Gary Gathers & Keith Mitchell v. United States
Annotate this Case
Download PDF
Notice: This opinion is subject to formal revision before publication in the Atlantic and Maryland Reporters. Users are requested to notify the Clerk of the Court of any formal errors so that corrections may be made before the bound volumes go to press. DISTRICT OF COLUMBIA COURT OF APPEALS Nos. 09-CO-422, 11-CO-1676, 11-CO-1677, 12-CO-1411, 12-CO-1412 GARY GATHERS & KEITH MITCHELL, APPELLANTS, v. UNITED STATES, APPELLEE. Appeals from the Superior Court of the District of Columbia (FEL-10270-93, FEL-11175-93) (Hon. Russell F. Canan, Motions Judge) (Argued January 16, 2014 Decided October 23, 2014) Seth Rosenthal, with whom Matthew Beville was on the brief, for appellant Gary Gathers. Amit Mehta was on the brief for appellant Keith Mitchell. David B. Goodhand, Assistant United States Attorney, with whom Ronald C. Machen Jr., United States Attorney, and Elizabeth Trosman, John P. Mannarino, and T. Anthony Quinn, Assistant United States Attorneys, were on the brief, for appellee. Paul F. Ensinna and James P. Bair were on the brief for amicus curiae, National Association of Criminal Defense Lawyers, in support of appellants. Philip T. Inglima and Tiffany V. Wynn were on the brief for amici curiae, W. Thomas Dillard, et al., in support of appellants. 2 Before BLACKBURNE-RIGSBY and BECKWITH, Associate Judges, and STEADMAN, Senior Judge. STEADMAN, Senior Judge: Appellants Gary Gathers and Keith Mitchell were convicted of first-degree murder and related offenses in 1994. In this appeal, they assert that at the trial, the government relied on false evidence that was crucial to the government s case and that a new trial is therefore mandated under the holding by the United States Supreme Court in Napue v. Illinois, 360 U.S. 264 (1959). The government admits a Napue violation occurred, but argues that appellants are barred from relief on both procedural and substantive grounds. We are persuaded that the appellants have made their case, however belatedly, and accordingly reverse the trial court s denial of their motion for a new trial. I. Factual Background The basic scenario of this case may be simply put. On August 8, 1993, Wayne Ballard was shot and killed by a gunman in another car as Ballard sat in his own car waiting for a stoplight to change. The passenger in Ballard s car, Eric Lindsay, then seventeen years old, testified that Gary Gathers was the gunman and Keith V. Mitchell was the driver of the car from which Gathers shot Ballard. Lindsay knew Gathers and Mitchell prior to the incident. Both defendants were convicted in a jury trial and their convictions were sustained on direct appeal by an 3 unpublished Memorandum Opinion and Judgment in 1997.1 Before us now is an appeal from denials of appellants motions for collateral relief filed under D.C. Code § 23-110 (2012 Repl.). Among other claims,2 appellants contend that they are entitled to relief under Napue because the government secured the convictions using false testimony from a Metropolitan Police Department ( MPD ) detective ( the Napue claim ). Following a series of evidentiary hearings, the motions court in an extensive order covering the diverse claims denied all relief. The Napue claim arises from the government s motive evidence presented at appellants initial trial. The government s theory was that appellants killed Ballard because he planned to testify against Gregory Gary Gathers s brother in a different trial for the murder of Carlton Gillis. To support this theory, the government presented MPD Detective Ray Crawford, who testified that Ballard 1 Gathers & Mitchell v. United States, Nos. 94-CF-875 and 94-CF-1233 (D.C. April 24, 1997). 2 In particular, appellants assert a claim for a new trial under Brady v. Maryland, 373 U.S. 83 (1963). This claim is not frivolous, see generally Vaughn v. United States, 93 A.3d 1237, 1244 (D.C. 2014), and involves the government s alleged knowledge of alternate-perpetrator evidence in that the leaders of Ballard s crew had threatened to kill Ballard as an eyewitness to two other unrelated shootings. Appellants also seek further review under the Innocence Protection Act but only if no new trial is granted. Given our Napue holding, we need not reach either issue on this appeal. 4 told investigators that he was driving the car in which Gillis was shot. Detective Crawford testified that Ballard had identified Gregory Gathers as the gunman shortly after the shooting. Ballard was the government s only eyewitness in its case against Gregory Gathers. While Crawford s testimony established that Ballard was a cooperating witness, it did not establish that Gary Gathers and Mitchell knew Ballard would testify against Gregory and, therefore, had a patent motive to murder. The government filled that gap with Crawford s testimony at appellants trial relating his own testimony at an earlier preliminary hearing in the Gregory Gathers case, as follows: Q: When you testified at the preliminary hearing [in the Gregory Gathers case], was Mr. Gregory Gathers present in the courtroom? A: Yes. Q: And were you questioned about the eyewitness who had identified Gregory Gathers as the shooter in the Gillis homicide? A: Yes, I was. Q: What information did you provide about who it was that identified Gregory Gathers as the person that killed Carlton Gillis? A: It was that Wayne Ballard was the person that identified Gregory. 5 Q: Did you use the actual name or did you just identify him as someone involved in the scenario of the events on Waller [sic] Place? A: His name was used. Q: And following the preliminary hearing, did Wayne Ballard continue to cooperate with the United States? A: Yes, he did. (Emphasis added). In closing argument, the prosecutor made use of Crawford s testimony, arguing: At a preliminary hearing [in the Gregory Gathers case] on June 2nd, Officer Crawford during cross examination told Gregory Gathers, told a public courtroom that Wayne Ballard was the Government s only witness. The only reason that this man s [Mitchell s] friend[, Gregory,] was in jail facing a murder charge and the only reason that that man[, Gary s, . . .] brother was in jail facing a murder charge was because Wayne Ballard had the courage to stand up. That s the why. That is why Wayne Ballard was executed on August 8th. And later in the closing, the prosecutor reiterated that Ballard s cooperation with the police had been announced in an open . . . courtroom. Finally, in its brief on the direct appeal to this court, the government again repeated this assertion, saying: [i]n late June of 1993, Detective Crawford 6 testified at Gregory Gathers preliminary hearing, where he stated that Wayne Ballard had identified Gregory Gathers as the alleged murder. However, the testimony of Defective Crawford was in fact false. The transcript from the Gregory Gathers preliminary hearing plainly showed that nobody mentioned Ballard s name in any way but referred merely to the driver of Gillis s car as the lone eyewitness against Gregory. II. Legal Analysis A. The Napue Claim A Napue claim invokes the Due Process Clause and arises when a conviction is obtained through the government s use of false evidence. See supra, 360 U.S. at 269. As we recently had occasion to set forth in Longus v. United States, 52 A.3d 836, 845 (D.C. 2012): Napue claims are reviewed de novo and, under this standard, we must conclude that appellant s due process rights were violated necessitating a new trial if: (1) the government knew (or should have known) that testimony proffered by Detective [Crawford] was false, but failed either to correct the falsehood before the jury or to 7 apprise the court and defense counsel about Detective [Crawford s] false testimony in a manner that would have allowed it to be corrected before the jury; and (2) the government cannot show, beyond a reasonable doubt, that the false testimony was harmless in the context of appellant s trial. Once the appellant has made sufficient demonstration of uncorrected false testimony then the burden shifts to the government to show, beyond a reasonable doubt, that the false testimony was harmless in the context of appellant s trial. Id. (citing Napue, 360 U.S. at 271-72) (other internal citations omitted). The touchstone of Napue claims, like all claims based on the Due Process Clause, is the fairness of the trial. Id. at 846-47. Fundamental to the fairness of any criminal prosecution is the government s adherence to its obligations to correct misstatements by any witness and to avoid any mistakes of fact arguing for conviction. See id. at 847-48. During appellants hearing for post-conviction relief, the motions court found, inter alia, that appellants had satisfied the first prong of Napue in that it is clear that the government knew or should have known of the falsity of the testimony, which the government does not contest.3 Nonetheless, the motions In its brief to us, the government frankly acknowledges the prosecutor s error. [T]he government understands that any such clarification effort does not (continued ¦) 3 8 court rejected appellants Napue claim, concluding that they had failed to meet their burden of showing that there was a reasonable likelihood that the false testimony affected the verdict. The motions court inexactly articulated the second Napue prong because, as the subsequently decided Longus case makes clear, the burden of showing harmlessness is on the government rather than the appellants and harmlessness must be proven by the constitutional standard of beyond a reasonable doubt. See id. at 845.4 Applying the Longus standard and making the harmlessness assessment de novo, as we must, Napue, supra, 360 U.S. at 271-72, we cannot conclude that the Napue violation was harmless. ( ¦continued) excuse the prosecutor s inexplicable failure to correct Detective Crawford s misstatement once any clarification effort had failed. Similarly, the government makes no excuse for the prosecutor s equally inexplicable use of this false testimony in his closing argument. The prosecutor in the case was also the prosecutor at the preliminary hearing in the Gregory Gathers case where Ballard s name was not in fact revealed. 4 Our cases have variously described the harmlessness test as whether there is a reasonable likelihood that the perjured testimony could have affected the verdict, O Brien v. United States, 962 A.2d 282, 315 (D.C. 2008) (internal citations omitted), or no reasonable possibility that the falsehood affected the jury's verdict, Woodall v. United States, 842 A.2d 690, 696 (D.C. 2004), or, as in Longus, supra, 52 A.3d at 845, harmless beyond a reasonable doubt. While it has been asserted that at least the first two formulations are substantially equivalent, O Brien, supra, 962 at 314 n.45, Longus, supra, 52 A.3d at 845, correctly cites the statement in United States v. Bagley, 473 U.S. 667, 679-80, 679 n.9 (1985), that the Napue Court s standard whether there was any reasonable likelihood that the false testimony could have influenced the jury is equivalent to that standard applicable to constitutional errors under the later-decided case of Chapman v. California, 386 U.S. 18 (1967); i.e., beyond a reasonable doubt. 9 The government s case here rested in the last analysis on two basic propositions: first, Lindsay s identification of Gathers as the shooter and Mitchell as the driver; and second, the establishment of a motive through Crawford s false testimony that he had revealed Ballard s name as the cooperating witness at the Gregory preliminary hearing. The prosecutor repeatedly stressed the importance of Crawford s false motive testimony throughout the course of the trial.5 Indeed, in its written opposition to a pretrial defense motion in limine, the government characterized this evidence as crucial to the government s theory of prosecution. And at the pretrial hearing, the prosecutor described the fact that Crawford would testify that he identified the driver of the car that Carlton Gillis was in as the most crucial piece of evidence against Gathers and Mitchell.6 Moreover, the A prosecutor s stress upon the centrality of particular evidence in closing argument tells a good deal about whether the admission of the evidence was meant to be, and was, prejudicial. Morten v. United States, 856 A.2d 595, 602 (D.C. 2004). 5 The government argues that Crawford s testimony was harmless as cumulative of other motive evidence principally from Hattie Barber, whom the government itself characterized as not your ideal citizen perhaps. Barber s testimony, as developed in cross-examination, was less than entirely clear as to any theory of general knowledge of Ballard s cooperation, nor did the prosecutor in closing make any significant use of the Barber testimony as establishing motive but, rather, placed his reliance on the Crawford testimony. And while the Gillis shooting occurred in broad daylight affording the killer a possible opportunity to identify Ballard as the driver, these bits of testimony paled in comparison to the (continued ¦) 6 10 testimony of the government s sole identification witness, Lindsay, was attacked in significant respects beyond the chaotic and relatively fleeting nature of the sighting, such as the fact that Lindsay did not make a positive identification of the killers for over a month following Ballard s death and the shifting identification of the car of the killers.7 None of this suggests that the evidence was insufficient to convict appellants, but, in the posture of one of the two major props of the government s case resting on false testimony, the outcome of the case falls significantly short of constitutional impregnability.8 We turn then to the government s procedural arguments for rejecting the belated Napue claim. ( ¦continued) conclusive evidence of Gathers s knowledge provided by Crawford s testimony as repeatedly stressed by the prosecutor. 7 The government introduced as supporting evidence allegedly incriminating statements by Gathers, one during a police interrogation and another supposedly overheard by Hattie Barber. But neither could come close to the challenged Lindsay on-the-spot identification. 8 We so conclude from the trial record itself. In fact, at the § 23-110 hearing in 2011, Lindsay made a complete recantation of his trial identification, although the trial court did not credit it. Moreover, as already mentioned, that hearing also brought out evidence of possible alternate perpetrators of the murder with their own discrete motive. 11 B. The Procedural Obstacle: Cause and Prejudice In Shepard v. United States, 533 A.2d 1278, 1280 (1987), we established a procedural bar to a § 23-110 motion where appellant demonstrably knew or should have known of a potential collateral claim but did not raise it by a § 23110 motion as part of the direct appeals process. Otherwise put, in bringing a subsequent attack in a § 23-110 motion for collateral relief, the movant must establish cause for the failure to raise the claim at the time of direct appeal as well as resulting prejudice. See id.9 In the instant case, the motions court rejected appellants motion because it found that appellants had actual knowledge that Crawford s testimony was contrary to the proffer and, by failing to raise the issue on direct appeal, were barred from making such a claim at this late date. Citing, inter alia, Bruce v. United States, 617 A.2d 986, 992-93 (D.C. 1992), the motions court reasoned that appellants had waived their Napue claim and could not satisfy cause and prejudice. 9 Subsequently, in Doe v. United States, 583 A.2d 670 (D.C. 1990), we made clear that part of appellate counsel s duties is to consider whether the client s interests require the filing of a § 23-110 motion based on ineffectiveness of counsel and, if so, to take steps to that end in addition to pursuing the direct appeal. 12 We must disagree with this conclusion in the context of a Napue claim and hold that, to the contrary, appellants have established cause and prejudice sufficient to overcome any procedural obstacle presented by Shepard.10 A fuller exposition of the facts relating to the Crawford testimony will put the issue in perspective. 10 Appellants urge that the Shepard holding is limited to situations where the claim is ineffectiveness of trial counsel and further note that a Napue claim involves a need to develop information outside of the record on appeal. In the instant case, we will assume that Shepard applies and, given our ultimate holding, we need not reach the issue of whether this potential procedural bar of Shepard applies to all collateral claims of which an appellant knew or should have known. Furthermore, whether or not appellants knowledge vel non was sufficient to trigger the Shepard cause and prejudice requirement, as we shall explain infra, appellants for this Napue claim have satisfied the cause and prejudice requirements of Shepard to warrant collateral consideration here. 13 The first mention of the Crawford testimony came in the government s written opposition to Gathers s motion to suppress certain evidence relating to the Gregory Gathers case. In its opposition, the government stated that among other things, it planned to introduce evidence that, at a preliminary hearing in the Gregory Gathers case, Ballard s identity as the sole government eyewitness was disclosed and that he was cooperating with the government. As already noted, the government asserted that evidence establishing that appellants knew about Ballard s cooperation against Gregory Gathers was crucial to the government s theory . . . that Ballard was executed because he was a cooperating witness. Subsequently, at a pretrial hearing on the motion to suppress, the prosecutor reiterated that he intended to call Crawford to testify about his prior testimony in the Gregory Gathers case. The prosecutor said that Crawford had earlier testified that the driver of the car that Carlton Gillis was in when he was killed was the sole Government eyewitness to Gillis s murder. In the course of a somewhat rambling discussion about what the government intended to show, the following colloquy ensued: [DEFENSE COUNSEL]: I have one question, Your Honor, I want to I m not clear about. Was Wayne Ballard identified by name at the preliminary hearing of 14 Gregory Gathers? I think [the prosecutor] needs to just tell us that again. [PROSECUTOR]: No. That s why it's important that [THE COURT]: He was identified as the driver of the car? [PROSECUTOR]: Right. Which is why I think it's relevant to show that 5:40 in the afternoon and he is driving down the middle of an unbusy street when the shooting takes place. That is to show that Gregory Gathers knew who the driver of the car was. [DEFENSE COUNSEL]: Well, let me just ask something. I m still not clear. Was the testimony of Gregory Gathers preliminary hearing that the driver of that car in which Gillis was killed was the only eyewitness to the Gillis shooting? Is that in effect the [THE COURT]: That s what I understood. [PROSECUTOR]: The only eyewitness that was offered that we had, that was offered at that point to hold Gregory Gathers, yes. [DEFENSE COUNSEL]: But it s also true that Mr. Ballard himself was not identified by name as the driver of that car when that hearing took place? [PROSECUTOR]: That s right. (Emphasis added). However, as already set forth, at the trial itself, Detective Crawford s testimony was squarely to the contrary and was not corrected in any way by the prosecutor but instead was used extensively to convict. 15 An argument that appellants were on notice of the falsity of the Crawford testimony rests on the exchange at the pretrial hearing. The critical difficulty, however, is in equating knowledge of inconsistency with knowledge of actual falsity. Unlike the prosecutor, the defense counsel would be, at best, only on inquiry notice. It would not have been totally unthinkable for defense counsel at trial and on appeal to accept that the first-hand testimony from Detective Crawford on the stand, which conformed to the government s written proffer at the preliminary hearing, was accurate rather than that of the prosecutor made orally in a somewhat confused earlier discussion, especially in light of the ongoing use by the government of the false testimony both in the course of the trial and on appeal.11 And any such lapse by defense counsel pales as against that of the government. The knowledge of the prosecutor markedly exceeded that of defense 11 This case is quite distinct from Bruce, supra, 617 A.2d at 992-93, cited by the motions court and the government, where both parties were indubitably aware of the actually falsity of the testimony that gunshots were fired and, unlike here, the prosecutor unfailingly presented the government s case on the basis of the true fact of an inoperable gun. No obvious reason appears here why Gathers s trial counsel would not have corrected the untruth so adverse to his client had counsel been aware of the falsity, or why appellate counsel, if aware, would not have raised the issue on direct appeal. The fact that counsel attempted to use Detective Crawford s testimony to his client s advantage by rhetorically asking why Lindsay was not killed as well as Ballard appears to be a matter of playing the hand one is dealt. 16 counsel,12 and it is difficult, if not impossible, to reconcile the motions court s finding as to defense counsel with its failure to find actual knowledge on the part of the prosecutor, rather than simply that he should have known. The knowledge of the prosecutor was of the statement s actual falsity, while that of the defense counsel was at best of inconsistency. It is also noteworthy that the judge at the actual trial was also present at both the motion hearing and the trial and yet apparently overlooked the discrepancy, as may have been the actual case of counsel on both sides. But any negligence by defense counsel was utterly outweighed by that of the government. Moreover, a more fundamental principle is at stake, even in the context of prosecutorial should have known. The presumption, well established by tradition and experience, that prosecutors have fully discharged their official duties is inconsistent with the novel suggestion that conscientious defense counsel have a procedural obligation to assert constitutional error on the basis of mere suspicion 12 See supra note 3. Moreover, although the Gregory Gathers transcript was technically available as a public record, the government did not actively share it with Gathers and Mitchell until 2010. The failure formed part of appellants Brady claim. While the government asserts that it is possible that appellants could have had access to the transcript at an earlier point in time, every indication is that this was not the case. 17 that some prosecutorial misstep may have occurred. Strickler v. Greene, 527 U.S. 263, 286-87 (1999) (internal citations and quotation marks omitted). Further, [o]rdinarily, we presume that public officials have properly discharged their official duties. We have several times underscored the special role played by the American prosecutor in the search for truth in criminal trials. Courts, litigants, and juries properly anticipate that obligations to refrain from improper methods to secure a conviction plainly resting upon the prosecuting attorney, will be faithfully observed. Banks v. Dretke, 540 U.S. 668, 696 (2004) (internal citations, quotation marks, and alterations omitted).13 Although judgment finality is undoubtedly of great importance, it must sometimes yield to higher considerations.14 The legislature has specifically chosen not to set any time limits on challenging a conviction on constitutional grounds in § 23-110 ( A motion for such relief may be made at any time. ). It is markedly 13 In the analogous context of Brady violations, this court has consistently upheld the rule that a prosecutor must seek justice before victory. Brady does not authorize the government to engage in a game of hide-and-seek, or require the defense to scavenge for hints of undisclosed Brady material. Vaughn, supra note 2, 93 A.3d at 1256 (internal citations and quotation marks omitted). As the Supreme Court has noted, The procedural-default rule is neither a statutory nor a constitutional requirement but is rather a doctrine adhered to by the courts to conserve judicial resources and to respect the law s important interest in the finality of judgments. Massaro v. United States, 538 U.S. 500, 504 (2003). 14 18 disquieting to think that appellants should stand convicted on what is plainly false evidence highly prejudicial to the outcome where the government knew or should have known of the falsity, however belatedly this falsity may have come to the forefront. We must conclude that, at least in this Napue context, sufficient cause exists to excuse the delay in raising the claim.15 For the foregoing reasons, we deem ourselves compelled to reverse the denial of the motion for a new trial and to remand the case for further proceedings consistent with this opinion. Reversed and remanded. 15 For the reasons already stated, prejudice to appellants here cannot be gainsaid, even with a burden on appellants rather than the government. The trial court equated the prejudice standard for procedural purposes to be equivalent to that on the merits, an approach that has some support in our case law. See McCrimmons v. United States, 853 A.2d 154, 161 (D.C. 2004). We agree with this approach for a Napue claim.
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.
