8 of the 44 counts, "reassures us that [it] complied with the court's directives." Greenleaf, 692 F.2d at 187. On this record, we conclude that the district court committed no reversible error in denying appellants' motion for severance.

IV. SUFFICIENCY OF EVIDENCE TO SUPPORT PERJURY CONVICTIONS

Individual appellants take exception on various grounds to their convictions under the different perjury counts.21 These challenges include claims that, in some counts, a defendant's statement was literally true (whatever other implications it may have suggested), and hence nonperjurious. As to other counts, appellants claim that the supposed perjurious responses lacked materiality to the inquiry in which they were rendered. And as to still others, some appellants challenge outright the legal sufficiency of the prosecution's evidence. Other errors are also charged.

We now address these matters individually. We affirm all except the convictions of Colon Berrios under Counts 37 and 38. We discuss Colon Berrios's convictions first.

The indictment charged Colon Berrios with three counts of perjury allegedly committed during his January 16, 1980 deposition in the civil action brought by Rosado's and Soto Arrivi's survivors. At trial, following the government's case-in-chief, Colon Berrios moved for a judgment of acquittal pursuant to Fed.R.Crim.P. 29. The district court denied the motion, an action it repeated when Colon Berrios renewed his motion after defendants rested. After two days of deliberations, the jury acquitted Colon Berrios on Count 36, but returned verdicts of guilty on Counts 37 and 38. He asserts on appeal that the government produced insufficient evidence to support his convictions under the two counts. Because our analysis of Count 37 depends in part on whether the conviction under Count 38 stands, we shall begin with the latter.

In his deposition, after being asked "who was your normal supervisor," Colon Berrios responded "Lieutenant Gonzalez." (Emphasis added.) Colon Berrios then answered "no" when asked whether "Gonzalez [had] anything to do with the operation at Cerro Maravilla," a contention the government alleges is perjurious. As stated in the indictment, "Nelson Gonzalez Perez was present at Cerro Maravilla on July 25, 1978, when the shots were fired which killed Carlos Soto Arrivi and Arnaldo Dario Rosado." Although the indictment assumed that "Lieutenant Gonzalez" and "Nelson Gonzalez Perez" were the same person, an assumption shared by the government at trial, we are unable to draw such a conclusion in the absence of a sufficient evidentiary foundation.

The government's evidence clearly established that Nelson Gonzalez Perez was a member of the Police of Puerto Rico assigned, at the time of the Cerro Maravilla shootings, to the same section of the force as was Colon Berrios, the intelligence division. As did other government witnesses, Jose Montanez Ortiz, a police officer present when the independentistas were shot, testified that Nelson Gonzalez Perez was at Cerro Maravilla on July 25, 1978. Thus, Colon Berrios perjured himself in responding that "Lieutenant Gonzalez" did not have anything to do with the Cerro Maravilla operation provided the government proved that, by "Lieutenant Gonzalez," Colon Berrios meant Nelson Gonzalez Perez and not someone else.

The record, however, does not afford such proof. The government has directed us to no document or witness which, when mentioning Nelson Gonzalez Perez's rank, calls him anything but sergeant.22 By itself, this disparity would not be fatal, if the record as a whole permitted a fair inference that Nelson Gonzalez Perez was one and the same as "Lieutenant Gonzalez." But the mystery of the true "Lieutenant Gonzalez" is deepened, and rendered in the end inexplicable, by evidence of another Lieutenant Gonzalez, different from Nelson Gonzalez Perez, who was also in the intelligence division. Antonio Mendez Rivera, a former Second Lieutenant in the intelligence division, testified as follows:

Q. Who were the other second Lieutenants, if you recall?

A. Lieutenant Sebastian Ortiz, Lieutenant Jaime Quiles, and I am not sure if by that time there was Lieutenant Francisco Gonzalez.

Given the above, the indication that Nelson Gonzalez Perez may have been a sergeant, and the government's total failure to present easily obtained clarifying evidence--for example, to show that Colon Berrios's supervisor was in fact Nelson Gonzalez Perez--the record lacks sufficient evidence, even when construed most favorably to the government, to allow a rational trier of fact to find Colon Berrios guilty beyond a reasonable doubt. See United States v. Drougas, 748 F.2d 8, 15 (1st Cir.1984); United States v. Marolda, 648 F.2d 623, 624 (9th Cir.1981).

In his January 16, 1980 deposition, Colon Berrios was asked if he had discussed the Cerro Maravilla incident with fellow officers Bruno Gonzalez or Reveron Martinez "since the time it happened." Colon Berrios replied, in part, "I have not made any comments like that." When examined on whether he had discussed the incident with "any of the other defendants" in the civil suit, Colon Berrios once again said "no." The government alleges that these are perjurious statements in that Colon Berrios "met with Juan Bruno, Luis Reveron, Jose Rios Polanco, Rafael Torres and others at Cerro Maravilla on August 2, 1978, and discussed the events of July 25, 1978."

The most damaging evidence against Colon Berrios was offered by Modesto Delgado Garcia, an employee of the corporation that owned the Channel 7 tower that Rosado and Soto Arrivi were allegedly intending to sabotage. He testified that on August 2, 1978, Colon Berrios came to Cerro Maravilla along with Rios Polanco, Reveron Martinez, Rafael Torres, and a number of district attorneys.23 Although the evidence indicates that the district attorneys discussed the shootings with each of the defendants present, and that the defendants may have talked among themselves, the government offered no evidence that Colon Berrios, although present, participated in the discussions.

The government, moreover, produced no evidence, with the exception of the August 2 episode, showing that Colon Berrios talked at any other time with any of the defendants. Although such a meeting could perhaps be inferred if Colon Berrios had engaged in perjury closely related to that of other defendants, the jury convicted Colon Berrios of only one other count of perjury, a verdict we found to be unsupported by sufficient evidence. Given this inadequate record, we reverse the conviction under Count 37.

In Bronston v. United States, 409 U.S. 352, 93 S. Ct. 595, 34 L. Ed. 2d 568 (1973), the Supreme Court held that 18 U.S.C. Sec. 1621 (1982), the general perjury statute, did not extend "to answers unresponsive on their face but untrue only by 'negative implication.'... [A]ny special problems arising from the literally true but unresponsive answer are to be remedied through the 'questioner's acuity' and not by a federal perjury prosecution." Id. at 361, 362, 93 S. Ct. at 601, 602; see also id. at 357-58, 93 S.Ct. at 599-600 (false negative implication not sufficient for perjury); United States v. Finucan, 708 F.2d 838, 847-48 (1st Cir.1983) (statement must be literally false). Different appellants make claims of "literal truth" as to Counts 3, 5, 23, 29, and 44, which we now consider seriatim.

Count 3 charges appellant Perez Casillas with perjury for testifying in a March 10 deposition, in substance, that on the morning of the shootings, Lieutenant Quiles sent Agent Montanez to Rio Piedras (a section of the city of San Juan far removed from Cerro Maravilla) to observe whether Soto Arrivi and Rosado boarded a bus leaving for Guanica.24 Perez Casillas argues on appeal that his testimony "was literally true although it conveyed false information by implication."

We think that the government adduced sufficient proof to support the jury's finding that Perez Casillas perjured himself under Count 3. Carmelo Cruz Arroyo, a member of the Intelligence Division, testified that on July 25, 1978, between 7:00 and 7:30 in the morning, Captain Jaime Quiles "instructed [him] to report to the Rio Piedras area where supposedly [Soto Arrivi and Rosado] were leaving from." He claimed that the only person who accompanied him was then-Lieutenant Antonio Mendez. Cruz further testified that after the independentistas secured a publico car (i.e., cab), he and Mendez followed them from San Juan to the base of Cerro Maravilla. According to Cruz, shortly after he and Mendez arrived at Cerro Maravilla, he talked with Perez Casillas, telling "the Colonel that the car in which the terrorists were driving had just past [sic] by."Montanez's own testimony at trial, if believed also directly contradicted the version of events claimed by Perez Casillas. Montanez stated that on July 24, 1978, pursuant to instructions given him by Perez Casillas, he had personally set up surveillance at Toro Negro (Cerro Maravilla is a mountain in the Toro Negro area of Puerto Rico). Montanez testified that nothing unusual happened that day.

According to Montanez's continued testimony, the next day, July 25, 1978, he reported to work between 7:00 and 8:00 a.m. at the intelligence division's headquarters in Hato Rey. He then received instructions from Perez Casillas to accompany him to the Isla Grande airport along with Jaime Quiles and Nelson Gonzalez.25 Montanez did so, and once at the airport, he and Perez Casillas separated and in two flights departed for Mercedita Airport in Ponce. Upon arriving at Ponce, all the officers, including Montanez and Perez Casillas, "boarded some vehicles" and went to Cerro Maravilla, where the shootings later took place.

From the testimony given by Cruz and Montanez, the jury could have reasonably concluded that Montanez did not go to Rio Piedras the day of the shootings, including during the early morning hour when Perez Casillas claimed Quiles dispatched Montanez to watch the bus terminal. More importantly, the evidence indicates that Perez Casillas knew that Montanez was not at Rio Piedras. From the time Montanez reported to work on July 25 until the shootings took place, he was with or near Perez Casillas either at or en route to Cerro Maravilla. We conclude that the jury had sufficient evidence on which to return a conviction under Count 3.

In Count 5, Perez Casillas is charged with perjury for stating during a March 10, 1980 deposition essentially that when he arrived at the area where the shootings had taken place,26 between two and five minutes after they occurred, one person was dead and a second was being placed into a car. Despite Perez Casillas' apparent claim of error, we believe that the government presented sufficient evidence of falsity to support the conviction.

Officer Montanez testified that after the first volley of shots was fired, the two independentistas were still alive:

Q. Mr. Montanez, after Gonzalez Malave [an undercover police officer who was injured during the initial volley of shots] was taken away to the hospital,27 what if anything happened to Dario and Soto?

A. They stayed at the place where they were arrested.

....

Q. [W]ere they alive at the time, or dead at the time you saw these assaults taking place?

A. They were alive.

Q. Did you stay in the area where these assaults were taking place?A. Uh, I believe not, I was driving Don Julio Ortiz Molina to the location of the police tower.

Q. Did you return to the scene where the assaults were taking place?

A. Yes, sir.

Q. What happened when you returned?

A. After I came back the atmosphere was fully charged where anything could happen, then I became aware that Colonel Perez was leaving the place and I noticed, or maybe by intuition that something maybe not good was going to happen, and I didn't want to stay there, so I left.

....

Q. What happened next ...?

A. Well, what I remember is not very, very clear, but I believe that a detonation was heard.

Q. What kind of a detonation, if you know?

A. It was a detonation like the shooting of a shotgun.

Montanez's testimony indicates that after the initial volley of shots, the two independentistas were both alive and that they remained so until after Perez Casillas left the area. This and similar testimony28 was sufficient to allow the jury to conclude that Perez Casillas knew they were alive and had perjured himself as charged in Count 5 when he spoke of one being dead.

Count 23 alleges that on March 6, 1980, Rafael Torres Marrero committed perjury before a grand jury when he asserted that by the time the wounded undercover agent was "leaving, when the car was already leaving, they were putting the other injured man inside the car." Because the evidence establishes that the only other people allegedly injured were the independentistas, and that neither victim had been shot until after the undercover agent had been taken to the hospital, see evidence discussed in conjunction with Count 5, supra, the record contains sufficient evidence to support this conviction.

Count 29 charges Bruno Gonzalez with perjury for stating before a grand jury, among other things, that by the time he arrived at the area of the shootings, Soto Arrivi's body had already been removed. Appellants' assert that "[t]his answer was literally true although it conveyed false information by implication."

The government produced sufficient evidence to support a conviction under Count 29. Montanez testified that Bruno Gonzalez was present when the exchange of shots between the independentistas and the officers occurred. Montanez claims that after the initial volley of shots, the independentistas gave up their arms. At that point, they were put in Bruno Gonzalez's control while the other officers conducted a brief examination of the area. Thus, if Montanez is believed, Bruno was "at the scene" well before Soto Arrivi was killed and "carried away."

Similarly, Cartagena Flores testified that after their capture, the independentistas were kneeling together. Four agents, Reveron, Bruno, Moreno, and Torres, surrounded them. Shortly thereafter, the independentistas were shot. The jury could certainly infer from this testimony that Bruno Gonzalez was present at the time the independentistas were shot, and thus before Soto Arrivi's body had been carried away.

Count 44 charges Mateo Espada with perjury for stating before a grand jury that after July 25, 1978, he never discussed the shootings with Perez Casillas. Given Miguel Cartagena Flores' testimony that he, Perez Casillas, and Mateo Espada met and talked about the shootings on September 12, 1983,29 we find appellants' claim that the government presented insufficient evidence of literal falsity to be without merit.

To convict a person of perjury, 18 U.S.C. Sec. 1621 (1982) requires that the allegedly perjurious statement pertain to a "material matter." See United States v. Goguen, 723 F.2d 1012, 1019 (1st Cir.1983); United States v. Indorato, 628 F.2d 711, 717 (1st Cir.), cert. denied, 449 U.S. 1016, 101 S. Ct. 578, 66 L. Ed. 2d 476 (1980). Recently, in United States v. Scivola, 766 F.2d 37 (1st Cir.1985), we elaborated on the meaning of materiality, stating that "the test for materiality is a broad one. A statement is material if it is 'capable of influencing the tribunal on the issue before it.' The statement need not be material to any particular issue in the case, but rather may be material to any proper matter of the jury's inquiry, including the issue of credibility." Id. at 44 (citations omitted).

Here, appellants claim that Counts 10, 11, 12, and 28 should have been dismissed for lack of materiality.30 The government disagrees, arguing that all four counts point to the same scheme--the attempt by appellants to conceal the presence of Jose Montanez Ortiz at the shootings. Thus, the government contends, defendants devised the story that Cruz, not Montanez, took the second flight from San Juan to Ponce on July 25, 1978 (Counts 10 and 11). Similarly, Bruno Gonzalez's statement that Colon Berrios was in charge of the agents near the tower (Count 28) allegedly hid the fact that Montanez Ortiz was at Cerro Maravilla and in command of those agents.

Although the government produced no direct evidence proving that the perjurious statements about Cruz's activities were designed to hide Montanez's role,31 the record contained enough circumstantial evidence to allow such a conclusion to be drawn. The government established that Montanez was in command of the agents operating around the television tower and that he flew with Nelson Gonzalez Perez from the Isla Grande Airport in San Juan to the Mercedita Airport in Ponce. Moreover, Cruz testified that he did not fly to Cerro Maravilla on July 25. Indeed, Cruz asserted that for his benefit, months after the shootings, Perez Casillas and other officers retraced the route they had taken to Cerro Maravilla on July 25. They did so to allow Cruz to testify falsely before the senate that he had accompanied Perez Casillas and others to Cerro Maravilla on the day of the shootings.

We believe that this evidence is sufficient to allow a reasonable jury to infer that defendants attempted to conceal Montanez' involvement by substituting Cruz in his place.

We are also satisfied that the concealment of Montanez's role falls within the materiality requirement of section 1621. The perjury charges before us arose out of three cases: two involved federal grand juries investigating the killings and the third pertained to a civil action alleging violations of Rosado's and Soto Arrivi's federally protected rights. In each case, knowledge of the precise people involved in the shootings and the role each person played was "capable of influencing the tribunal on the issue before it." Scivola, 766 F.2d at 44. For example, in assessing liability, the jury trying the civil case would certainly be interested in knowing whether Montanez was at the shootings. Accordingly, we affirm appellants' convictions on Counts 10-12 and 28.

D. Failure of Proof.

Appellants next attack the convictions returned under Counts 21, 22, 30 and 40, claiming that the government produced insufficient evidence. We find these arguments to be without merit.

Count 21 charges Torres Marrero with perjury for the following testimony before a grand jury:

Q. With respect to the man who was taken to the hospital who had been shooting at the police, did you see anyone strike or hit that person at any time?

A. No.

Q. Referring to the one who was just wounded?

A. No. Nobody was hit.

Torres Marrero asserts that because no evidence established that he actually saw an independentista being struck, his conviction under Count 21 cannot stand. The government does not respond to Torres Marrero's claim, but argues that the jury properly concluded that Torres' unqualified statement that "[n]obody was hit" was knowingly false.

We believe the government presented sufficient evidence to support the conviction. A number of witnesses testified that Torres Marrero was one of the four agents who surrounded the independentistas immediately after their capture. See, e.g., testimony of Miguel Cartagena Flores ("Q. And who were these agents that were around these two individuals [Soto and Rosado]? A. Agent Reveron, Bruno, Moreno, and Torres ") (emphasis added); testimony of Antonio Mendez Rivera ("Q. Who was present, to your knowledge, in the immediate area where the shots that you have described, were fired ...? A. Rafael Moreno, Rafael Torres, and Luis Reveron") (emphasis added).

The government also demonstrated that immediately prior to the fatal shootings, the two independentistas were repeatedly assaulted. Montanez testified "[a]fter the shooting event ... the atmosphere was very hot because believing that the [undercover] agent, the fellow worker that was hurt was such a damaging hurt that it could cause his death and there was a collective hysteria going on by which the arrested persons were beaten." Similarly, Julio Ortiz Molina (the publico, i.e., cab, driver) stated that "[t]hey [the independentistas] were being kicked and hit with the butt of rifles."

This evidence was sufficient to allow the jury to conclude either that (1) Torres Marrero must have seen the beatings going on or (2) even if he did not actually see the beatings, his presence in the immediate area made it inevitable that he knew of them and thus lied by stating that "nobody was hit."

In Count 22, Torres Marrero was charged with perjury for the following testimony before the grand jury:

Q. After the initial shooting did you hear any shots fired later?A. No, no fires--nothing was shot.

On appeal, he claims that the government presented no evidence that he actually heard the shots.

As shown in the discussion of Count 21, Torres Marrero was one of the four agents surrounding the independentistas after their capture. The testimony of Miguel Cartagena Flores, among other people, demonstrated that Torres Marrero was still nearby when Rosado and Soto Arrivi were killed. Indeed, Antonio Mendez Rivera, a police officer present at Cerro Maravilla, testified that "Mr. Moreno grabs the weapon that Rafael Torres [Marrero] was using, and fired at the one who turned out to be Soto Arrivi." See also the testimony of Montanez Ortiz (Moreno Morales told Montanez that "someone had not wanted or had not dared to shoot at Soto Arrivi and that [Moreno Morales] took away his weapon and shot him").

In light of this evidence, the jury was certainly entitled to conclude that Torres Marrero was lying when he stated that "no, no fires--nothing was shot." Because people far away from the shootings heard the gun fire, it is reasonable to infer that Torres Marrero, who was within six feet of the independentistas, must have heard the shots.

Count 30 alleges that Bruno Gonzalez perjured himself before a grand jury by responding "no" when asked "After the first shooting was over, did you hear throughout the rest of the next hour any other shot fired?". Because the evidence demonstrated that Bruno Gonzalez was one of the four agents surrounding the independentistas before their death, see discussion of Count 21, the jury could have reasonably inferred that he must have heard the shots.

Count 40 claims that Rios Polanco committed perjury before a grand jury when he stated that he did not hear any shots after the initial volley. Although Rios Polanco does not appear to have been as close to the independentistas as were Bruno Gonzalez and Torres Marrero, a number of witnesses testified that he was in the general vicinity of the Channel 7 facility. Given this evidence, it was the jury's responsibility to evaluate whether Rios Polanco, like others not in the immediate area, heard the shootings.

E. "Failure to Pin Down "

Nelson Gonzalez Perez claims that his convictions under Counts 32, 33, 34, and 35 must be reversed. While before the grand jury, Gonzalez Perez stated that he "did not remember" seeing Jose Montanez Ortiz (Count 32), Luis Reveron Martinez (Count 33), Rafael Moreno Morales (Count 34), or Rosado and Soto Arrivi (Count 35) between the hours of 7:30 a.m. and 3:00 p.m. on July 25, 1978. The government, however, produced evidence showing that he was present at Cerro Maravilla at those times and that those individuals were likewise there, permitting the jury to infer, given the enormity of the events and his association with Montanez, Reveron and Moreno, that he was lying when he said he did not remember.

Gonzalez now argues that,

It is the responsibility of the lawyer to probe.... If a witness evades, it is the lawyer's responsibility to recognize the evation [sic] and to bring the witness back to the mark, to flush out the whole truth with the tools of adversary examination.... [Here,] the prosecutor took his [Gonzalez Perez's] answers and failed to follow them up.

Brief at 53.

We assume that Gonzalez Perez is arguing that his answers were too vague to support a perjury conviction. We disagree. Gonzalez clearly stated that he "did not remember" and the jury was entitled to conclude that, under these circumstances, he did remember, and was perjuring himself when he declared otherwise.32 United States v. Ponticelli, 622 F.2d 985, 989 (9th Cir.), cert. denied, 449 U.S. 1016, 101 S. Ct. 578, 66 L. Ed. 2d 476 (1980).

Appellants contend that Counts 31, 32, 33, 34, 35, 42, and 44 should have been dismissed because they pertain to declarations made before a grand jury whose investigation was already complete. Appellants present no evidence indicating that the grand jury's investigation had concluded by the time the perjurious statements charged in Counts 31-35, 41-42 were made. Accordingly, we affirm the convictions returned under the counts.

Count 1 charged all defendants with conspiracy to commit, among other offenses, perjury before the grand jury and in depositions. The count alleged that defendants engaged in the scheme in order to "prevent the citizens of Puerto Rico and law enforcement authorities of Puerto Rico and the United States from learning that Arnaldo Dario Rosado and Carlos Soto Arrivi had been unlawfully brutalized and killed by officers of the Police of Puerto Rico ... at Cerro Maravilla." The jury returned convictions against each defendant on Count 1, a decision all appellants except for Moreno Morales and Perez Casillas challenge for insufficiency of evidence. We reverse the conviction against Colon Berrios, while affirm as to the other appellants.

The government presented direct evidence of Quiles Hernandez's and Mateo Espada's involvement in the conspiracy. Miguel Cartagena Flores testified that "days after the shooting," in the general police headquarters at Hato Rey, he met with Moreno Morales, Perez Casillas, and Quiles Hernandez "to get together as to what we were going to say" when questioned pursuant to the investigation being made by the Department of Justice. According to Cartagena Flores, Perez Casillas "gave him instructions" on the substance of his testimony, including telling him falsely to state, if asked, "that Montanez had not travelled by airplane and that it was Carmelo Cruz who had traveled by airplane." See discussion of materiality, supra, section IV C. Cartagena Flores also asserted that on September 2, 1983, he met with Mateo Espada, Perez Casillas, and Quiles Hernandez to discuss how to testify before the Senate. Carmelo Cruz Arroyo corroborated Cartagena's testimony (stating that Mateo Espada and Quiles Hernandez were at the September 2 meeting and claiming that Perez Casillas provided him with false testimony to give before the Senate).

The government also proved through circumstantial evidence that Torres Marrero, Bruno Gonzalez, Gonzalez Perez, and Rios Polanco participated in the conspiracy. The evidence demonstrated that they were all at Cerro Maravilla when the shootings occurred. They are each members of the same division within the police force as those defendants who either concede the existence of the conspiracy or who were convicted on direct evidence of meeting in furtherance of the conspiracy. Most importantly, however, each was found to have committed multiple perjury before the grand jury with the effect of "preventing the citizens of Puerto Rico and law enforcement authorities" from learning the details surrounding the deaths of Rosado and Soto Arrivi.

The similarity of their perjury gives rise to the inference that they were all in league. It strains credulity, for example, to conclude that Torres Marrero (Count 22), Bruno Gonzalez (Count 30), and Rios Polanco (Count 40) independently arrived at the false conclusion that only one round of shots was fired. This example, which is but one of a number, permitted the jury to conclude that the police officers acted in concert to devise a story intended to conceal the truth about the shootings.

The government, of course, was entitled to prove conspiracy through circumstantial evidence. As the Supreme Court stated in Glasser v. United States, 315 U.S. 60, 80, 62 S. Ct. 457, 469, 86 L. Ed. 680 (1942), "[p]articipation in a criminal conspiracy need not be proved by direct evidence; a common purpose and plan may be inferred from a 'development and a collocation of circumstances.' " (Citations omitted). See United States v. Cincotta, 689 F.2d 238, 241 (1st Cir.), cert. denied, 459 U.S. 991, 103 S. Ct. 347, 74 L. Ed. 2d 387 (1982); United States v. Patterson, 644 F.2d 890, 893 (1st Cir.1981); United States v. Bithoney, 631 F.2d 1, 5 (1st Cir.1980), cert. denied, 449 U.S. 1083, 101 S. Ct. 869, 66 L. Ed. 2d 808 (1981). Thus, we affirm the convictions under Count 1 of Bruno Gonzalez, Gonzalez Perez, Rios Polanco, and Torres Marrero.

We reverse, however, the conspiracy conviction of Colon Berrios. Although a number of witnesses testified that Colon Berrios was present at Cerro Maravilla, the government did not present sufficient evidence, as we have already determined supra, for us to affirm. Since perjury was not shown, there is insufficient other evidence to prove that he took part in a conspiracy to cover up the true facts of the events at Cerro Maravilla.

Perez Casillas asks us to reverse his convictions because he received ineffective assistance from his trial attorney. This assertion has no merit.

In Strickland v. Washington, 466 U.S. 668, 104 S. Ct. 2052, 80 L. Ed. 2d 674 (1984), the Supreme Court adopted a two part standard for assessing such a claim. A defendant must show (1) that his attorney made errors so serious that the sixth amendment right to counsel was violated; and (2) that but for the errors, there is a reasonable probability that the result of the proceeding would have been different. Id. at 687, 694, 104 S. Ct. at 2064, 2068. The Court advised that "[j]udicial scrutiny of counsel's performance must be highly deferential.... [A] Court must indulge a strong presumption that counsel's conduct falls within the wide range of reasonable professional assistance." Id. at 689, 104 S.Ct. at 2065; see also Therrien v. Vose, 782 F.2d 1, 3 (1st Cir.), cert. denied, --- U.S. ----, 106 S. Ct. 2285, 90 L. Ed. 2d 727 (1986) (without specific showing of errors and prejudice, ineffective assistance of counsel claim must fail); United States v. Fuller, 768 F.2d 343, 346 (1st Cir.1985) (same).

Here, Perez Casillas argues that his attorney committed grievous error by failing to cross examine three prosecution witnesses, Marte Ruiz, Cartagena Flores, and Cruz Arroyo. These witnesses provided highly incriminating evidence against Perez Casillas. For example, Officer Cartegena testified that Perez Casillas stated that "he had received confidential information that a group of armed terrorists, armed and dangerous, were thinking about attacking some towers at the Toro Negro area. That among them there was a hostage, that everything possible must be done to save the hostage's life and that the terrorists were not, or should not come down alive." Perez Casillas contends that leaving such testimony unexamined deprived him of effective assistance of counsel "as thoroughly as if trial counsel had been absent."

We cannot say, however, that the failure to cross examine was an unreasonable trial tactic. Experienced trial attorneys may choose not to cross examine witnesses where the probable result is a mere repetition and strengthening of the direct testimony. Counsel here could have reasonably concluded that cross examination would at best be futile and at worst self-destructive. None of the other defense attorneys chose to cross examine these witnesses. Appellant provides only generalized speculation as to what cross examination might have uncovered; he points to nothing in the record which indicates that an examination of these particular witnesses was reasonably likely to produce a different verdict.

To show prejudicial error, Perez Casillas had to demonstrate that cross examination stood a likely chance to undermine the probative value of the witnesses' testimony (and thereby the verdict itself). Having failed to make that showing, he failed under Strickland v. Washington to assert a valid claim of ineffective assistance of counsel.33

All appellants contend that the terms of imprisonment imposed upon them are disproportionately severe.34 A court of appeals' power to intervene in respect to length of sentences, however, is extremely limited. As we stated in United States v. Francesco, 725 F.2d 817, 823 (1st Cir.1984), "[t]he reviewing court may overturn a sentence only when it exceeds the statutory limit, or is so disproportionate to the offense for which it was imposed that it constitutes cruel and unusual punishment." (Citations omitted). The sentences did not exceed the limits permitted under the relevant federal perjury statutes, 18 U.S.C. Secs. 371, 1621-1623, and they cannot be said to constitute cruel and unusual punishment.

We find no merit in appellants' remaining arguments. Accordingly, with the exception of all of Colon Berrios's convictions (Counts 1, 37, 38), the judgments below are affirmed.

Affirmed in part; reversed in part.

TORRUELLA, Circuit Judge (Dissenting in part and concurring in part).

There has been sufficient impassioned comment on the issues revolving around this case without my adding to the fire by engaging in excessive hyperbole. I am, however, reminded of an old Spanish saying to the effect that "one cannot block out the sky with one's hand" (no se puede tapar el cielo con la mano ). In light of the pretrial publicity established by this record I cannot help but conclude that affirming that the appellants received an impartial trial, is like trying to "block out the sky with one's hand."It is hardly an overstatement that this case represents the paradigm of pretrial publicity. I have searched in vain for a factual situation that even approaches that of the present case and have found none. If the standard set by this case is allowed to prevail, no criminal defendant can hope to overcome it. Furthermore, the burdens placed on a criminal defendant by the majority clearly run against specific constitutional guaranties. U.S. Const. Art. III, Sec. 2, cl. 3; Art. VI.

It goes without saying that in considering the questions before us, the guilt or innocence of the accused, or the heinousness of the offenses charged, should be irrelevant to the issue of whether they are entitled to, and received, a fair trial. In fact, it is when there are charges of opprobious crimes that courts are required to exercise heightened sensitivity to an accused's need for protection against undue pretrial publicity. See Coleman v. Kemp, 778 F.2d 1487 (11th Cir.1986), cert. denied, --- U.S. ----, 106 S. Ct. 2289, 90 L. Ed. 2d 730 (1986); Delaney v. United States, 199 F.2d 107 (1st Cir.1952).

It is axiomatic that the right to trial by jury is a fundamental right. Duncan v. Louisiana, 391 U.S. 145, 88 S. Ct. 1444, 20 L. Ed. 2d 491 (1968). This right would be meaningless if it did not encompass the right to be tried by fair and impartial jurors. Irvin v. Dowd, 366 U.S. 717, 722, 81 S. Ct. 1639, 1642, 6 L. Ed. 2d 751 (1961).

I accept that under today's modern constitutional thinking, however, an impartial juror is not necessarily one who is totally ignorant of current affairs. It is now settled law that prior knowledge about the case by prospective jurors is not anathema to their impartiality. Patton v. Yount, 467 U.S. 1025, 104 S. Ct. 2885, 81 L. Ed. 2d 847 (1984). "[T]he relevant question is not whether the community remembered the case, but whether the jurors ... had such fixed opinions that they could not judge impartially the guilt of defendant[s]." Patton, supra, at 1035, 104 S.Ct. at 2891; Irvin, supra, 366 U.S. at 723, 81 S. Ct. at 1642.

In determining the extent to which pretrial publicity may have entrenched prejudice in the minds of prospective jurors, it is of course relevant to consider the record of that publicity as well as the voir dire testimony of the jurors. As a rule, a juror's profession of impartiality is entitled to deference, as is a trial court's finding to such effect. Patton, supra, 467 U.S. at 1038, 104 S. Ct. at 2892; Coleman, supra, at 1490. In law, such a ruling can only be reversed on appeal upon a showing of manifest error. Id.

There is abundant case law, however, to the effect that under appropriate circumstances a trial court is duty bound to reject jurors' claims of objectivity where the prospective jurors have been subjected to a "barrage of inflammatory publicity immediately prior to trial." Murphy v. Florida, 421 U.S. 794, 798, 95 S. Ct. 2031, 2035, 44 L. Ed. 2d 589 (1975). In such a situation "little stock need be placed on jurors' claim to [impartiality]." Holbrook v. Flynn, 475 U.S. 560, 106 S. Ct. 1340, 1347, 89 L. Ed. 2d 525 (1986); Sheppard v. Maxwell, 384 U.S. 333, 351, 86 S. Ct. 1507, 1516, 16 L. Ed. 2d 600 (1966); Irvin, supra, 366 U.S. at 728, 81 S. Ct. at 1645; Marshall v. United States, 360 U.S. 310, 312, 79 S. Ct. 1171, 1172, 3 L. Ed. 2d 1250 (1959); Coleman, supra, at 1490 (prejudice presumed where community saturated by pretrial publicity). Where the record establishes the probability of prejudice by reason of the pretrial publicity, failure of the trial court to grant appropriate relief constitutes manifest error. Patton, supra, 467 U.S. at 1031, 104 S. Ct. at 2888; Estes v. Texas, 381 U.S. 532, 543, 85 S. Ct. 1628, 1633, 14 L. Ed. 2d 543 (1965); Sheppard, supra, 384 U.S. at 352, 86 S. Ct. at 1516; In re Murchinson, 349 U.S. 133, 136, 75 S. Ct. 623, 625, 99 L. Ed. 942 (1955); Coleman, supra. See also American Bar Association, Standards for Criminal Justice, 2d Ed., "Fair Trial and Free Press," Vol. II (Supplement) Standard 8-3.3, p. 8.40, Little, Brown & Co., Boston.

In determining the probability that pretrial publicity may have had a prejudicial effect upon the prospective jurors, the following are the relevant factors to be considered, based upon the record of the publicity and testimony of the jurors during voir dire:

(1) The volume and reach of the publicity. Estes, supra, 381 U.S. at 535, 85 S. Ct. at 1629; Mayola v. Alabama, 623 F.2d 992, 998 (5th Cir.1980), cert. denied, 451 U.S. 913, 101 S. Ct. 1986, 68 L. Ed. 2d 303 (1981); United States v. Mazzei, 400 F. Supp. 17 (W.D.Pa.1975). See also Ruiz v. State, 265 Ark. 875, 582 S.W.2d 915, 920 (1979); Commonwealth v. Cohen, 489 Pa. 167, 413 A.2d 1066, cert. denied, 449 U.S. 840, 101 S. Ct. 118, 66 L. Ed. 2d 47 (1980).

(2) The nature of the pretrial publicity. Irvin, supra, 366 U.S. at 725-27, 81 S. Ct. at 1644-45 ("inflammatory," "hostile," "emotionally charged"); United States v. Medina, 761 F.2d 12, 19 (1st Cir.1985) ("pervasive, intense and inflammatory").

(3) The time elapsed between the adverse publicity and the trial, and the effect that such a hiatus may have had on the minds of the jurors. Patton, supra (four years between adverse publicity and second trial; no prejudice found); Murphy, supra, (seven months elapsed between publicity and trial, but jurors remembered only vaguely or not at all the contents of the publicity; no prejudice); Beck v. Washington, 369 U.S. 541, 556, 82 S. Ct. 955, 963, 8 L. Ed. 2d 98 (1962) (nine and one-half months sufficient for impact of publicity to subside).

Because this issue and the relevant cases are so fact bound, in considering this appeal, the record cannot be glossed over but must be considered in detail. In so doing we must keep in mind that both the pretrial publicity and the criminal charges centered on the same issue. The crux of both lay in competing versions as to how Soto-Arrivi and Rosado, the two alleged terrorists, were killed. This incident is popularly referred, even by the trial court in voir dire, as the "Cerro Maravilla incident." It is principally around this controversy that evolved, commencing in July 1978, the publicity which gave course to this appeal. As will be presently discussed, it was defendant-appellants' alleged lack of credibility, the very essence of the charges made against them in the present cases, that became the central focus of this pretrial publicity. Because of this, the jury that tried this case, not only was exposed beyond permissible bounds to prejudicial pretrial publicity, but was privy to knowledge not in the trial record. Furthermore, it is clear that the jury used this extra-judicial knowledge in the course of its deliberations. A prime example of the latter is the verdict against appellant Colon-Berrios in Counts 1, 37 and 38, which we have been forced to overturn. It is obvious that the jury in that case used knowledged received ex record in reaching its verdict. That is a situation which is just as likely to have happened as regards the other defendant-appellants and the remaining charges.

Although the publicity concerning the Cerro Maravilla incident commenced immediately after that event took place on July 25, 1978, and continued throughout the trial, it is the publicity that took place after 1980 that more immediately concerns us. Cf. In re San Juan Star, 662 F.2d 108 (1st Cir.1981); Colon-Berrios v. Hernandez Agosto, 716 F.2d 85 (1st Cir.1983). In 1980 elections were held in Puerto Rico. The Cerro Maravilla incident was a principal issue in the political campaign that preceded it. One of the major political parties, the Popular Democratic Party (PDP), promised to reinvestigate the entire matter if it won the elections. The PDP lost the gubernatorial race by the closest margin in Puerto Rican political history, but won a majority of the seats in both houses of the local legislature.

Shortly after the new Senate was sworn in, it proceeded to conduct its own investigation of the Cerro Maravilla incident.35 This investigation culminated with a six month "live" televised hearing which started in June and lasted until December 1983. These proceedings, which have also become known locally as las vistas ("the hearings"), will presently be discussed more in detail. For the time being it is sufficient to say that the vistas reached their dramatic climax when Miguel A. Marte (Marte), a television technician who witnessed the shoot-out from a nearby location, and three members of the police stake out team, Carmelo Cruz, Jose Montanez and Cartagena Flores, requested and received immunity from the Senate and the Federal government, and thereafter testified "live" on television that Soto-Arrivi and Dario Rosado had been shot by the police during a second volley of gunfire, after they had surrendered and were disarmed. This version was contrary to that of appellants, who steadfastly contended that Soto-Arrivi and Rosado were killed during the original exchange of shots, as part of the officers' legitimate self-defense.

As a direct result of the evidence adduced in the vistas, the federal authorities also reopened a prior criminal investigation and reconvened a new grand jury, which thereafter issued the indictment in this case, charging conspiracy and various individual counts of perjury and obstruction of justice.

The political campaign for the 1984 election again made Cerro Maravilla a central issue, this time however, with the Senate hearings as the immediate backdrop.

In the meantime this case was progressing below. In the face of urgings by the Government that the case proceed to trial, opposed by defense motions seeking an indefinite continuance because of the pretrial publicity, the trial judge made a crucial ruling on August 22, 1984. See United States v. Perez-Casillas, 593 F. Supp. 794 (D.C.P.R.1984).

This order, which decided the defenses' motion for continuance, is particularly useful in that it conveniently summarizes the most outstanding events related to the pretrial publicity issue prior to that date. It also serves as the focal point to a determination of that question because it narrows the scope of inquiry to deciding whether the circumstances prevalent on August 22, 1984 had sufficiently ameliorated by February 5, 1985 to permit the empaneling of an impartial jury.

The district court took judicial notice of "the contents of all newspapers of general circulation in Puerto Rico during the period of 1979 to the present directly related to the Cerro Maravilla incident." It also took notice "that the Senate hearings on the Cerro Maravilla incident ... were televised ... and transmitted by radio." Perez-Casillas, supra, 593 F.Supp. at 796-797. The court indicated, citing from our own opinions, that:

It would be shunning reality not to recognize that the Cerro Maravilla events were followed by a deluge of publicity which has yet to abate. This much was recognized by our circuit as early as 1981 when it said: "Coming as it did in the midst of a heated debate over the Island's political future--and on the eve of a closely contested electoral campaign--the Cerro Maravilla incidents attracted widespread popular and political attention." In Re San Juan Star, 662 F.2d 108, 111 (1st Cir.1981). Again in 1983: "The incident at Cerro Maravilla had and has significant implications and has been the subject of intense media coverage and popular attention." Colon-Berrios v. Hernandez-Agosto, 716 F.2d 85, 87 (1st Cir.1983).

Id. It then found that "[s]ince 1978 the press, radio and television coverage, heightened by the Senate public hearings, has had the cumulative effect of making the Cerro Maravilla the media event of the years 1983-1984. The hearings, undoubtedly, aroused the excitement and the curiosity of a community that was already saturated by the earlier publicity stemming from this incident." Id. at 798 (emphasis in the original). The court proceeded to enumerate and analyze a plethora of newspaper articles and reports, which I cannot herein reproduce but which are well worth reading if one wishes to acquire a flavor for the nature and extent of the publicity preceding the trial of defendant-appellants. Id. 798-805. Suffice it to merely include some of the district court's findings:

The reporting of the news, frequently off-course in the objective presentation of the facts, contained conclusory, slanted commentaries and subjective opinions of the reporters. The reporting of the testimony of the policemen at the hearings was constantly characterized as contradictory, altered, or belied by other testimony.

Id. at 803.

The community has also been constantly exposed to remarks by respected public figures of the entire political spectrum and by members of the Senate Judiciary Committee proclaiming belief in defendants' guilt or expressing doubts on the veracity of their testimonies which appeared in newspapers, television and radio coverage ...

Id.

This last finding was reinforced by the quoted statements of various prominent public figures: Miguel Gimenez-Munoz, former Secretary of Justice of Puerto Rico (stated that "the murder of the independentistas is 'unpardonable' "), Rafael Hernandez-Colon, Governor of Puerto Rico during the years 1972-76, who was re-elected to that post as a result of the 1984 election (characterized the incident as "the summary execution of dissidents and the official lie to hide this from the people"), Hernan Padilla, then mayor of San Juan, (indicated that the murders at Cerro Maravilla have been proven, "Nobody who is certain of what he says needs to invoke the Fifth Amendment"), Calixto Calero-Juarbe, then member of the Senate Judiciary Committee (opined that the "killings" and the cover-ups were "monstrous acts"), Jose Granados-Navedo, a New Progressive Party legislator (referred to acts committed by a "handful of policemen," as "crimes that defy imagination"), Jose R. Lebron Velazquez, head of the Puerto Rican Evangelical Council (urged a purge of the Cerro Maravilla offenders), Francisco Aponte-Perez, President of the Senate Judiciary Committee (said defendant-appellant Perez-Casillas would tell the truth when faced with incarceration), Miguel Hernandez-Agosto, President of the Senate of Puerto Rico (commenting on the Fifth Amendment privilege invoked by defendant-appellants Torres-Marrero and Rios-Polanco stated: "By resorting to such refuge, what they are implying is that if they repeat the testimony that they previously offered to another Grand Jury and the one offered in the civil action for their participation in the Cerro Maravilla incident, they would be exposing themselves to perjury charges"). Id. at 803-805.

The district court's findings regarding the vistas and their aftermath are of importance:

The subject of most of this news reporting, the Senate televised hearings themselves, enthralled the public and made the Cerro Maravilla incident a conversation piece and a household word.

These hearings were not the usual type of legislative investigation. As they developed, defendants, in effect, stood trial before an entire community and guilt was adjudicated without the constitutional protections afforded a criminal defendant in a court of law and without the safeguards of evidentiary principles.... [W]ith all due respect to the Senate's exercise of its legitimate investigative function, the Maravilla hearings were conducted in an adversative atmosphere which, nonetheless, lacked the balance and the impartiality of a judicial process where an accused enjoys a presumption of innocence, is entitled to confront and cross-examine witnesses, compel the attendance of witnesses to testify in his favor, an active and effective assistance of counsel and the right to remain silent. These televised hearings ended in a dramatic note after the Senate announced that several policemen present at the Cerro site had been granted immunity and would offer revealing testimony. Before an expectant community, these witnesses revealed, in a testimony filled with emotional overtones, that Soto and Rosado were summarily executed. Their testimonies triggered a massive public reaction reflected in the media. From that moment onward, charges of murder against defendants were constantly voiced and exposed by the media. The Senate investigation moved on to what the press termed the cover-up stage, taking for granted the guilt of defendants as expressed by the last witnesses at the hearing.

To this date, the Cerro Maravilla issue remains in the limelight because of constant media exposure.⁸ This is due in part to the ongoing Senate investigation and the fact that Cerro Maravilla is a major issue in the upcoming elections. Recent political propaganda as pointed out by one of the codefendants, includes a television commercial which shows the corpses of Soto and Rosado with the word "cover-up" superimposed in red ink.

It is illusory to think that this type of publicity is going to wear off as the election day approaches, which is but seventy-six (76) days away....

Id. at 805-806 (emphasis supplied).

Finally, concluding that the totality of the circumstances impeded the holding of a fair trial, the court continued the trial until at least early 1985.

The above, which constitutes the undisputed record of this case, is in marked contrast with the bland treatment given by the majority to these facts.

In analyzing these subsequent events to determine probability of prejudice it is required, as previously indicated, that consideration be given to: (1) the volume and reach of the publicity, (2) the nature of the publicity, and (3) the time elapsed between the publicity and the trial.

I commence with the undeniable proposition that after August 22, 1984, the electoral campaign in Puerto Rico continued in crescendo until November 1984. As the district court predicted in its August 22 order, the political controversy regarding Cerro Maravilla continued through at least the electoral period. As will be seen it also continued thereafter. Furthermore, this publicity developed not only in the press, but also on television and radio, as it had prior to the court's order. A nationwide news program on "60 Minutes" was not only broadcast in Puerto Rico but was also widely commented upon by the local newspapers.

Notwithstanding the suspenseful atmosphere prevalent in Puerto Rico regarding the Cerro Maravilla incident caused by the above, something well known to the Government and on which I shall presently comment, the Government continued to pressure the district court for a trial setting. Finally, on November 20, 1984 the court ordered that the trial commence on February 5, 1985. A timely motion opposing this setting was ruled moot by the district court sans a hearing or specific findings. Another motion for continuance, based on similar grounds, was filed on January 15, 1985. It was again denied on January 18 in a mercurial footnote order, also without a hearing or detailed findings.

The January 15th motion is important because, absent the holding of any hearing or the making of any specific findings by the district court, the 17 page appendix attached to that motion containing references to newspaper reports between October 9, 1984 and January 14, 1985 is the only evidence in the record of what took place with regards to pretrial publicity between the August 22 order and the commencement of trial on February 5, 1985. I cannot, in the light of this appendix, which I shall presently discuss, as well as other pertinent events to be recounted, dismiss the district court's failure to hold a hearing or make specific findings, as lightly as is done by my colleagues in the majority who merely say that such a procedure would have been "preferable," speculating as they do on the assumption "that the judge, a resident of Puerto Rico and the same judge who [decided the August 22 ruling] ... [must have] felt that the climate had sufficiently improved to permit a fair trial." See maj. op., ante at 731. This is putting the shoe in the wrong foot. It is precisely because of the strong findings in the August 22 order that a hearing and specific findings were mandated. If any assumption were to be made by this Court, particularly in view of the supporting evidence presented in the motion as well as the other matters referred to, it should be that the situation found by the August 22 ruling had not changed. Other factors, including later rulings by the district court, bear this out.

As indicated, the appendix to the January 15 motion points to 189 newspaper articles dealing with the Cerro Maravilla incident between October 9, 1984 and January 14, 1985, as well as to related television and radio items. Those references included such headlines as: "Concealment is disclosed," (El Nuevo Dia, October 15, 1984, p. 3); "Charges of murder filed against eight policemen," (El Nuevo Dia, October 20, 1984, p. 1), with photographs of five of the appellant-defendants; "Today; Hearing regarding Maravilla entrapment," (El Reportero, October 22, 1984, p. 3); "At the Senate they cried, however at the Court, they were smiling," (El Vocero, headlines, October 22, 1984); "Entrapment found," (El Mundo, October 24, 1984, front page); "Maravilla's Crime was planned," (El Reportero, October 24, 1984, front page); "More contradictions in Maravilla Case," (El Reportero, October 30, 1984, p. 2); "Probable cause for concealment charges," (El Nuevo Dia, November 3, 1984, p. 2); "60 Minutes cites Maravilla coverup," (El Mundo, November 5, 1984, p. 10-B); "The Corruption Pyramid," (United Citizens Against Corruption, El Nuevo Dia, November 5, 1984, p. 95); "Infamy," (El Nuevo Dia, November 7, 1984, p. 39); "Ortiz Molina asks R.H.C. to proceed with Maravilla investigation," (El Mundo, November 15, 1984, p. 9-A); "Today Maravilla Issue is renewed," (El Nuevo Dia, December 12, 1984, p. 18). The motion also included a clipping from the New York Times (January 3, 1985, p. A14), regarding Governor Hernandez-Colon's inaugural address of January 2, 1985 in which he was quoted as condemning the "entrapment and summary execution of two independence supporters by police at Cerro Maravilla."

The government's opposition unabashedly admitted that "[i]t is undisputed that the events which occurred at Cerro Maravilla on July 25, 1978, have engendered substantial publicity." It insisted, however, that the cure for the publicity was to be found in the voir dire.

On January 28, 1985 defendant-appellants sought reconsideration of the district court's ruling denying the continuance, calling attention to the fact that "at both the opening session of the Puerto Rico Legislature and [at the swearing in ceremonies of] the new Commonwealth Secretary of Justice ... great official and news media prominence was given to Mr. Jose Ortiz-Molina and to his version of the events of 'Cerro Maravilla'."36 An addendum to that motion included an article in El Reportero, a San Juan newspaper of general circulation, with a detailed discussion of the voir dire proposed at the pretrial conference held on January 29, 1985. On its front-page was depicted a drawing of the faces of Dario Rosado and Soto-Arrivi, with a superimposed sketch of two kneeling figures being shot by persons with weapons in their hands.

On February 1, 1985, the district court, again without a hearing or specific findings denied the request for continuance. The court instead issued an order seeking "the voluntary self-restraint in the public expression of those who would write off the judicial system in the adjudication of the guilt or innocence of the defendants," urging "the high public officials of the executive and legislative branches of this Commonwealth to be mindful of the fact that the right of each American citizen to a fair trial is an integral part of the Constitution that they serve to defend and uphold." This statement is, in itself, the clearest admission, absent the specific findings that should have been made, that "something [was] rotten in the state of Denmark."37

The failure to make these findings, considering the record of this case, should be enough to cause reversal of appellants' convictions. The speculation that is otherwise required of this court regarding the pretrial publicity issue, in view of the specific findings contained in the August 22 order, is totally unacceptable.

Because of the lack of specific post-August 22 findings by the trial court regarding appellants' allegations of prejudicial pretrial publicity, the voir dire is of particular relevance as a measure of the extent to which such publicity was still fresh in the prospective jurors' minds. Since the jury in the District of Puerto Rico is "selected at random from a fair cross section" of the residents of that district,38 the testimony received through voir dire is presumptively reflective of the opinion and views of the Puerto Rican community at large.

Two comments about the majority's statistical analysis of this issue. The first is that I am not as enthusiastic about the quality of the district court's voir dire as are my colleagues. Although I recognize that the trial judge did take some unusual steps such as engaging in individualized voir dire in an effort to weed out biased jurors, and that she was under great constraints because of the nature and length of the case in question, I find that the voir dire was in general perfunctory, leading and not as incisive as should have been expected considering the massive and pervasive nature of the pretrial publicity. Although the length of time that the voir dire took is impressive, the content of the record is not. Furthermore, because the proposed voir dire was commented upon in detail in the newspaper El Reportero on January 29, 1985, ante at 758, less than a week before the trial commenced, a problem not inquired into by the trial judge, it is more than likely that a substantial number of the jury pool had prior knowledge not only of the Cerro Maravilla incident but also of the voir dire.

Second, I would like to comment on the majority's statistical conclusions as to the jury composition. I am well aware that different people reading the same record can reach different conclusions, and that this phenomena may explain why my statistical exercise differs in its conclusions from the majority's. I would point out, however, that even the government concedes to higher figures regarding pretrial publicity exposure by the jury that heard the case, than does the majority. In its brief the government stated the following:

Of the 12 jurors who actually decided the case, 2 indicated that they had been exposed to what can be considered a great deal of media publicity, 4 a moderate amount, and 6 a small amount. As to the Senate hearings, 6 said that they had great exposure, 2 had moderate exposure, and 4 minimal exposure.

Brief for the United States, p. 16 (footnotes omitted). Of course, the latter statistic regarding exposure to the Senate hearings is of crucial importance, as will be later discussed, because any juror who saw those hearings had in effect already "tried" the criminal case once, in addition to being privy to information not admissible at the real trial. See Rideau v. Louisiana, 373 U.S. 723, 83 S. Ct. 1417, 10 L. Ed. 2d 663 (1963).

Be that as it may, the following is my statistical analysis of the venires and of the petit jury that tried the case. I emphasize again that the attitudes of the entire venires are important, in addition to that of the petit jury, because since we are dealing with presumed prejudice in addition to actual prejudice, a measure of such attitudes in the jury pool is more indicative of the general community's views than if we limit the inquiry to the petit jury. See Coleman, 778 F.2d at 1490.

A total of 244 prospective jurors were individually voir dired. Of these, 55 were excused without in depth questioning, thus leaving 189 jurors that were questioned with some particularity regarding the pretrial publicity issue.

The record reveals that of this base figure of 189 prospective jurors, 186 (or 98.41% ), admitted to having prior knowledge about the facts of this case, about defendants, or about the Cerro Maravilla incident. All of the jurors on the panel that tried appellants fall within this grouping: Jurors No's. 13, 14, 64, 79, 88, 108, 130, 145, 148, 162, 169, 227, and alternates 113, 120, 176, and 220.

A total of 160 prospective jurors, that is 84.4% of the base group, admitted to having received knowledge through the newspapers. This included panel members No's. 13, 14, 64, 79, 88, 108, 130, 162, 169, 227, 245 and 248, and alternates 113, 120 and 176.

A total of 169, or 89.4% of the base group, admitted to having received knowledge via television accounts. All of the members of the panel and all of the alternates were in this grouping.39

Ninety-two prospective jurors admitted to having received information through the radio. Only four regular jury members, No's. 169, 227, 245 and 248, and two alternates, 176 and 270, admitted to this grouping.

A total of 141 prospective jurors, 74.6% of the base group, watched or heard the Senate hearings. Eleven of the twelve petit jurors (No's. 13, 14, 64, 79, 88, 108, 162, 169, 227, 245, and 248) and all four alternates (113, 120, 176 and 270) fall in this category.

I have indicated that prior knowledge about a case does not necessarily disqualify a prospective juror. See Patton, supra, and other cases cited ante. However, the existence of prior knowledge does raise a presumption of possible bias, Irvin, supra, 366 U.S. at 728, 81 S. Ct. at 1645, requiring closer scrutiny to determine not only the existence of opinions but additionally to conclude whether the circumstances surrounding the formation of the juror's views support a finding that they can reasonably be set aside. As previously indicated, the inquiry made in voir dire was pro forma and tainted by the juror's prior knowledge of the voir dire. An analysis of the quantity and quality of the pretrial information admittedly known by the panel is relevant to such analysis.

This analysis shows that 143 prospective jurors, that is 75.6% of the base group, admitted to substantial familiarity with the facts, issues and evidence to be presented in the trial of this case. Within this grouping are panel members No's. 64, 79, 88, 108, 130, 162, 169, 227, 245 and 248, and alternates 113, 120, 176, and 270. The nature of the information known by jurors is important. For example, prospective jurors that admitted to having knowledge regarding key witnesses included: Jurors No's.: 2 (Ortiz-Molina), 31 (Ortiz-Molina, Marte), 41 (Gonzalez-Malave), 50 (Cartagena), 52 (Ortiz-Molina, Marte), 56 (Marte), 62 (Marte), 68 (Gonzalez-Malave, Marte), 69 (Gonzalez-Malave), 71 (Ortiz-Molina), 82 (Ortiz-Molina), 84 (Ortiz-Molina, Marte), 92 (Ortiz-Molina), 103 (Figueroa-Vivas, Quinones, Colton), 108 (Ortiz-Molina), 113 (Marte), 120 (Marte), 155 (Cartagena), 175 (Marte), 218 (Ortiz-Molina), 219 (Ortiz-Molina; "I remember everyone who testified there"), 226 (Marte), 234 (Ortiz-Molina, Carmelo Cruz, Gonzalez-Malave), 238 (Gonzalez-Malave), 245 (Marte), 250 (Cartagena, Gonzalez-Malave). Knowledge regarding important participants in the Senate hearings: Jurors No's. 113 (Sen. Hernandez-Agosto, chief counsel Rivera-Cruz), 117 (Rivera-Cruz), 146 (Rivera-Cruz), 153 (Rivera-Cruz), 157 (Rivera-Cruz), 169 (Rivera-Cruz), 171 (Rivera-Cruz), 226 (Rivera-Cruz), 2 (Rivera-Cruz), 3 (Rivera-Cruz), 31 (Rivera-Cruz), 118 (Rivera-Cruz). Knowledge regarding individual defendants or the alleged victims: Jurors No's. 52 (the defendants), 3 (all the defendants), 10 (defendant Torres-Marrero), 39 (Dario Rosado and Soto-Arrivi), 56 (defendant Perez-Casillas), 50 (defendants Perez-Casillas and Quiles-Hernandez), 55 (Dario Rosado and Soto-Arrivi), 69 (Dario Rosado and Soto-Arrivi), 84 (Soto-Arrivi and Dario Rosado), 86 ("I remember many of the policemen"), 90 (Soto-Arrivi), 96 (defendant Quiles-Hernandez), 117 (defendant Gonzalez-Perez; this juror pointed out Gonzalez-Perez at defendants' table), 121 (defendant Perez-Casillas), 157 (defendant Torres-Marrero), 167 (Dario Rosado), 175 (Soto-Arrivi), 185 (defendants Quiles-Hernandez, Gonzalez-Perez, and Perez-Casillas), 203 (identified the defendants in court), 250 (defendants Perez-Casillas and Gonzalez-Perez, the latter of which was identified in court), and 278 (defendant Perez-Casillas).

We cannot here reproduce all of the instances of detailed knowledge of evidence by jurors, brought out in voir dire. It was, however, overwhelming. Suffice it to point to the more salient examples.40

THE COURT: [Do] you know anything at all about the facts of this case that is going to be tried?

Juror: Well, who doesn't know about Maravilla? I have read it, I see it on T.V., I speak with other persons ...

THE COURT: What do you remember seeing concerning the defendants on television, try to remember.

Juror: Well, how they, these young men was was go there to Cerro Maravilla located where they was the tower. I see in the television the incident, the tower, channel 7 tower, and I see there the boys, when they was dead, everything that they pass in the television news, and....

Well, I remember when, I can't remember the name of the driver [Ortiz-Molina] when he give his testimony, and ...; And the other guy, the guy who works in channel 7, I can't remember the name now [Marte]; Well I know two guys [were] killed in Cerro Maravilla, and what everyone knows; Well, I remember that, two guys was killed in that place and ... [they were killed] by the police.

That the police ... circled them and killed them ... The two boys, the young boys, Arnaldo and the other one, and Carlos Soto-Arrivi.

THE COURT: Have you read anything about this case or concerning the defendants or the Cerro Maravilla incident in the newspapers of Puerto Rico?

Juror: Definitely, I think everybody has.

He remembered the testimony at the Senate hearings, "About how they slain the victims, how they were supposedly shot".

Well, that the officers killed these guys with no reason to kill them.

Well, my opinion is from the time that the driver [Ortiz-Molina] who took the two men to Cerro Maravilla, say that they were pleading ... [f]or not to be killed.

[I remember] ... when they interviewed Marte ... and the driver [Ortiz-Molina], and maybe two or three other persons ... [W]hen he [Marte] cried, and he told what happened and everything ... about how many shootings he heard, and what they told him ... [t]he policemen, they told him not to talk....

When the first investigation was made by the policemen, after that the case was closed. Okay, the second investigation made by the Senate of Puerto Rico, I read another version, another situation, came to my mind. Well, in that investigation I read, I heard that those people, the two boys that was in Cerro Maravilla, they were still alive, under what circumstances they died, I don't know ... After that I made my own personal question, if they were there still alive, why they were not arrested, or something.

... I remember that two young men were taken to Cerro Maravilla by a detective, undercover [Gonzalez-Malave], and they took a driver [Ortiz-Molina], a taxi driver, and they went to Cerro Maravilla, and they met the police there and had trouble there, they started shooting, and they the two young men were killed; [I expect defendants to testify] [b]ecause they [defendants] should say the truth of what happened, what really happened in Cerro Maravilla.

The part that I remember was the way the killing of these people, because I don't like; that because for make my decision on that.

That two young pro-independence young men were killed and that there have been many versions of what happened exactly. People involved who have said one thing at one time and later changed their story.

Well, that two persons were killed by policemen. Well that these two persons, at the time of the killing were in defense,41 they were not armed at the time of the killing....

I would say it is a case that is involved, some members of the police, and the Senate tried to prove that, and that is the main thing ... [F]or me, my opinion the Senate is trying to prove that the Government or some people working for the government did something wrong.

Well, I think everything has been said at the Senate hearing about the case of Maravilla....

... [E]verybody ... comment about the Cerro Maravilla ... How they took two guys, how they were killed....

Well, I know that from the hearings, all the previous case, or the events that happened there changed, that is what I can tell you about the hearings. You know that everything changed ... Well, what happened over there at Cerro Maravilla, the way that the people died.... Well at the first time I recall that they said that those two died in a shooting. From the hearings they say that they died in another form, not in exchanging shots.

I would say that two persons died and the persons say there was an entrapment; [Marte's] name was mentioned, by all people in my office.... I believe that he was the person in the [T.V.] station.

... The first thing that I heard was something happening up in Cerro Maravilla, where a couple of independentistas were gunned down, when they were trying to explode the communication tower. A couple of weeks later, or months later, something came in the news that there was something wrong with the official version, then nothing happened for a couple of months or year, then all of a sudden all of this these things exploded. That there was a cover up and all that.... Well, apparently there is a discrepancy between the official version and what actually happened....

I don't remember his name [a policeman], I know he was saying he was innocent and it came out to be he was guilty ... He was saying that he wasn't directly involved in what had happened, and then one of the other policemen some talks came out to be that he was there, or something like that.

THE COURT: In your mind ... do you associate that person ... with this case?

Juror: Yes, because it is the same case.

Juror: Well, as I understand, some teenagers went up there to destroy some electrical poles, or something to do with communications, and they were stopped by the police, and the police interfered with them and didn't really give them a chance to defend themselves ... the police arrived and without notifying or identifying themselves or giving these people a chance or, identifying themselves as police, well they just shot them.

Well, I heard that a group of policemen took these two guys to Cerro Maravilla, and over there they killed them, that is what I heard ... I heard also that they give false information to the people that was investigating, the judge and the attorney.

Cerro Maravilla was just like it was described on the way the television described it during the Senate investigation.

... [T]wo members of the Independence party went to Cerro Maravilla mountain with the purpose to destroy one or two communicate towers. Located in this place. I know also that an undercover agent infiltrated in the members of the Independence party ... The undercover agent had informed all of the information that he know, that he discover to the first lieutenant and sergeant that supervise their work. Then the two youngest boys member of the Independence party went to the Cerro Maravilla ... And that place was situated in many agents in different position to observe when the two boys come ... to destroy the towers ... then after that information all I know is that the two members of the Independence party have guns and when the Police agent advised the youngest boys to quit and halt. And one of them perhaps two take out guns and try to shot to the agents and that place establish a fight shooting fight between two both sides Police agents and the two members of the Independence party ... And the two youngest boys died in the firing shooting fire and then the things the general comments in different form to analyze act in this place. And then after the witness Marte [Marte] who declare ... he said that the agent killed the two youngest boys when the boys are on the ground. And without arms and kneeling, were kneeling for the purpose for the defense. Lying on the ground and they began to push and to kick the boys and to hit them with arms with guns and with different arms and to take to punish the boys.

... [T]he way ... they killed the two guys that were over there ... that they were pushed and they were kicked with their feet and after that they said they have to be killed, something like that.

Well, I have heard that there were two guys that were going to bomb a tower and they were brought there by a guy named ... Gonzalez-Malave, and as I have heard, the policemen that they say they killed them, in the first place they say there was a shooting but they later, they found that there was no such a shooting; that they killed them in cold blood. That was it. I found out that instead of shooting, it was cold-blooded murder.... I saw it in the case. It was shown on T.V.

THE COURT: Are you referring to the Senate hearings?

Juror: Yes, ma'am.

A total of 81 prospective jurors (or 42.85% of the base group) either admitted to having an opinion as to the guilt or innocence of appellant defendants, or such a conclusion is clearly surmisable from the nature of their answers. This group included seven of the jury panel that heard the case (No's. 79, 88, 108, 130, 227, 245 and 248) and one alternate member (No. 270).

Again, we reproduce only the more salient examples of jurors' stated opinions during voir dire:

The only belief I have is that it was always wrong.

Well, I live in Puerto Rico and the persons that live here say no, they are not talking the truth....

The Senate hearing gave a clear picture of what happened over there.

Well, this case is, they committed perjury, I really can't explain, but I don't have any opinion.

... [W]hat everyone knows ... Well, when I saw or read the news, well, I got my own conclusions, and the reason I came here in the case ... I got my opinion from the news ... I think that the last two persons was murdered.

... [T]he police officers killed these guys with no reason to kill them.

... [T]he results [of the investigation] were shocking. It sort of, you lose faith in mankind sometimes, when you hear things like this. I thought the guilt was proven, like I told you, I thought that [the Senate hearings] was the trial. I think that it was a general, people were shocked in general....

Well, I have discussed much of the case with my family, and with my fellows at work, and we almost have a unanimous opinion about the case.

THE COURT: What has been the basis for your discussions? What has been your sources of information?

Juror: The newspapers, the radio and the T.V.

I believe in the guilt of some.

THE COURT: On what do you base that opinion?

Juror: On the facts that I heard ... on the television.

Well, the way I think is that somebody killed them, who did the shooting, I'm not sure, because I wasn't there. But I mean somebody, some of them are guilty, I know somebody shoot.

Well, in the first investigation, by 1978, I would say, was one version, and the second, made by the Senate of Puerto Rico, was the other version. Now, listening to the second investigation at the Senate, I made my own thing in this case ...

... [T]o tell you the truth, I have been following this case, in T.V., since the beginning, and I read in the paper almost everyday, so I think I have my decision already made.

I just spoke with my father about the prejudice ... [p]erjury I mean ... I said if they committed perjury, well, that was wrong on my opinion, that was not good ... [H]e think that the police committed perjury ... [B]ased on the public hearings, well I think they committed perjury ... [t]hree or four [of them] ... Based on what I saw on T.V., because of the hearings.

It is a moral problem of mine.

THE COURT: What is your moral problem?

Juror: That I think that the persons must have principles and tell the truth.THE COURT: What does that have to do with this case?

Juror: Well, what I have read and heard, I think that it was a problem, because the persons don't, I think, that is what I think....

THE COURT: Who are those persons, who do you refer to?

Juror: The group.

THE COURT: What group?

Juror: The police.

Well, I think that I will be prejudiced....

Yes, I have a formed opinion.... It is an opinion on guilt, yes.... I had the opinion, I have to be sincere, when the incident occurred, I always had the feeling that something was wrong ... [since 1978, 1979] and the hearings, the second part of the hearings that what I saw on video cassette, was like a confirmation of my feeling about the case.

Well, I have my own personal opinion ...

THE COURT: On what do you base that opinion?

Juror: On the hearings in the Senate, which brought out the truth about what happened at Cerro Maravilla ... [T]he truth that has been held for such a long time, and it finally came out.

Yes, I have an opinion [as to guilt or innocence].

... I think that some of them are lying, and some people did force others to lie....

Some persons commit, assassinate to them.

Well, I think that some of what I heard, that it was wrong....

THE COURT: What opinion do you have?

Juror: Well, that the persons who were killed there were killed unjustified. It wasn't a murder that should have been realized.... I am afraid that I have some opinion regarding the incident itself, as it came out in the Senate ... I believe there was two murders there ... As far as I can see, and as I understand it, the police involved in that incident had the opportunity to arrest the suspects, and I think they had it under control, until somebody decided they had to be killed.

It could be that I am influenced, from what I heard in the news ... Well, I believe that a person like myself is influenced, by what I have heard on the news, radio, newspaper, comments ... Well, I would take one of the sides, without really hearing what is said in court. I would hear, but I would already have an idea formed.

I believe there was a murder there....

Like what happened there is like a murder happened there ... that two persons were ... entrapped in a place without defense ... But about the evidence that was shown, a number of persons can conclude that happened, what everybody knows ... Personally, I have a tendency to think that [they are] maybe guilty.

... I heard that the information they gave the judge was not really what happened at Cerro Maravilla ... Well, I may have some idea, according to what I saw in television.

THE COURT: What idea may you have?

Juror: Well, that maybe there something strange happened over there at Maravilla, and the people never knew the real truth of what happened over there.

Well, I guess they could have gave the boys an opportunity at Maravilla ... [T]hey should have arrested them and not killed them, you know.

Juror No. 253:There were entrapment, the obstruction in general ... What I heard is that the whole truth at the beginning was not told ... After the Senate hearings so many declarations was change....

... [A] murder was committed there ... There was committed a murder and that the persons that were in the place lied about what happened there ... [T]he policemen that were there ...

THE COURT: On what source of information did you base your opinion on?

Juror: On the information that was presented [at the Senate hearings].

THE COURT: ... I am asking you if you have an opinion on the guilt or innocence.

Juror: Guilty.

THE COURT: Could you read an impartial verdict based only on the evidence without taking into consideration the extensive publicity that the case has had?

Juror: No.

... I found out that instead of shooting it was cold-blooded murder.

THE COURT: And is that opinion an opinion on the guilt or innocence?

Juror: I think they killed them. They killed them.

Other incidents during the course of voir dire

While voir dire proceeded with the jury still unsequestered, media coverage continued unabated, a fact commented upon by the trial judge in various ways.

On February 7, 1985, the third day of jury selection, defense counsel for Colon-Berrios called the court's attention to the fact that the voir dire was being made public and "was a lead story in one of the major daily newspapers." The court observed that "this voir dire is being read by everybody." On the following day, even after disqualifying a juror, the judge opted to continue the entire line of voir dire questions "because the press is speculating."

On February 18, 1985 the tenth day of jury selection, Moreno-Morales' counsel moved the court for a mistrial or continuance on the basis of an article published in El Reportero. This motion, which was joined by the other appellants was denied in open court without further ado. When the motion was renewed the following day, with similar results, the judge indicated: "I wish to ... remind the attorneys that each juror is appraised that he or she is not to read the newspapers or watch news or listen to the radio, anything about Cerro Maravilla. I give that admonition to each juror that passes through this court. See, I cannot be pinpointing every time there is something in one of the newspapers, because I will be adding an endless list of things. I did add that concerning the Governor of Puerto Rico, because there was no way really that a potential juror could know that the Governor's message to the Legislature, which takes up multiple subjects, would focus on anything having to do with Cerro Maravilla ..." (Emphasis added).

The Governor's State of the Commonwealth Address

As previously indicated, in his inaugural address on January 2, 1985, an address which received live television coverage throughout Puerto Rico as well as wide press and radio coverage, the newly elected governor made various references to the Cerro Maravilla incident, including a condemnation of the "entrapment and summary execution of [the] two independence supporters by police at Cerro Maravilla." These statements were made barely one month before the commencement of the trial on February 5, 1985.

The district court's remarks regarding the Governor of Puerto Rico, however, referred to an episode which took place on the night of the 14th of February, the ninth day of the selection of the jury, as part of the Governor's annual State of the Commonwealth speech before a joint session of the Legislature. This event also received news coverage throughout the Island, including live television and radio attention. It was front page news in the newspapers that appeared throughout Puerto Rico on February 15.

The Governor, during the course of his speech, stated that Soto-Arrivi and Dario Rosado had been the "victims of official acts of violence."42 He expressed sympathy for the families of the deceased and indicated that he would introduce a special bill in the legislature to compensate them for their losses.

On the day following the speech, all appellants moved for a finding of contempt against the Governor, claiming a violation of the court's order of January 18, 1985. They also moved for mistrial, alleging in substance three basic points: (1) that prospective jurors, even if they followed the court's instructions not to expose themselves to Cerro Maravilla publicity, had no way of knowing beforehand that this would be a topic included in the Governor's speech; (2) that because of the nature and circumstances of the speech, there was great likelihood that prospective jurors, including those as to whom voir dire had been completed, had been exposed to the Governor's remarks regarding Cerro Maravilla; and (3) that the Governor's speech amounted to an official statement that defendant-appellants were guilty.

The court denied both motions, ruling that the Governor was not covered by the order of January 18th, and opting on the second point to inquire in the voir dire regarding juror exposure to the Governor's speech. Those jurors already questioned would be recalled for questioning regarding their possible post-voir dire contamination.

Fifty-three prospective jurors admitted to having seen, heard or read the Governor's address. These included three of the jurors who actually heard the case (No's. 130, 227 and 245) and one alternate (No. 176). The testimony of several of the prospective jurors on this issue is of interest:

... I heard he was recommending the legislature to some legislation about compensating the victims of the persons that died in Cerro Maravilla ... I think it was a mistake on the Governor to make such a recommendation to the legislature ... I think that the Government should wait till the cases that are being here in the court, should be resolved.

... [T]hat he was going to compensate the families for the death of the boys ... Well, I think he was not supposed to talk about that, because he knows right now is a trial.

On the eighteenth day of trial, February 28, 1985 the jury was finally empanneled, and forthwith sequestered.

Conclusions regarding the pretrial publicity issue

From the record of the publicity and the voir dire it is clearly established that its volume and reach were all-encompassing. The issues surrounding this case received massive media attention throughout the District of Puerto Rico, commencing immediately after the Cerro Maravilla events took place in 1978, and continuously thereafter until the jury was empanneled in mid-February, 1985.

The notoriety of Cerro Maravilla reached such proportions that it became a popular pastime to visit the scene of the incident. For example, Juror No. 169, who sat on the panel that decided appellants' fate, visited Cerro Maravilla with his wife and children about six months prior to the time of his voir dire. When he arrived there were so many cars he could not drive in. He found a grotesque circus-like spectacle, with "music [playing and] people dancing in the main road." Hot dogs were for sale as well as "[T-]shirts ... [with] the picture of Hector Rivera-Cruz [chief Senate investigator]."43

The evidence also abundantly proves the "pervasive, intense and inflammatory" nature of this publicity. Cf. United States v. Medina, 761 F.2d at 19. The very issues tried before the district court were first litigated in the news media and, more specifically, in the televised vistas. In this first "trial" before the populace, appellants were publicly branded as murderers and perjurers, a viewpoint repeatedly echoed by the news media and by exalted community representatives. This continued up to and including the time of trial.

The passage of time is irrelevant in this case, because there was no hiatus in the barrage of pretrial publicity. It continued unabated for eight years, up to the trial of appellants. We cannot discount the cumulative effects that eight years of unremitting publicity had on the conscious and unconscious attitudes and opinions of those subjected to its influence. See generally Irvin, Public Opinion and Propaganda, Crowell Company, New York (1950); Goggin & Hanover, Fair Trial v. Free Press: The Psychological Effect of Pre-Trial Publicity on the Juror's Ability to be Impartial; A Plan for Reform, 38 So.Cal.L.Rev. 672, 676-83 (1965); Surette, supra, at 20-21, 36-37, 127, 153-56, 247-49; Hennessy, supra, at 305-06, 313-14; Graber, Crime News and the Public, Praeger Publishing, New York (1980), pp. 26, 28-29, 40-42, 50-51. See also Padower-Singer, Singer & Singer, Legal and Social-Psychological Research on the Effects of Pre-Trial Publicity on Juries, Numerical Makeup of Juries, Non-Unanimous Verdict Requirements, 3 Law & Psychology Rev., pp. 71, 75-76 (1977); Kline & Jess, Prejudicial Publicity: Its Effect on Law School Mock Juries, 43 Journalism Quarterly, pp. 113-114, 116. Although the vistas, with their reenactment of the Cerro Maravilla shoot-out and the adverse testimony of Marte and the unindicted co-conspirators, took place in the later half of 1983, the voir dire testimony of the prospective jurors makes it clear beyond reasonable doubt that its dramatic impact was still fresh in the minds of the venire. Even as late as February 28, 1985, the district court, in ruling on a "Jencks" motion indicated that "[o]ne must not forget that ... the abundant pretrial publicity and televised investigation surrounding this case was undoubtedly harmful to defendants in the sense that they were tried by the Senate and the media without the due process safeguards afforded to any defendant ..." Furthermore, from the nature of the voir dire answers it is impossible not to conclude that a general opinion prevailed regarding appellants' guilt or innocence.

This jury, with its intimate knowledge of the case, would have made an ideal jury in England during the Middle Ages. See 3 W. Blackstone, Commentaries 375. But that notion of what a jury should be "was explicitly rejected by the Sixth Amendment provision that a defendant is entitled to be tried by an 'impartial jury'." Gannett Co. v. DePasquale, 443 U.S. 368, 385, 99 S. Ct. 2898, 2908, 61 L. Ed. 2d 608 (1978). If to this we add the Governor's speech, taking place in the middle of perhaps the most delicate stage in the process, the jury selection, we should forcefully conclude that there existed a probability of prejudice by reason of the pretrial publicity.

At first glance, it would appear that Patton v. Yount, supra, presents a difficult hurdle for appellants to overcome. In that case the voir dire showed that all but 2 of the 163 veniremen had heard about the case, and that 126, or 77%, admitted they would carry an opinion into the jury box. Patton, supra, 467 U.S. at 1029, 104 S. Ct. at 2887. Furthermore, 8 of the 14 jurors and alternates actually seated admitted that at some time they had formed an opinion as to defendant's guilt. Id. The Court of Appeals overruled the trial court's finding of impartiality, holding that under Irvin, supra, the adverse pretrial publicity had created a presumption of prejudice which overcame the jurors' claims of impartiality. Id. at 1031, 104 S. Ct. at 2888. In reversing, the Supreme Court indicated that the Court of Appeals had failed to apply the appropriate standard of review for overturning the trial court's findings of impartiality, which the Court ruled was a question of "historical fact." Id. at 1036, 104 S.Ct. at 2891. Such a finding could only be reversed on appeal for "manifest error." Id. at 1031, 104 S.Ct. at 2889.

The Court then proceeded to enumerate the factors present in Patton, which in contrast to Irvin, prevented a finding that the district court had committed manifest error. In Patton, when defendant was first tried in 1966, there existed extensive adverse publicity and a sense of outrage in the community. However, defendant's second trial occurred four years later, in 1970, "at a time when prejudicial publicity was greatly diminished and community sentiment had softened." Id. at 1032, 104 S.Ct. at 2889. In support of this conclusion the Court pointed to the fact that in the year and a half preceding the commencement of voir dire, each of the two county newspapers had an average of less than one article per month related to the trial; that even then, most of the articles that appeared were brief announcements related to trial dates and scheduling; and that the transcript of the voir dire not only showed the sparse nature of the publicity and minimal public interest in the second trial, but also that the lapse of time had softened or effaced previously held opinions. Id. at 1032-33, 104 S.Ct. at 2889-90.

Those circumstances, ruled the Court, contrasted with the Irvin situation, where there had existed "during the six or seven months immediately preceding [that trial], 'a barrage of newspaper headlines, articles, cartoons and pictures ... unleashed against [the defendant]'." Id. at 1032, 104 S.Ct. at 2889.

There is no question but that the present situation more closely resembles that faced by defendant in Irvin, than that in Patton.

To begin with, as I have previously indicated, the evidence is clear that in the present case the prejudicial publicity unabatedly pursued appellants for 8 years and into the trial. The appendix to the January 15, 1985 motion for continuance alone points to 189 newspaper articles dealing with the Cerro Maravilla incident in the period since October 9, 1984. To this we must add at a minimum the Governor's inaugural speech on January 2, and the critical State of the Commonwealth Address in the middle of voir dire on February 14, 1985. In contrast to Patton, the voir dire record shows that the jury in this case was still fully cognizant of Cerro Maravilla and of the "first trial" of appellants before the television cameras of the Senate of Puerto Rico. As the district court indicated:

As [the hearings] developed, defendants, in effect stood trial before an entire community and guilt was adjudicated ...

Perez-Casillas, supra, at 805.

The majority emphasizes that in Patton, 77% of the venire members had an opinion as to defendant's guilt or innocence, while in this case, even if my statistical analysis is accurate, only 42.85% of the venire is so contaminated. See ante at 735 n. 12. There is no question that such a distinction between the Patton and the present situation exists. But focusing on such a distinction misses the forest for the trees. The whole point of this exercise is, that when the pretrial publicity reaches a certain level, there is a presumption of prejudice irrespective of jurors' answers. Holbrook v. Flyns, supra; Coleman v. Kemp, supra ; Delaney v. United States, supra. Statistics are then irrelevant, because regardless of whether the jurors claim they have or have not opinions, or whether they say they will put them aside, the risk of trying defendants before jurors so exposed is too great. As I have indicated the publicity surrounding the present case establishes a situation clearly distinguishable from Patton.

The present situation is more akin to Rideau v. Louisiana, 373 U.S. 723, 83 S. Ct. 1417, 10 L. Ed. 2d 663 (1962), wherein defendant's act of confession was broadcast over the local television station. I find the majority's attempt to distinguish that case from the present situation totally unconvincing. In overturning defendant's conviction in the face of prejudicial pretrial allegations, the Court stated:

For anyone who has ever watched television the conclusion cannot be avoided that this spectacle, to the tens of thousands of people who saw and heard it, in a very real sense was [defendant's] trial--at which he pleaded guilty to murder. Any subsequent court proceeding in a community so pervasively exposed to such a spectacle could be but a hallow formality.

Id. at 726, 83 S.Ct. at 1419 (emphasis in original).

The transcript of the voir dire in the present case shows not only that the veniremen were fully aware of the evidence presented at the "first trial" in the Senate, but some of them actually believed that the Senate vistas were a trial.44

As previously indicated, I am concerned also with a related matter, posed by the Court in Gannett Co. v. DePasquale, 443 U.S. 368, 378, 99 S. Ct. 2898, 2904, 61 L. Ed. 2d 608 (1978), in the context of publicity concerning a suppression hearing:

Publicity concerning pretrial suppression hearings such as the one involved in the present case poses special risks of unfairness. The whole purpose of such hearings is to screen out unreliable or illegally obtained evidence and insure that this evidence does not become known to the jury. Publicity concerning the proceedings at a pretrial hearing, however, could influence public opinion against a defendant and inform potential jurors of inculpatory information wholly inadmissible at the actual trial.

The danger of publicity concerning pretrial suppression hearings is particularly acute, because it may be difficult to measure with any degree of certainty the effects of such publicity on the fairness of the trial. After the commencement of the trial itself, inadmissible prejudicial information about a defendant can be kept from a jury by a variety of means. When such information is publicized during a pre-trial proceeding, however, it may never be altogether kept from potential jurors.

Citations and footnotes omitted. As previously alluded to, one of the problems faced by appellant-defendants was an apparent supplementation by the jury of facts not in the record. In this respect, it should be remembered that eleven of the twelve panel members, and all four alternates admitted having watched or heard the Senate hearings.

The political prominence of Cerro Maravilla, as evidenced by its electoral importance as well as its being the subject of island-wide speeches by the Commonwealth's chief executive, are in sharp contrast with the situation in Patton. Last, but not least, the volume of press attention, both before and after August 22, 1984, also show a dissimilar factual setting to Patton, with this case receiving a tremendous amount of coverage in contrast to Patton. When all of this is taken together with the findings of the district court in its August 22, 1984 order, see ante at 755-57, particularly regarding statements made against appellants by prominent Puerto Rico community and religious leaders, ante at 755-56, I find the majority's conclusion that a quasi-benign attitude existed in the "Puerto Rico public" towards appellants to be nothing short of remarkable. See ante at 736-737.

From the totality of the circumstances it must forcefully be concluded "that the pretrial publicity was so pervasive, intense and inflammatory as to produce a 'trial atmosphere utterly corrupted' " by this publicity. United States v. Medina, 761 F.2d 12, 19 (1st Cir.1985).

I reach this conclusion even in the face of the district court's efforts to preserve an atmosphere appropriate to a trial in which citizens' liberties were at stake. Notwithstanding such efforts, there comes a point when only so much can be expected from the efficacy of voir dire.45 Sheppard, supra, 384 U.S. at 351, 86 S. Ct. at 1516; Irvin, supra, 366 U.S. at 728, 81 S. Ct. at 1645. The propriety of a trial court's acceptance, at face value, of jurors' denials of bias also has its limit. This limit is surpassed when circumstances such as have coalesced in this case take place. As observed by this court in Delaney v. United States, 199 F.2d 107, 112-13 (1st Cir.1952):

One cannot assume that the average juror is so endowed with a sense of detachment, so clear in his introspective perception of his own mental processes, that he may confidently exclude even the unconscious influence of his preconceptions as to probable guilt, engendered by a pervasive pre-trial publicity.

In view of its findings in its August 22, 1984 order and the events that transpired thereafter, the trial court committed manifest error by not continuing the trial. Id.

In so ruling I cannot overlook the fact that this error was induced at least in part by the government's insistence that the court set the case for trial, even after the findings in the August 22 order, and while the electoral process in Puerto Rico continued full blast. Unless the government sought to gain advantage from the effects of the pretrial publicity, its insistence on an immediate trial is particularly incomprehensible if we consider the government's admission during oral argument that its case would not have been prejudiced in any way by a continuance of even up to a year's duration. This is a position in marked contrast to that taken by the government before the district court. It would appear that in its eagerness to win convictions, the government lost sight of the affirmation of the Supreme Court in Brady v. Maryland, 373 U.S. 83, 87, 83 S. Ct. 1194, 1196, 10 L. Ed. 2d 215 (1962) to the effect that: "Society wins not only when the guilty are convicted but when criminal trials are fair; our system of the administration of justice suffers when any accused is treated unfairly. An inscription on the walls of the Department of Justice states the proposition candidly for the federal domain: 'The United States wins its point whenever justice is done its citizens in the courts.' "

The majority, although not directly hanging its hat on this point devotes sufficient attention to it to warrant a response. Without quite deciding the issue, the majority implies that defendant-appellants' failure to move for a change in venue affects their pretrial publicity claim, or at least "is significant." See ante, at 739.

In my view what is significant is why this was not done and why the failure to so move is in any event irrelevant.

This Court is at least partially responsible for defendant-appellant's failure to move for a change in venue. Our decision in In re San Juan Star Co., 662 F.2d 108 (1st Cir.1981) cannot be as lightly treated as is done by the majority. That case involves the civilian counterpart of the present criminal case. See also Colon-Berrios v. Hernandez-Agosto, 716 F.2d 85 (1st Cir.1983). That action involved a suit by the relatives of the two suspected Cerro Maravilla terrorists alleging violation of their federal civil rights by, among others, the present appellant-defendants. At the request of the civilian defendants the district court issued a protective order to prohibit dissemination to the news media of material discovered through depositions taken in that proceeding. The district court determined that this order was necessary because "the amount of publicity being generated would make it 'difficult if not impossible for the defendants to obtain an impartial jury'." Id. at 116. In upholding the district court's ruling, we said:

[I]t is clear that the court found at least a reasonable likelihood of a material harm to defendants' right to a fair trial. We think this conclusion well-supported by both the massive amount of publicity undisputably attending this case and by the emotionally-charged nature of the trial itself ...

Id. at 117. This court answered a challenge of overbreadth in the protective order by saying that:

[G]iven the probability that any additional publicity of deposition contents would be damaging to defendants' fair trial rights ... we think this arguable overbreadth alone insufficient to invalidate the court's order.

Id. (Emphasis in the original). It is important to keep in mind, that at the time of this pronouncement by this court, the televised hearings had not even begun. I therefore fail to see how, if in 1981, this court ruled that any additional publicity to mere deposition contents was damaging to defendants' rights in a civil trial, a similar result is not mandated for a criminal trial commencing in early 1985, after the televised hearings of 1983, the election of 1984 and the other incidents previously recounted, which really generated massive publicity.

More directly on the change of venue issue, this court went on to say that "Puerto Rico is singularly unsuited to a change of venue." In re San Juan Star Co., 662 F.2d at 117. The majority downgrades this holding by implying that this citation is out of context with the ruling and because it "was not a major pronouncement." Id. Considering the latter statement first, I was not aware, until now, that the decisions of this court were subject to some type of ranking. I was under the impression that the public was entitled to rely, to the same extent, on all of our decisions.

As to the first point, whether or not this statement is out of context, it is perhaps best to reproduce the entire pronouncement:

The conclusion that the alternatives that should be considered [i.e., alternatives to the protective order], would be unavailing finds support in several factors peculiar to this case. As appellees have argued, Puerto Rico, is singularly unsuited to a change of venue, and postponement of a trial of such urgent proportions could seriously jeopardize important interests in its resolution. On the district court's conclusions that such intermediate measures as voir dire, instructions, or sequestration would be unavailing in the face of the jury pool's "saturation," we must defer to that court's discretion and its closer familiarity with the nature of the publicity involved.

Id. (citation and footnote omitted). It appears to me that the cited language regarding venue is in context and on point. This court was considering the very issue before us, i.e., the effect of pretrial publicity on the ability of the district court to select an uncontaminated jury in Puerto Rico in a case involving substantially the same issues as are presented in this appeal. Defendant-appellants in this case had good grounds for relying on that ruling and thus not requesting a change in venue. Any error in this respect was induced by this court. We are at least morally estopped in this regards.

The government, however, is legally estopped from raising this issue on appeal. The government in opposing the various requests for continuance never once suggested a change in venue as an alternative. In fact at oral argument counsel for the government candidly admitted that the government would have opposed such a motion had it been filed by defendant-appellants. I might add also, that the district court never invited such a motion either, obviously cognizant of the inappropriateness of such a procedure because of the "several factors peculiar to this case." Id. Cf. Dennis v. United States, 302 F.2d 5, 8 (10th Cir.1962) ("In ruling on [defendant's] pretrial motions for severance and continuance on the grounds of prejudicial publicity, the trial court indicated that the proper remedy was change of venue, and invited such motion."), rev'd on other grounds, 384 U.S. 855, 86 S. Ct. 1840, 16 L. Ed. 2d 973 (1966).

The reason why the government raises this issue on appeal, but failed to promote a change in venue before the district court is obvious: it was quite content with reaping the benefits of a situation which it helped to promote.46 The testimony of the key witnesses in the Senate hearings, Marte, Cartagena, Carmelo Cruz and Martinez, the same persons who were key witnesses for the government in this case, could not have taken place without the granting of immunity from prosecution by the federal government from the various perjuries they had previously committed. It was these grants of immunity that permitted them to appear with impunity in the Senate hearings. The federal authorities, in granting this immunity had full knowledge that these persons would incriminate defendant-appellants in a proceeding which would receive the widest possible diffusion in the news media, including television. Their testimony would include a reenactment on television of the Cerro Maravilla incident amounting to a "trial." See ante 762 note 41. The federal government is thus accountable for the pretrial effects of this massive pretrial publicity and should be the one to bear the burden of its impact on the trial, not appellant-defendants. See Rideau v. Louisiana, supra.

Irrespective of how it is phrased, the effect of the majority's ruling regarding this issue is to require that defendant-appellants give up their constitutional right to vicinage⁴⁷ in order to correct a situation that not only is not of their own making, but is one which they attempted to contain. See In re San Juan Star, supra; Colon-Berrios, supra. The majority, without citing any authority for its novel theory, imposes "a significantly heavier burden" upon appellants to show prejudice by reason of the widespread publicity, because they failed to seek a change in venue. See ante at 738. Cf. United States ex rel. Darcey v. Hardy, 351 U.S. 454, 463, 76 S. Ct. 965, 970, 100 L. Ed. 1331 (1956) (failure to move for change in venue not dispositive). The majority's position is constitutionally unsupportable. See Johnston v. United States, 351 U.S. 215, 220-21, 76 S. Ct. 739, 742-43, 100 L. Ed. 1097 (1956); Salinger v. Loisel, 265 U.S. 224, 232-33, 44 S. Ct. 519, 522, 68 L. Ed. 989 (1924); Delaney v. United States, 199 F.2d 107 (1st Cir.1952). See also, Note, The Right to Venue and the Right to an Impartial Jury: Resolving the Conflict in the Federal Constitution, 52 Univ. of Chic.L.Rev. 729 (1985); Broeder, The Impact of the Vicinage Requirement: An Empirical Look, 45 Neb.L.Rev. 99 (1966). I know of no authority that allows placing a criminal defendant in such a "Catch 22" situation. Cf. Carter v. Kentucky, 450 U.S. 288, 301, 101 S. Ct. 1112, 1119, 67 L. Ed. 2d 241 (1980) ("a defendant must pay no court-imposed price for the exercise of his constitutional privilege not to testify"); Griffin v. California, 380 U.S. 609, 614, 85 S. Ct. 1229, 1232, 14 L. Ed. 2d 106 (1964) (same). I might add that I know of no other constitutional right that is guaranteed by two provisions of the Bill of Rights such as is the right to vicinage, an obvious indication of the importance of vicinage in the scale of constitutional rights.

The cases relied upon by the majority for its contention are inapposite. Although in many of them a change in venue was requested and denied, in none of them does the Court require such a procedure as the quid pro quo for validating a criminal defendant's right to a fair trial. The only case on point is in fact one from this circuit. It holds contrary to the majority's ruling. See Delaney v. United States, 199 F.2d 107 (1st Cir.1952).

In Delaney we reversed the district court's failure to grant a third indefinite continuance because of pretrial publicity, notwithstanding defendant's failure to file a motion for change of venue despite the district judge's specific urging to said effect. Id. at 115. In rejecting the position taken by the panel majority in this appeal, our court said in Delaney:

Under the Sixth Amendment the accused enjoys the constitutional right to a speedy and public trial, "by an impartial jury of the State and district" wherein the alleged crimes are charged to have been committed--in this case the District of Massachusetts. The right to apply for a change of venue is given for the defendant's benefit and at his option. He is not obliged to forego his constitutional right to an impartial trial in the district wherein the offense is alleged to have been committed; and under the circumstances of this case we do not think that the defendant's appeal stands any worse for failure on his part to apply for a change of venue.

Id. at 116. In attempting to distinguish the circumstances referred to in this quotation from those in the present case, the majority distinguishes the "indistinguishable." See Chardon v. Fernandez, 454 U.S. 6, 8, 102 S. Ct. 28, 29, 70 L. Ed. 2d 6 (1981).

The publicity involved in Delaney came as a result of hearings conducted by a Congressional committee concerning Delaney, who had been Collector of Internal Revenue for the District of Massachusetts until suspended from office and indicted for various crimes. Motion pictures and sound recordings were permitted at the hearings, which received wide publicity,48 although it "was intensified in the Boston area." Id. at 111. However, the nature and extent of the publicity was not even remotely comparable to that in the present case. As in the present case, however, many of the witnesses who testified at the hearings also testified against Delaney at his trial. Our court expressed its consternation about this situation:

In this respect the committee hearing afforded the public a preview of the prosecution's case against Delaney without, however, the safeguards that would attend a criminal trial ... [T]he witnesses who testified were not subjected to cross examination by counsel for the accused.

Even more damaging, perhaps, was the fact that the testimony thus publicly heard by the committee ranged far beyond matters relevant [i.e., admissible] to the pending indictments ...

Id. at 110.

We rejected the district court's reliance on the efficacy of last minute trial procedures to cure the effects of the pretrial publicity, and said:

No doubt the district judge conscientiously did all he could, both in questions he addressed to the jurors at the time of their selection and in cautionary remarks in his charge to the jury, to minimize the effect of this damaging publicity, and to assure that defendant's guilt or innocence would be determined solely on the basis of the evidence produced at the trial. But ... [quoting Mr. Justice Jackson] [T]he naive assumption that prejudicial effects can be overcome by instructions to the jury ... all practicing lawyers know to be unmitigated fiction. One cannot assume that the average juror is so endowed with a sense of detachment, so clear in his introspective perception of his own mental processes, that he may confidently exclude unconscious influence of his preconception as to probable guilt, engendered by a pervasive pretrial publicity. This is particularly true in the determination of issues involving the credibility of witnesses.

Id. at 112-13 (emphasis supplied).

I need only to remind the reader of the overwhelming, massive and continuous nature of the pretrial publicity in the present case up to and including the trial, in contrast to that in Delaney, which was mild by comparison. I cannot see how it is possible to ignore the unqualified circuit precedent established by Delaney, on the merits of the pretrial publicity issue as well as on the majority's contention regarding venue.

Delaney has been cited with approval not only by the Supreme Court, but also by most circuits, including our own. See Sheppard v. Maxwell, 384 U.S. 333, 354 n. 9, 86 S. Ct. 1507, 1518 n. 9, 16 L. Ed. 2d 600 (1965); Scales v. United States, 367 U.S. 203, 258, 81 S. Ct. 1469, 1501, 6 L. Ed. 2d 782 (1960); Irwin v. Dowd, 366 U.S. 717, 727, 81 S. Ct. 1639, 1645, 6 L. Ed. 2d 751 (1960); Smith v. American Indus. Research Corp., 665 F.2d 397, 399 (1st Cir.1981); Reinstein v. Superior Court, 661 F.2d 255, 258 (1st Cir.1981); United States v. Guido, 597 F.2d 194, 198 (9th Cir.1979); Ramos Colon v. U.S. Attorney for District of Puerto Rico, 576 F.2d 1, 8 (1st Cir.1978); United States v. Perrotta, 553 F.2d 247, 249 (1st Cir.1977); United States v. Bailey, 480 F.2d 518, 521 (5th Cir.1973); United States v. Dellinger, 472 F.2d 340, 373 n. 48 (7th Cir.1972); United States v. Spock, 416 F.2d 165, 180 n. 35 (1st Cir.1969); Rolon Marxuach v. United States, 398 F.2d 548, 552 (1st Cir.1968); United States v. Stromberg, 268 F.2d 256, 270 (2d Cir.1959). We did not hesitate in that case to exercise our supervisory powers to prevent an injustice.

Nor will I concede, as the majority claims, that if my position prevails, in a situation in which a defendant fails to seek a change in venue, the result will be that he/she will be "insulated from trial indefinitely." See ante at 734. The first thing to be kept in mind in this regard is that the holding of a speedy trial in criminal cases is principally a protection against the undue pretrial detention of a person who, although charged, is still presumed innocent. Although there is also an undisputed societal interest in seeing that justice be speedily applied, the paramount interest is to see that justice be done. A trial should not be so speedy as to bring about unfairness to a criminal defendant. The government should not be allowed to use its quest for a speedy trial as a weapon to place a criminal defendant at a disadvantage. More in point in this case, by the government's own admission, a continuance of up to one year would have made no difference to its case. A continuance within such a period would have been appropriate and would have avoided this unseeming haste. Although no absolute assurance is possible, such a continuance, or one within such a period, would almost certainly have eliminated or greatly reduced the prejudice caused by the massive publicity and the election furor. Such a postponement would not have been an indefinite insulation from trial.

The record is patently clear that the pretrial publicity, particularly that generated by the Senate hearing, was prejudicial and inflammatory. See Order of August 22, 1984, Perez-Casillas, 593 F. Supp. at 805-06 ("From that moment onward, charges of murder against defendants were constantly voiced and exposed by the media ... [t]aking for granted the guilt of defendants as expressed by the last witnesses at the hearing."). It is impossible for anyone who has read the record to deny that the jury pool, including the eventual petit jury that tried the case, was fully aware of this inflammatory publicity. It is thus the height of naivete to conclude that jurors exposed to such pernicious influence could, notwith-standing their good faith assertions to the contrary, give defendants a fair and impartial trial free from such pressures. I would not want such a jury for myself, or my brother, or my son--or for that matter, anyone.

I shall not discuss further the majority's conclusion regarding appellant Colon-Berrios, with which I concur, except to restate, that the jury's verdict in that case was obviously influenced by its exposure to extra-judicial information received as a result of the massive pretrial publicity previously detailed. That is a problem pointedly censured by the Supreme Court in Gannett Co. v. De Pasquale, supra. There is every reason to believe that it also tainted the jury as regards the remaining defendants.

The remaining defendants, Rios-Polanco, Bruno-Gonzalez, Gonzalez-Perez, Quiles-Hernandez, Mateo-Espada, Torres-Marrero and Perez-Casillas, appeal rulings by the district court concerning (1) severance, (2) the admission of inflammatory evidence, and (3) the length of the sentences imposed. Although these are discrete issues and shall be so discussed, in my opinion these matters are inexorably related to the pretrial publicity issue, a problem which weaves through this entire case like a snake in the grass.

Appellants' Motion in Limine to Exclude Details of the Alleged Murder

The remaining appellants contend that the district court erred in admitting evidence concerning the details of the alleged murders. The evidence vividly portrayed how the two young men, while unarmed, handcuffed and kneeling, were kicked and pummelled with rifle butts by several officers, until redfaced and crying. Evidence was presented to show that Dario Rosado was killed with a single shotgun blast to the chest while handcuffed and that Soto-Arrivi was shot twice. It is the same evidence presented before the Senate in the televised hearings by substantially the same witnesses. See Perez-Casillas, 593 F. Supp. at 805-06. Appellants argue that the evidence was unduly inflammatory and prejudicial, and was irrelevant to prove the elements of the government's case, which charged conspiracy and perjury.

Rule 401, Fed.R.Evid. provides that evidence is relevant if it has "any tendency to make the existence of any fact that is of consequence to the determination more probable or less probable than it would be without the evidence."

Relevant evidence generally is admissible under Fed.R.Evid. 402. Fed.R.Evid. 403 states in part that relevant evidence "may be excluded if its probative value is substantially outweighed by the danger of unfair prejudice...." Wide discretion as to the relevancy of evidence is vested in the trial judge, both as to its probative value and its prejudicial impact. United States v. Marler, 756 F.2d 206, 217 (1st Cir.1985). The trial judge's ruling on the issue will not be disturbed absent an abuse of discretion. United States v. Coast of Maine Lobster Co., 557 F.2d 905, 908 (1st Cir.), cert. denied, 434 U.S. 862, 98 S. Ct. 191, 54 L. Ed. 2d 136 (1977).

The trial court admitted the evidence at issue, without any limiting instruction, because it felt that it was necessary to prove the conspiracy and perjury charges.

All appellants were charged with perjury. However, only four of the perjury counts charged against two of the remaining appellants, Moreno-Morales, and Torres-Marrero, refer to whether the officers lied when they said that they neither assaulted, nor saw anyone else assault the two men.

In order to prove violations of 18 U.S.C. Secs. 1621 and 1623, it is necessary to prove that the answer given to a question is false. That is, the government must show that the allegedly perjured statement is not true because it is contrary to fact. See generally 2 Modern Federal Jury Instructions Sec. 48.02, p. 48-15 and cases cited therein. Accordingly, for the government to establish that these two lied when denying that Soto and Rosado were assaulted and struck by police officers while in custody and before being shot, it had to present evidence that the assault had indeed occurred. See, e.g., United States v. Kehoe, 562 F.2d 65, 67-68 (1st Cir.1977); see also United States v. Masters, 484 F.2d 1251, 1253 (10th Cir.1973). Thus, there was no way to avoid presenting the testimony at issue as regards the above-mentioned counts against appellants Moreno-Morales and Torres-Marrero.

The correctness of the admission of this evidence as to the conspiracy and the remaining perjury charges presents a different picture, however. Under Count 1 of the indictment, the defendants were charged with conspiring to obstruct justice by participating in a cover-up of the Cerro Maravilla events. As overt acts, in furtherance of the conspiracy the government alleged that the appellants had: (1) agreed to conceal the identity of material witnesses; (2) met at the scene of the incident on August 2, 1978 and created a false version of the events; (3) given false written statements in August, 1978 reporting among other things, that there had been only one volley of shots with the two men when killed; (4) made false declarations before a federal grand jury regarding the same matters testified to in the written statements; and (5) given false testimony in depositions in the civil suit brought by the alleged victims' relatives. It was additionally charged that (6) codefendant Angel Perez-Casillas met with Miguel Marte-Ruiz, the Channel 7 television technician who had been at Cerro Maravilla on July 25, 1978, and urged him to give false testimony concerning the events at Cerro Maravilla before a federal grand jury; and (7) several codefendants, Angel Perez-Casillas, Jaime Quiles-Hernandez, Nazario Mateo-Espada, and other persons both known and unknown to the grand jury, had met at Cerro Maravilla and agreed to continue to tell their previously adopted version of the events of July 25, 1978. The remaining perjury charges involved the alleged false testimony given in support of the conspiracy.

As demonstrated by the above, the remaining charges are devoid of references to the details surrounding the killings. The government certainly could have proven them by presenting evidence of the facts alleged as overt acts in furtherance thereof. Thus, contrary to what the district court ruled, and the majority affirms, details on how the killings occurred were not a necessary link in the government's conspiracy case, nor in the perjury charges against the defendants other than Moreno-Morales and Torres-Marrero.

The government argues that under Fed.R.Evid. 404(b) evidence of uncharged conduct may be offered to establish motive. In this instance the government contends that the challenged evidence is particularly probative of motive: given the heinousness of the offenses, the incentive to conceal them was great. Rulings under Rule 404(b) however, are also subject to the constraints of Rule 403 regarding cumulativeness and unfair prejudice.

The fact that two citizens were deliberately killed by law enforcement officers while in custody is in itself serious enough to have lead those involved to conspire as charged to prevent prosecution. Thus, there was no need to present additional testimony bearing on the issue of motive. Moreover, the impact of evidence of police officers assaulting them under the circumstances described is highly inflammatory. Considering the questionable additional probative value that it may have had in establishing motive, it would appear to me unquestionable but that this evidence should have been limited. It also appears to me that in admitting this evidence the jury was reminded of the inflammatory Senate hearings. Thus, the government again unduly reaped the benefits of this prejudicial pretrial publicity.

As conducted, this was not a conspiracy and perjury trial, it was a trial for murder.

Admittedly, the government was entitled to present evidence as to the details of the deaths of the alleged terrorists in its case against defendants Moreno-Morales and Torres-Marrero. To prevent a miscarriage of justice, however the other appellants should have been separately tried. Given the massive pretrial publicity and the inflammatory nature of the evidence presented by the Government, it is pure fiction to conclude that the cumulative effect of this situation was not one that brought about a "spill over effect" against these other defendants.

A judge has a continuing duty at all stages of the trial to grant a severance if prejudice appears. See Schaffer v. United States, 362 U.S. 511, 516, 80 S. Ct. 945, 948, 4 L. Ed. 2d 921 (1960). There is no question but that this prejudice appeared in all its force during voir dire and during the government's presentation of the inflammatory evidence. Irrespective of which of the various standards enunciated by this circuit is used,49 the remaining appellants should have received a trial separate from defendants Moreno-Morales and Torres-Marrero. The unquestioned effect of admitting the inflammatory evidence and the failure to sever the trials, when taken together with the massive pretrial publicity, resulted in appellants' trial, conviction and sentencing for murder, not for conspiracy and perjury as charged.

In my view, appellants have no valid grounds for overturning the sentences imposed, as they do not exceed the allowable statutory limits nor do they per se constitute cruel and unusual punishment. United States v. Francesco, 725 F.2d 817, 823 (1st Cir.1984). However, even though they may not provide independent grounds for reversal, the length of the sentences imposed is relevant to the issues previously discussed because they are demonstrative of the extent to which the massive pretrial publicity and the inflammatory evidence presented at trial influenced even the trained mind of the judge who supervised the trial. Although undoubtedly perjury is a serious offense, I doubt that there are many, if any, cases in which sentences as harsh as those imposed in this case have been imposed for such violations. Even admitting to their legality, when I see how a seasoned trial judge is carried away under the influence of the events previously described in this opinion, I cannot but reaffirm my belief that appellants are constitutionally entitled to a new trial.