Griffin v. United States, 336 U.S. 704 (1949)

Mr. Justice Frankfurter

delivered the opinion of the Court.

This case was brought here under § 1254 (1) of Title 28 of the United States Code to review the dismissal by the Court of Appeals for the District of Columbia of an appeal from the denial of a motion for a new trial on the ground of evidence discovered after the petitioner had been convicted of murder in the first degree. 335 U. S. 866. The decisive issue is the admissibility of that evidence. The question arises not through its exclusion at trial but on a motion for a new trial in order to be able to introduce it as newly discovered.

The petitioner, Baxter Griffin, was convicted of the murder of Lee Hunter. The killing was the outcome of a quarrel. Admitting that he shot Hunter, Griffin claimed that he did so in self-defense. His story was that the deceased and he were playing a card game called blackjack, that Hunter demanded a larger share of the pot than was his right, and that upon his refusal to pay,

Hunter “jumped up and started around the table, with his hand in his pocket, and told me he would kick my teeth out of my head.” On cross-examination Griffin added that Hunter threatened to kill him. Accordingly, so his story continued, Griffin shot Hunter as Hunter advanced toward him with his hand in his pocket. This version of the occurrence was contradicted by five Government witnesses. Each testified that petitioner started the argument, and that it had nothing to do with the card game which, according to their account, was over before the fracas began. According to them, this is what happened: Griffin made some remark to Hunter about taking Hunter’s wife and baby around to Griffin’s house; Hunter replied that he would kick petitioner’s teeth down his throat; Griffin thereupon left the house and returned within ten minutes with a gun, and on his return shot Hunter, who had made no move from the spot where he was standing. Griffin admitted that he saw nothing in Hunter’s hand at the time he shot Hunter. On the evidence, as summarized, the jury on March 28, 1947, found Griffin guilty of murder in the first degree; on April 18, 1947, he was sentenced to death; on December 8, 1947, the conviction was affirmed, 83 U. S. App. D. C. 20, 164 F. 2d 903; on March 15, 1948, this Court denied certiorari, 333 U. S. 857.

On May 7, 1948, a little more than a month before the day set for execution, Griffin began the present proceedings for a new trial. It was based on affidavits of his then counsel who averred that it had recently come to his knowledge that the attendant at the morgue had found an opened penknife in the trousers’ pocket of the deceased and that the prosecutor knew of this at the time of the trial but failed to introduce this circumstance in evidence or make it available to the defense. The affidavits further alleged that there was evidence that playing cards were on the floor immediately after the shooting, a fact which

would, had it been known to the defense, have tended to corroborate Griffin’s statement that the card game was in progress at the time of the shooting. An extended hearing was had on the motion for a new trial. The allegation regarding scattered playing cards on the floor at the time of the fatal shooting was adequately met, and this ground for a new trial need not detain us.

As to Griffin’s discovery, after his conviction was affirmed, of the undisclosed knife in the pocket of the deceased, the Government conceded that it knew of this circumstance at the time of the trial and despite that knowledge neither introduced the fact in evidence nor felt any duty to make it known to the defense. The Government justified this on the ground that in its view the circumstance of the knife was inadmissible, since knowledge of its presence in the pocket of the deceased had not been communicated to Griffin either by sight or otherwise. The District Judge took this view of the law and denied the motion for a new trial. In an unreported opinion, he stated, “The question whether a person is justified in attacking an assailant in self-defense must be determined by the facts which were presented to the person who pleads self-defense. He [Griffin] did not know, it appears, that the deceased had an open knife in his pocket, and therefore its existence is irrelevant.” An appeal having been taken, the Government moved to dismiss the appeal on the ground that “the issues raised by appellant’s motion for a new trial were fully explored in the court below and that the disposition made of them by the trial court was manifestly correct.” The appeal was dismissed by a unanimous Court of Appeals, presided over by a judge than whom no one is more alert in protecting the rights of the accused.

Unfortunately, the Court of Appeals evidently thought that the ground for dismissing the appeal was too clear to require explication. It dismissed the appeal without

an expression of views regarding the admissibility of the evidence on which the claim for a new trial rests. It may well have done so on the ground that in the District evidence of this nature is inadmissible. That this was the reason for the dismissal is the view of some members of this Court. The opinion of the Court of Appeals on a later appeal from the denial of a petition for habeas corpus by Griffin lends support to such an interpretation of the summary dismissal of the appeal now under review'. See Griffin v. Clemmer, 83 U. S. App. D. C. 351, 169 F. 2d 961.1 But solicitude for life bars reliance on such an inference, especially since the issue on habeas corpus is quite different from that on appeal from a denial of a motion for a new trial. It seems to us more appropriate for the Court of Appeals to address itself directly to the issue of admissibility. This is so in order to rule out the inference that the Court of Appeals may, in applying United States v. Johnson, 327 U. S. 106, have deemed the denial of a motion for a new trial on the basis of newly discovered evidence solely a matter for the trial court’s discretion.

Were the Court of Appeals to declare that the controverted evidence was admissible according to the law

prevailing in the District, it would have to consider further whether it would not be too dogmatic, on the basis of mere speculation, for any court to conclude that the jury would not have attached significance to the evidence favorable to the defendant had the evidence been before it. If the Court of Appeals had decided that the disputed evidence was not admissible in the District of Columbia on a claim of self-defense and on that ground had sustained the denial of the motion for a new trial, there would have been an end of the matter. It is not to be assumed that this Court would have granted a petition for certiorari to review the ruling since the determination would have been a matter of local law as are the rules of evidence prevailing in the State courts.

We are told, however, that a ruling which did not permit the introduction of “uncommunicated threats” would constitute “egregious error” to be corrected by this Court. Fisher v. United States, 328 U. S. 463, 476. Wig-more is vouched as authority that uncommunicated threats are admissible in “virtually all Courts.” But Wigmore immediately follows the words quoted with a series of qualifications and limitations which prove that there are few questions of admissibility in trials for murder that have occasioned a greater contrariety of views. See 1 Wigmore, Evidence § 111 (3d ed., 1940).2 By way of

example, most jurisdictions hold that evidence of uncommunicated threats is inadmissible where there is clear proof that the defendant took the initiative, or where there is no evidence that the deceased was the aggressor other than the proffered uncommunicated threats. Were this the rule in the District, the dismissal of the appeal may well have been rested on it, since there was weighty

proof that the petitioner was the aggressor. Indeed, for all we know the Court of Appeals might have had in mind a rule concerning uncommunicated threats that would admit them and yet guard against the danger of fabrication by placing upon the trial judge the responsibility of excluding such alleged threats against the defendant in the absence of proof satisfactory to him of some hostile manifestation by the deceased relevant to the killing. At least one State has some such rule. State v. Carter, 197

La. 155, 158, 1 So. 2d 62, 63. This is not to reject as unreasonable a rule, followed by some courts, that would let the evidence in, even where all other witnesses oppose a defendant’s version of the killing.

One thing is clear. There is no “federal rule” on this subject. The decision in Wiggins v. Utah, 93 U. S. 465, does not purport to lay down a general rule, nor does it even formulate the evidentiary problem now in controversy. In that case, in light of the fact that there was no other identification of the aggressor, proof was offered that the deceased had exhibited a pistol a few minutes before the shooting and had said, though out of the hearing of the accused, that “he would kill defendant before he went to bed that night,” and this Court naturally held that this evidence should have been admitted. It did so because “it would have tended strongly to show where that first shot came from, and how that pistol, with one chamber emptied, came to be found on the ground.” Wiggins v. Utah, supra at 470.

But even assuming that the “federal rule” is that the evidence described in the motion for a new trial would be admissible, it does not follow that it must also be the rule for the District of Columbia. This Court, in its decisions, and Congress, in its enactment of statutes, have often recognized the appropriateness of one rule for the District and another for other jurisdictions so far as they are subject to federal law. Thus, the “federal rule” in first-degree murder cases is that, unless the jury by unanimous vote agrees that the penalty should be death, the court must fix the sentence at imprisonment for life. 35 Stat. 1151, 1152, 18 U. S. C. § 567, now 18 U. S. C. § 1111 (1948), Andres v. United States, 333 U. S. 740. But a defendant convicted of first-degree murder in the District cannot look to the jury to soften the penalty; he must be given the death sentence. 31

Stat. 1321, 43 Stat. 799, D. C. Code § 22-2404, Johnson v. United States, 225 U. S. 405. Furthermore, the Court’s decision in Fisher v. United States, 328 U. S. 463, makes clear that when we refused to reverse the Court of Appeals for the District we were not establishing any “federal rule” in interpreting the murder statutes which apply in places other than the District of Columbia over which Congress has jurisdiction. In fact, this Court has been at pains to point out that “Congress . . . recognized the expediency of separate provisions” pertaining to criminal justice applicable exclusively to the District of Columbia in contradistinction to the Criminal Code governing offenses amenable to federal jurisdiction elsewhere. Johnson v. United States, 225 U. S. 405, 418.

Many statutes reflect this distinctive position of the District in matters of criminal law. Compare 35 Stat. 1149, 18 U. S. C. § 516 (“federal” adultery statute), now repealed, 18 U. S. C. p. 2415 (1948), with 31 Stat. 1332, D. C. Code § 22-301 (District adultery statute); compare 35 Stat. 1143, 18 U. S. C. §§ 2031, 2032 (1948) (“federal” rape statute) with 31 Stat. 1322, 41 Stat. 567, 43 Stat. 798, D. C. Code § 22-2801 (District rape statute); compare 35 Stat. 1144, 18 U. S. C. § 2111 (1948) (“federal” robbery statute) with 31 Stat. 1322, D. C. Code § 22-2901 (District robbery statute); compare 35 Stat. 1144, 18 U. S. C. § 466 (“federal” larceny statute), now repealed, 18 U. S. C. p. 2415 (1948),3 with 31 Stat. 1324, D. C. Code § 22-2201 (District larceny statute). In fact, it requires two volumes to contain “all the general and permanent laws relating to or in force in the District of Columbia, on January 3, 1941, except such laws as are of application in the District of Columbia by reason of being laws of the

United States general and permanent in their nature.” See Preface to District of Columbia Code (1940 ed.). If Congress can enact substantive rules of criminal law exclusively for the District of Columbia,4 the Court of Appeals for the District of Columbia ought not to be denied opportunity to formulate rules of evidence appropriate for the District, so long as the rules chosen do not offend statutory or constitutional limitations.

The position of spouses as witnesses strikingly illustrates that the District stands apart from the rule of evidence prevailing generally in the federal courts. The federal courts have held that one spouse cannot testify against the other unless the defendant spouse waives the privilege. Miles v. United States, 103 U. S. 304; Bassett v. United States, 137 U. S. 496; cf. United States v. Mitchell, 137 F. 2d 1006, 1008 (C. A. 2d Cir.). Since this Court in the Funk case left open the question whether this rule should be changed, Funk v. United States, 290

U. S. 371, 373, it presumably is still the “federal rule” for the lower courts. In the District, however, the rule has long been otherwise. Halback v. Hill, 49 App. D. C. 127, 261 F. 1007; Buford v. Buford, 81 U. S. App. D. C. 169, 170, 156 F. 2d 567, 568; cf. Dobbins v. United States, 81 U. S. App. D. C. 218, 157 F. 2d 257; 31 Stat. 1358, D. C. Code § 14-306. Another example is afforded by the fact that the statute just cited also provided that one spouse could testify in favor of the other in cases in the District when the “federal rule” was still to the contrary. Compare Jin Fuey Moy v. United States, 254 U. S. 189; Hendrix v. United States, 219 U. S. 79, both overruled in Funk v. United States, supra.

The problem of the admissibility of the evidence set forth in the motion for a new trial is serious and its wise solution full of difficulty. The problem was apparently not explored below, and at the bar of this Court counsel did not give it the consideration appropriate for determination of a federal issue of general importance. It was not even argued in their briefs. Under such circumstances it is not for us to announce a rule for the District of Columbia. Nothing that has been said concerning the various possible choices is intended as an expression of preference among the competing rules about the admissibility of uncommunicated threats, nor as the slightest restriction upon the freedom of the Court of Appeals to make its own choice. We purposely withhold any expression of opinion on the merits of any of the permissible views on admissibility of this evidence. Certainly nothing in our decisions forecloses the Court of Appeals from selecting any one in the range of choices open to it, each one having some rational basis. That court has heretofore been recognized as the appellate tribunal for determining the local rules of evidence; it also is a court that has active experience with the just

and practical considerations governing trials for murder, plainly outside the preoccupation of this Court.

It is precisely for such reasons that for a decade the Court has declined to review all convictions for first-degree murder in the District of Columbia, with a single exception, and in every one of these cases some local rule of evidence was at least in part involved. The Appendix, infra, p. 719, gives a summary of the legal issues involved in the fourteen cases in which we denied a petition for certiorari. This course of disposition manifests uniformity of respect by this Court for District rulings on evidence.5 Reference to this course of disposition of attempts to secure review here for convictions of mürder in the District in no wise disregards our repeated admonition that denial of a petition for certiorari imports nothing as to the merits of a lower court decision. These denials do not remotely imply approval of the various rulings on evidence made in these cases by the Court

of Appeals for the District. What they do establish is that it has become settled practice for this Court to recognize that the formulation of rules of evidence for the District of Columbia is a matter purely of local law to be determined — in the absence of specific Congressional legislation — by the highest appellate court for the District.

Previous to this case, there was, as has been noted, a single exception to this Court’s consistent refusal, for the past decade, to bring here for review a conviction for murder in the District.6 The disposition of the exception powerfully underlines the significance of the necessity for the Court of Appeals to pass initially on this issue. The conviction in that case was affirmed essentially on the principle that the law of evidence and procedure governing criminal trials in the District of Columbia is in the keeping of the Court of Appeals for the District and is not to be exercised by this Court. “The administration of criminal law in matters not affected by constitutional limitations or a general federal law is a matter peculiarly of local concern. . . . Matters relat

ing to law enforcement in the District are entrusted to the courts of the District. Our policy is not to interfere with the local rules of law which they fashion, save in exceptional situations where egregious error has been committed.” Such were the views which determined decision in Fisher v. United States, 328 U. S. 463, 476. While the Fisher case evoked dissent, it was a decision rendered after the Court of Appeals had fully declared its views of the law, and none of the considerations that moved the dissenters in that case is even remotely present in the case now before us.

We must therefore remand the case to the Court of Appeals with instructions to decide, in the first instance, what rule should prevail in the District of Columbia. To do otherwise would constitute an unwarranted departure from a wise rule of practice in our consideration of cases coming here from the Court of Appeals of the District. “There are cogent reasons why this Court should not undertake to decide questions of local law without the aid of some expression of the views of judges of the local courts who are familiar with the intricacies and trends of local law and practice. We do not ordinarily decide such questions without that aid where they may conveniently be decided in the first instance by the court whose special function it is to resolve questions of the local law of the jurisdiction over which it presides. Huddleston v. Dwyer, 322 U. S. 232, 237, and cases cited. Only in exceptional cases will this Court review a determination of such a question by the Court of Appeals for the District.” Busby v. Electric Utilities Employees Union, 323 U. S. 72, 74-75.

[For dissenting opinion of Mr. Justice Murphy, see post, p. 721.]

[figure]

APPENDIX TO OPINION OF THE COURT.

Summary of Disposition of Petitions for Certiorari to the Court of Appeals for the District of Columbia to Review Death Sentences on Conviction for First-degree Murder since 1938.

Mr. Justice Murphy,

dissenting.

Baxter Griffin has been sentenced to die for the murder of Lee Hunter. His justification for the killing was self-defense. He has found that Hunter had an open knife in his pocket when he was shot. He seeks a new trial on the basis of that newly-discovered evidence. The first question is whether that evidence would be admissible at a new trial.

It is clear to me that it is admissible. Uncommunicated threats and designs on the defendant cannot show his motive in killing, but they may demonstrate that a design on the defendant did in fact exist. This is the rule in “virtually all Courts.” 1 Wigmore, Evidence (3d ed., 1940), § 111, p. 547. It is certainly the federal rule. Wiggins v. Utah, 93 U. S. 465; Trapp v. New Mexico, 225 F. 968. And it is a thoroughly desirable rule. A defendant should be entitled to present the jury with evidence lending credence to his theory of the case. Griffin’s case is a good example of the policy behind the rule: for the open knife is the only supporting evidence of his self-defense testimony.

There can be little question that the open knife is an element in the proof of a design on the defendant, and is admissible under the rule stated above. But some courts have made exceptions to this rule, three of which might be considered relevant in this case. Wigmore, supra, § 111 (3). The exceptions have a central foundation: distrust of the jury’s ability to evaluate this kind of evidence. Many rules of exclusion are bottomed on this distrust, of course. But it is clearly misplaced when directed at the jury’s capability in weighing the value of uncommunicated threats in a murder trial. The evidence is simple; it is not calculated to inflame; it is far more difficult to fabricate than are communicated threats; the prosecution can easily question its importance; and

it provides solid support for a defendant’s self-defense theory. While in Griffin’s case the evidence is stronger for the prosecution than it was in Wiggins’, supra, that difference is not a distinction. The very plea of self-defense raises doubt on that question. Defendant’s testimony, supporting his plea, raises further doubt.

It is clear that this evidence might change the jury’s verdict. To make admissibility depend upon mechanical and often illogical variations in the size of the doubt in a judge’s mind is an invasion of the jury’s function. “It is pertinent here to remark, that both the effect of [the witnesses’] testimony and [their] credibility were to be weighed by the jury.” Wiggins v. Utah, supra, at 469.

The Court makes little attempt to justify the exclusion of this evidence. Instead, it cites Fisher v. United States, 328 U. S. 463. The Fisher case declined to upset an evidence rule that had “long been the law of the District of Columbia”: that “mental deficiency which does not show legal irresponsibility” is not “a relevant factor in determining whether an accused is guilty of murder in the first or second degree.” The Court stated the general rule that “matters relating to law enforcement in the District are entrusted to the courts of the District” in a case in which a reversal would have been a “radical departure from common law concepts” and thus “more properly a subject for the exercise of legislative power or at least for the discretion of the courts of the District.” 328 U. S. at 471, 473, 476.

In Fisher, the Court considered the judiciary’s case-by-case method ill suited for the sweeping changes which were and are necessary in the law of insanity. It recognized that an indirect attack on the problem, by admitting evidence of one’s past life as relevant in premeditation, might lead to the trial of one’s whole life rather than of the specific offense charged.

Despite the radical nature of the change, three members of this Court thought that the judiciary should make an attempt to correct the injustice of the common-law rules. Those arguments were rejected. But they were rejected only upon the limited basis to which I have referred.

Today the Court extends the Fisher rule. It calls Fisher a holding that no District of Columbia rules of evidence are reviewable in this Court. The Fisher case is no authority for such a proposition. There is no warrant for it in statute. And our denial of thirteen petitions for certiorari in death cases in the District in the last ten years cannot establish such a proposition. In the last ten complete Terms of Court, only 5.1% of all petitions for certiorari in forma pauperis have been granted. And the percentage of petitions for certiorari, other than in forma pauperis, granted in the same period has fluctuated between 14.9 and 22.7.1 When we deny nineteen out of twenty petitions in forma pauperis, and four out of five of the other petitions, the denial of petitions in thirteen capital cases in ten years reflects no greater policy in those cases than it does in any other class of cases. This is particularly true when the sample — fifteen cases — is so small compared to the number of cases we are asked to review, and when the sample considers only murder cases. “Nothing is so fallacious as facts, except figures.” For figures which do not reveal the peculiar facts of each case cannot reflect a policy of any kind.

Self-limitation of our appellate powers may be a worthy thing, but it is not attractive to me when the behest of Congress is otherwise. Congress has given this Court

the ultimate power to review District of Columbia trials. No matter how the decision is phrased, the Court’s power in the premises is such that it is responsible for the evidence rule it asks the Court of Appeals to expound. There is no “radical departure from common law” rules in Griffin’s case, as there was in Fisher’s. We should declare the evidence admissible.

If the evidence is admissible, a motion for a new trial should be granted. A contrary determination would be an abuse of discretion,2 for there is manifestly a reasonable possibility3 that the jury would lessen the verdict of first-degree murder.

The Chief Justice, Mr. Justice Douglas, and Mr. Justice Rutledge join in this opinion.