United States of America, Appellant, v. Richard John Lafuente, Also Known As Ricky Lafuente, Appellee.united States of America, Appellant, v. John Emmanuel Perez, Also Known As John Perez, Appellee.united States of America, Appellee, v. Richard John Lafuente, Also Known As Ricky Lafuente, Appellant.united States of America, Appellee, v. John Emmanuel Perez, Also Known As John Perez, Appellant, 54 F.3d 457 (8th Cir. 1995)

Annotate this Case
US Court of Appeals for the Eighth Circuit - 54 F.3d 457 (8th Cir. 1995) Submitted Dec. 15, 1994. Decided April 28, 1995. Rehearing Denied June 7, 1995

Lynn E. Croods, Asst. U.S. Atty., Fargo, ND, argued. Dennis D. Fisher and Norman G. Anderson appeared on the brief, for appellant.

Jonathan T. Garaas, Fargo, ND, argued, for appellee.

Before ARNOLD, Chief Judge, FAGG and WOLLMAN, Circuit Judges.

WOLLMAN, Circuit Judge.


This is but the latest phase of the extensive litigation growing out of the August 1983 beating and murder of Jerome Edward ("Eddie") Peltier on the Devils Lake Indian Reservation in North Dakota.1  Richard John LaFuente was convicted of first-degree murder for driving the car that ran over and killed Peltier, and John Emmanuel Perez was convicted of second-degree murder. Both are in prison. LaFuente is serving a life sentence. Perez is serving a sentence of fifteen years on the second-degree murder conviction, to be followed by a five-year sentence on his conviction for witness tampering. (We refer to LaFuente and Perez collectively as "LaFuente.")

Following our rejection of his other challenges to his conviction, LaFuente moved for a new trial on the basis of newly discovered evidence and allegations of prosecutorial misconduct. The district court denied the motion without first conducting an evidentiary hearing. We reversed the order of denial and remanded the case with directions that the district court conduct an evidentiary hearing and make findings on a number of disputed issues. United States v. LaFuente, 991 F.2d 1406 (8th Cir. 1993).

On remand, the district court conducted a four-day evidentiary hearing, receiving testimony from some forty-five witnesses. The district court entered findings adverse to LaFuente on the issues set forth in our remand. Notwithstanding those findings, the district court ordered a new trial. United States v. LaFuente, No. C2-85-069 (D.N.D. May 5, 1994). The government has appealed. Both defendants have cross-appealed, challenging the district court's findings and arguing that the evidence they presented at the hearing established their innocence.2  We hold that the district court's findings on the issues on which we remanded are not clearly erroneous and that no new trial is warranted.

Our remand directed the district court to make findings on a number of issues. The first of these was LaFuente's allegation that Shirley Greywater, one of the principal witnesses at trial, had attended, in the absence of her counsel, an ex parte chambers conference with the government and the district court at which Ms. Greywater was told that she would be jailed if she spoke with defense counsel or the press.

The district court found that no such meeting ever occurred, citing the affidavits filed by the government (which included one from the district judge in question) and referring to Ms. Greywater's inability to remember whether the alleged incident occurred before, during, or after the trial, as well as her inability to provide any description of the chambers in which the meeting was supposed to have taken place. We conclude that the district court's findings on this issue are not clearly erroneous.

We directed that the district court review its earlier finding that Ms. Greywater's post-trial recantation of her trial testimony that she had been at the Juarez party did not satisfy the requirements of the five-part test set forth in United States v. Liebo, 923 F.2d 1308 (8th Cir. 1991).

On remand, the district court, without making specific findings with respect to the recantation, found that Ms. Greywater's testimony was not critical to the government's case and that her recantation was without sufficient impact to affect the verdict. Without recounting the details of Ms. Greywater's trial testimony, we conclude that the district court's finding that the absence of that testimony would not have affected the verdict is not clearly erroneous.

With respect to our direction that it consider the allegation that Ms. Greywater had been threatened by law enforcement officers and that she had been provided with information by one of these officers, the district court found that no credible evidence had been produced to support these allegations. This finding is not clearly erroneous.

The district court found that, in direct contradiction to LaFuente's allegations, LaFuente, 991 F.2d at 1409, John Wells gave testimony at the evidentiary hearing that had the effect of corroborating Mary McDonald's testimony. We quote a portion of the district court's finding on this issue, for it reflects, accurately we believe, the tenor of much of the evidence introduced at the evidentiary hearing:

On the witness stand, John Wells testified in a manner directly opposite to what was expected by LaFuente's counsel, and in effect corroborated McDonald's trial testimony. Neither side appears willing to claim him now as his testimony is harmful to the defense, and while helpful to the Government, the Government has shown he could not have known what he testified to. Mr. Wells seems to be typical of much of the proffered testimony.

Memorandum and Order at 12.

With respect to our direction that it consider LaFuente's allegation regarding Linda Alberts' proposed testimony, LaFuente, 991 F.2d at 1409, the district court found that testimony to the same effect had been given at trial and that anything said by Ms. Alberts, as well as by another witness to the same effect, would at best be cumulative, a finding with which we agree.

With respect to the proposed testimony of Kevin Mindt, LaFuente, 991 F.2d at 1409-10, the district court found (with a caveat regarding its later discussion of prosecutorial misconduct):

Kevin Mindt appears to the court to be a pathological liar. No one used his testimony at trial, as no one apparently trusted him to conform to any previous statement or sworn testimony. He plead guilty to a perjury charge and the court finds his recitations of hearsay regarding Fred Peltier to be as incredible as his differing versions of direct testimony.

Memorandum and Order at 12-13.

We conclude that this finding is not an unwarranted characterization of Mindt's testimony, and it is certainly not clearly erroneous.

Our remand also directed the district court to consider LaFuente's allegations of prosecutorial misconduct regarding the government's failure to disclose exculpatory evidence. LaFuente, 991 F.2d, at 1410-11. The district court found that no persuasive evidence was produced to support this claim, stating in part that " [m]aterial which looked impressive in affidavit form proved ephemeral in testimony." Memorandum and Order at 14. We conclude that this is an accurate assessment of the evidence offered in support of this claim, and we hold that the district court's finding is not clearly erroneous.

In response to our directive that it determine how many individuals were improperly subpoenaed and whether any prejudice resulted from the pretrial conferences conducted pursuant to these subpoenas, LaFuente, 991 F.2d at 1411-12, the district court found that no showing was made that any witness testified any differently as a result of such practice and that on at least one occasion that practice was used at the request of defense counsel. Accordingly, the district court found that no relief was warranted on the basis of improper use of subpoenas. We conclude that the record supports the district court's finding on this issue.

With respect to the allegation that LaFuente was prejudiced as a result of the improper payment of witness fees, LaFuente, 991 F.2d at 1412-13, the district court found that no prejudice had been shown to have resulted from such payments. We conclude that the record supports this finding.

Likewise, we believe that the record supports the district court's findings with respect to the material witness warrants and the protective custody of material witnesses issues. LaFuente, 991 F.2d at 1413-15.

In sum, then, after reviewing each of the issues of newly discovered evidence upon which we remanded for findings, the district court either found that no credible evidence existed to support the alleged evidence, or, in the case of Shirley Greywater's recantation, that the recanted testimony had no impact upon the verdict. The district court's findings are not clearly erroneous, and thus it did not err in denying the motion for new trial based upon the allegations of newly discovered evidence.

Likewise, we find no error with respect to the district court's denial of a new trial on the basis of the prosecutorial errors discussed above.

We turn, then, to the issues upon which the district court ordered a new trial. The district court identified three prosecutorial errors, the combined effect of which, the court concluded, deprived LaFuente and Perez of a fair trial.

The first prosecutorial error cited by the district court was the government's promise to dismiss perjury and other felony charges against key witnesses if their trial testimony helped the government. (The perjury charges stemmed from inconsistent testimony before the grand jury in this case; the witnesses initially denied any knowledge of the murder, but days later changed their story and gave detailed accounts of the crime. To add more confusion, two of the indicted witnesses who testified in favor of the government at trial--and had grand-jury based perjury charges dismissed as a result--recanted at the evidentiary hearing held on remand.)

Second, the government forced witnesses by subpoena to undergo interviews with prosecutors in violation of Fed. R. Crim. P. 17, which does not contemplate the use of subpoenas for ex parte information gathering.

Third, the district court cited irregularities in material witness proceedings.

Although the district court's opinion seems to turn on the cumulative effect of the foregoing prosecutorial errors, a close reading shows that the agreement to dismiss charges only after witnesses testified in support of the government's case is what most troubled the district court.

It is not uncommon for witnesses under indictment to trade testimony for lenient treatment later. E.g., Brewer v. Nix, 963 F.2d 1111, 1113 (8th Cir.), cert. denied, --- U.S. ----, 113 S. Ct. 273, 121 L. Ed. 2d 201 (1992). In general, such an arrangement will not taint a conviction unless the jury was not aware of the arrangement or the defendant was not allowed to cross-examine the witness about it. Brewer, 963 F.2d at 1113; Byrd v. Armontrout, 880 F.2d 1, 10 (8th Cir. 1989), cert. denied, 494 U.S. 1019, 110 S. Ct. 1326, 108 L. Ed. 2d 501 (1990). A properly informed jury capable of determining what weight to give interested witnesses is usually an adequate procedural safeguard against coerced testimony. See United States v. Hernandez, 779 F.2d 456, 461 (8th Cir. 1985); cf. United States v. Duke, 50 F.3d 571 (8th Cir. 1995) (jury's knowledge of witness's self-interest adequate safeguard against false testimony). Here, such protections were in place: the jury knew about the pending charges, and the defendants had the opportunity to cross-examine on any resulting witness bias. They were entitled to no more.

Although the government's tactics may have been heavy-handed, they also were above-board. Accordingly, we conclude that there was no prosecutorial misconduct in having witnesses testify while under indictment.

We note in passing that it is important to distinguish witness intimidation of the type alleged here from the far more serious problems posed when the prosecution or court uses its substantial coercive power to induce a witness not to testify, see Webb v. Texas, 409 U.S. 95, 98, 93 S. Ct. 351, 353, 34 L. Ed. 2d 330 (1972) (per curiam); United States v. Blackwell, 694 F.2d 1325, 1337 n. 14 (D.C. Cir. 1982); U.S. v. Teague, 737 F.2d 378, 384 (4th Cir. 1984), or when the prosecution or court threatens to renege on a duly concluded plea bargain, see Blackwell, 694 F.2d at 1336-39. Neither of such situations was present here.

Turning to the second element of prosecutorial misconduct relied on by the district court in granting a new trial, it is true that the use of subpoenas to secure ex parte witness interviews before trial is improper. The government does not dispute this, and indeed has acknowledged its mistake.

There is no doubt that such an abuse of process can form an important link in a chain of reversible prosecutorial misconduct. United States v. Morrison, 535 F.2d 223, 225-26 (3rd Cir. 1976). But without additional factors such as the witness badgering and intimidation that took place at the improperly secured interviews in Morrison, we are reluctant to say that bare abuse of process like that alleged here rises to the level of reversible error. LaFuente does not allege any Morrison-style improprieties at the tainted interviews, and the district court in fact found that the illegal subpoenas did not change any testimony. Memorandum and Order at 13. Viewing this abuse of process in the context of the entire criminal proceeding, then, we cannot with confidence say that it had any effect at all on the trial.

The last element of prosecutorial misconduct relied on by the district court was "doubtful material witness proceedings." While the opinion below does not make clear the dubious proceedings that swayed the district court, we surmise that the district court was referring to the fact that " [t]hree essential government witnesses ... were arrested as 'Material Witnesses' ... and their living expenses were paid by the government for a several month period prior to and during the trial." The district court also made reference to the fact that material witness arrest files, like many others in the case, were sealed and that the defense did not know about them before the trial.

This resembles LaFuente's accusation of witness bias due to pending indictments discussed above in part II A, and we resolve it in like manner. At trial, the defense attorneys, though they may not have known that persons were named (and arrested) as material witnesses per se, did know that certain witnesses were being maintained at government expense in some sort of protective custody due to security concerns. The defense had the opportunity to cross-examine on any links between the witnesses and the government, as well as on any bias created by the payments and other support given to these witnesses; a strategic decision was made not to pursue these topics. LaFuente cannot here revisit his decision not to present his inferences of witness bias to the jury. For reasons similar to those set out in our resolution of the issue of bias due to indicted witnesses, we reject this claim of prosecutorial misconduct.

After considering the cumulative impact of the three elements of prosecutorial misconduct relied on by the district court in granting a new trial, we conclude that LaFuente received a fair trial.

This conclusion is only bolstered by the fact that the district court's decision strongly suggests that any prosecutorial overreaching that may have occurred was harmless. For instance, the recantations of allegedly coerced trial testimony by two witnesses at the evidentiary hearing is key to LaFuente's arguments here. Yet the district court rejected the recantation of one critical witness (Frederick Peltier) as not credible. As to the second recanting witness (Shirley Greywater), the district court, as set forth above, found that her testimony was not critical to the conviction. The district court found that the trial testimony of Mary McDonald, who at trial testified that she was in LaFuente's vehicle when it ran over Peltier, survived LaFuente's post-trial attacks "intact." The other eyewitness deemed critical by the district court, Patricia DeMarce, is now dead. According to the district court, the testimony of DeMarce and McDonald alone "would in all probability be sufficient to sustain a verdict."

To obtain a new trial based on prosecutorial misconduct, a defendant must show not just that misconduct occurred, but that it "fatally infected" the trial. Rhodes v. Foster, 682 F.2d 711, 714 (8th Cir. 1982). We cannot focus microscopically on the alleged misconduct, but must instead consider its effect on the broader fairness of the whole trial. Hernandez, 779 F.2d at 458-59, 461. We conclude that any prosecutorial overreaching, in the context of the whole trial, did not affect the verdict.

This has been a troubling case, marked by a level of contentiousness on the part of counsel not seen in most criminal prosecutions. Perhaps the depth of the feeling on the part of both the prosecution and the defense was inevitable, given the complexity of a multi-defendant trial conducted nearly three years after the events that gave rise to the prosecution.

That the government engaged in some conduct that it now concedes was improper only heightened the tension pervading the trial and the subsequent proceedings, including this appeal.

That the government went to extraordinary lengths in preparing for and trying the case is beyond peradventure. That such conduct may have seemed necessary in view of the extraordinary difficulty the government faced in investigating the case is understandable. The initial investigation was stymied by a refusal of potential witnesses to reveal what they knew because of the threats of retaliation that had been communicated to them. Not until Patricia DeMarce, LaFuente's half-sister, came forward nearly two years after Eddie Peltier's death was the government able to renew its efforts to bring Peltier's slayers to book. Given the alcohol-clouded memories of many of the witnesses, young and old alike, and the residual fear flowing from the intimidating tactics of those who stood to be linked to the killing, the government faced a formidable task in preparing and presenting a case against the eleven defendants who were ultimately indicted and brought to trial. It would have been easy for the prosecutors to conclude that the sometimes sketchy memories and the reluctance to testify on the part of those at the scene of the early-morning events on August 27, 1983, made further investigation and prosecution problematical, if not impossible. The government would have been remiss in failing to bend every effort to vindicate the law by seeking out those who were responsible for Eddie Peltier's death. In bending every effort, the government went to lengths not sanctioned by law and rule. Such conduct is not to be condoned, but if, as we conclude, it did not prejudicially affect the verdict, it should not be held to vitiate the convictions.

The order granting a new trial is reversed.

 1

See United States v. LaFuente, 991 F.2d 1406 (8th Cir. 1993); United States v. Cavanaugh, 948 F.2d 405 (8th Cir. 1991); United States v. Grey Bear, 883 F.2d 1382 (8th Cir. 1989), cert. denied, 493 U.S. 1047, 110 S. Ct. 846, 107 L. Ed. 2d 840 (1990); United States v. Grey Bear, 863 F.2d 572 (8th Cir. 1988) (en banc); United States v. Grey Bear, 836 F.2d 1086 (8th Cir. 1987); United States v. Grey Bear, 828 F.2d 1286 (8th Cir. 1987)

 2

Perez raises a jurisdictional challenge to the government's appeal, arguing that the notice of appeal fails to mention him by name, referring only to LaFuente in the caption. The government argues, persuasively we conclude, that the notice of appeal, which names neither defendant in its text, simply mirrored the district court order's own caption, which likewise omitted Perez. It is clear that the government intended to appeal the entire order, which addressed and granted the new trial motions of both LaFuente and Perez. There is no indication that Perez did not get timely notice of the appeal

We also reject the additional argument (made by both defendants) that the government's notice of appeal is insufficient because it refers to a district court "Judgment," while the document in question is actually styled a "Memorandum and Order."

Some case metadata and case summaries were written with the help of AI, which can produce inaccuracies. You should read the full case before relying on it for legal research purposes.

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.