UNITED STATES COURT OF APPEALS FOR THE DISTRICT OF COLUMBIA CIRCUIT
November 15, 1966
NATHANIEL CLIFTON, APPELLANT
UNITED STATES OF AMERICA, APPELLEE 1966.CDC.204 DATE DECIDED: NOVEMBER 15, 1966
THIS APPEAL ARISES OUT OF THE HOLDING OF THE SUPREME COURT IN JACKSON
DENNO, 378 U.S. 368, 84 S. CT. 1774, 12 L. ED. 2D 908 (1964).
UNITED STATES COURT OF APPEALS FOR THE DISTRICT OF COLUMBIA CIRCUIT
Certiorari Denied, April 10, 1967, 386 U.S. 995, 87 S. Ct. 1312, 18 L. Ed. 2d 341.
Bastian, Senior Circuit Judge, and Burger and Leventhal, Circuit Judges. Leventhal, Circuit Judge (concurring in result).
DECISION OF THE COURT DELIVERED BY THE HONORABLE JUDGE BURGER
More than 20 years ago Appellant was tried and convicted on a charge of robbery. *fn1 No appeal was taken. In 1964 Appellant first attacked his 1946 conviction by a petition in the District Court, Clifton v. United States, Civil No. 2316-64, which was treated by the court as a petition for a writ of coram nobis. *fn2 That petition was denied. Following Jackson v. Denno (supra) a second petition was filed in 1965. In this second petition from which the instant appeal is taken Appellant rests on the single ground that the District Court in the 1946 trial had not afforded him an independent hearing such as is now required to pass on the voluntariness of the confession which was admitted in evidence against him. *fn3
Search for the stenographic verbatim record of the 1946 trial was unavailing but the District Judge who presided at that trial heard Appellant's petition and relied on detailed personal longhand notes of the trial which he had preserved. On the basis of these notes the District Judge made findings that he had in fact conducted an independent hearing on the issue of voluntariness of the confession and "found on the basis of the evidence that the confession was voluntary . . . and admitted the confession." Accordingly, the District Judge denied the contemporaneous petition, United States v. Clifton, 239 F. Supp. 49, 51 (D.D.C. 1965). *fn4
The question to be resolved is whether the procedures applied in the 1946 trial and the instructions to the jury were "fully adequate to insure a reliable and clear-cut determination of the voluntariness of the confession . . .," Jackson v. Denno (supra) 378 U.S. at 391, 84 S. Ct. at 1788.
The Government's position is that the District Judge who tried the case in 1946 made findings on the basis of his 1946 longhand notes, that he conducted an independent hearing on voluntariness out of the presence of the jury before deciding to submit that issue to the jury, and that this procedure, generally conforming to the so-called Massachusetts procedure approved in Jackson, fully satisfied the requirements of Jackson.
Appellant contends that the Court in Jackson expressed doubts as to the procedure in this jurisdiction. *fn5 He thus argues that "by stating that he [the District Judge] followed the 'traditional practice' in this district, the court . . . has virtually conceded that the procedures utilized in determining the voluntariness of appellant's confession do not meet the constitutional criteria laid down in Jackson." We do not agree. Doubts as to past practice do not necessarily mean that the requirements of Jackson were not met in Appellant's case. See Mitchell v. Stephens, 353 F.2d 129, 142 (8th Cir. 1965).
The critical issue raised by Appellant is that "the decision in Jackson v. Denno . . . requires that the trial judge, in addition to holding a preliminary hearing out of the presence of the jury, make a specific finding that the confession was voluntary beyond a reasonable doubt." Appellant also contends that the District Court did not instruct the jury that it was required to find the confession voluntary beyond a reasonable doubt in order to rely on it. The fact is that the Supreme Court did not pass on either of these questions in Jackson.
Jackson v. Denno requires only that the procedures applied in these circumstances must be "fully adequate to insure a reliable and clear-cut determination of the voluntariness of the confession," 378 U.S. at 391, 84 S. Ct. at 1788; it made no attempt to define the standard to be applied by the trial judge in his preliminary determination of voluntariness, *fn6 and the Court expressly disclaimed any ruling on jury instructions concerning the consideration of voluntariness, 378 U.S. at 375 n. 5, 84 S. Ct. at 1779.
In his charge to the jury in the 1946 trial, the District Judge did not specifically single out the confession, apart from the other evidence, and define the standard applicable to it. We note, however, that the confession was in fact the essence of the Government's case. The Judge instructed the jury that the Government had the burden of proving beyond a reasonable doubt that the defendant was guilty and had committed every element of the offense charged. It does not seem possible to us that the jury would have failed to understand that in order to convict they must believe beyond a reasonable doubt both its voluntariness and its substance, i.e., the truth of the facts recited in the confession.
The District Judge, however, did not in 1946 or in the hearing from which this appeal is taken, spell out by what standard he evaluated the confession before submitting it to the jury. We are not unmindful that the Fourth Circuit has held that the District Judge must find the confession voluntary beyond a reasonable doubt when making his preliminary determination. United States v. Inman, 352 F.2d 954, 956 (4th Cir. 1965). *fn7 The opinion of that court contains no extended discussion or treatment of the vexing problem left unresolved by Jackson. The determination of whether a confession is voluntary is, in substance, a ruling on its admissibility as evidence. State ex rel. Goodchild v. Burke, 27 Wis.2d 244, 249, 133 N.W.2d 753, 762 (1965), cert. denied, 384 U.S. 1017, 86 S. Ct. 1941, 16 L. Ed. 2d 1039 (1966); State v. Keller, 240 Or. 442, 448, 402 P.2d 521, 525 (1965). Nowhere in the law -- civil or criminal -- has it ever been thought that a trial judge, in passing on admissibility of evidence in a jury case, must use the reasonable doubt standard. *fn8 The Supreme Court in Jackson pointedly failed to make any comment on this crucial issue, as Mr. Justice Black sharply noted in his dissenting opinion, 378 U.S. at 404-405, 84 S. Ct. at 1795-1796.
It is one thing to call for this high standard of proof from the ultimate fact finders and quite another to ask that this issue be resolved preliminarily by the judge beyond a reasonable doubt contrary to all the law governing admissibility of evidence. We think Mr. Justice Black was correct in suggesting that to vest this essentially fact finding function in the judge would constitute a "downgrading of trial by jury," 378 U.S. at 405, 84 S. Ct. at 1796 (dissenting opinion). We are not persuaded that it is more logical or reasonable that a confession, alone among all the myriad of evidentiary material, be singled out for a unique standard of appraisal. It would be difficult to distinguish, so far as jury impact is concerned, the weight of a confession from other forms of damaging utterance -- inconsistent statements and admissions, for example, heard by persons other than the police. Cf. Ashcraft v. State of Tennessee, 327 U.S. 274, 66 S. Ct. 544, 90 L. Ed. 667 (1946). *fn9 Under the Massachusetts procedure we see no basis for a special and unique "admissibility" standard which places the judge in the temporary role of a "juror," denigrating the historic fact finding role of jurors.
The concurring opinion views Bram v. United States, 168 U.S. 532, 18 S. Ct. 183, 42 L. Ed. 568 (1897), as the source of the strict "reasonable doubt" standard for trial judges in the preliminary determination of voluntariness required under Jackson for the so-called Massachusetts procedure. The pertinent language in Bram is as follows:
The rule is . . . that, in order to render a statement admissible, the proof must be . . . sufficient to establish that the making of the statement was voluntary . . ., 168 U.S. at 549, 18 S. Ct. at 189.
Any doubt as to whether the confession was voluntary must be determined in favor of the accused . . ., 168 U.S. at 565, 18 S. Ct. at 195.
Judge Leventhal concedes that the most that can be said for the "any doubt" language of Bram is that it was a "quantum of proof" reference, used in a quantitative sense. Bram does not establish a standard which marks off on a qualitative scale the degree of certainty which a trial judge must employ before letting the jury evaluate both voluntariness and accuracy of a confession. *fn10
The only alternative interpretation of Bram is to say that "any doubt" as used there means that the slightest scintilla of qualitative doubt regarding voluntariness would require the trial judge to ban the confession from any jury determination of voluntariness under its obviously lesser "reasonable doubt" standard. This would go far beyond imposing a mere "reasonable doubt" standard on trial judges in their preliminary determination of voluntariness. Thus, if we are to read Bram literally we must conclude that it calls for an absolute "any doubt" qualitative standard, and no reason appears why this should be reduced arbitrarily to a "reasonable doubt" standard. The language in Bram is simply "any doubt" not "any reasonable doubt." This analysis seems clearly apparent from the very wording of the questioned sentence in the Bram opinion. That Court was simply saying that doubt must be determined in favor of the accused regarding voluntariness of confessions. *fn11 It was not addressing itself to framing a standard.
Additionally some mention should be made of a troublesome contradiction in the position that it is from Bram that we draw the "reasonable doubt" standard to be utilized by the trial judge in his preliminary determination but that Jackson's requirement of voluntariness hearings by the trial judge need not be retroactively applied to govern cases where some pre-admission voluntariness hearing was in fact held. It is reasoned that the required standard, "beyond a reasonable doubt," is an old one, established by Bram ; but its expression is considered to be new since Jackson, with respect to whether this standard should have retroactive application. Under the concurring interpretation, Jackson is retroactive only in its requirement for a preliminary determination of voluntariness by the trial judge. Jackson, it is reasoned, need not be retroactively applied when some judicial screening did take place, although imposing a less restricted standard of doubtfulness. Yet it is this very standard which Judge Leventhal claims was established long ago by Bram. For these reasons we are unable to accept the concurring interpretation of Bram and Jackson with respect to the issue now before us.
We repeat again that Jackson does no more than call for a procedure that will give a "reliable determination on the voluntariness issue . . .," 378 U.S. at 387, 84 S. Ct. at 1786; we reject the argument that a United States District Judge cannot make that "reliable determination" on conventional admissibility criteria, leaving any resolution of reasonable doubt as to that evidence in its traditional province -- the combined judgment of twelve jurors. We hold that the confession should not be admitted in evidence unless the trial judge makes a preliminary determination and an express finding that on all the evidence he is satisfied that the confession was voluntarily made. *fn12 We note that Judge Leventhal's opinion sees little difference between a standard that the District Judge be "satisfied" of voluntariness and that he be persuaded of that conclusion beyond a reasonable doubt. If the determination of the District Judge is to submit the confession to the jury, however, he should not indicate that he has made a preliminary decision that it was voluntarily made, but he should specifically instruct that they are not to give any weight to the confession unless they, as ultimate fact finders, are satisfied beyond a reasonable doubt on all the evidence that it was voluntarily given by the accused.
That a coordinate tribunal has reached a different conclusion, United States v. Inman (supra) has led us to ponder carefully on this problem. We are presented with one choice which seems to us to maintain the role of lay jurors in their historically high place and another which limits them to resolving only those fact issues which can be "safely" entrusted to their collective judgment. For a generation judicial trends have been to exclude from the jury all evidence thought to have been unfairly or improperly secured, and more recent holdings place stringent limits on the use of any utterance by an accused. In the future trial judges will be evaluating only those utterances of an accused which have already passed through the whole gamut of screening processes outlined in McNabb, Mallory, Escobedo, Massiah, and Miranda. The prospects now are that trial judges, otherwise much overburdened, will not be overworked in passing on the voluntariness of the few confessions which will survive the application of these cases. At the very least, it seems to us to be reasonable to continue functioning without departing from the traditional rule reaffirmed by this holding. The future experience of this and other circuits may indicate a different rule is desirable. Perhaps this is why the Supreme Court left these issues unresolved in the face of Mr. Justice Black's astringent comments -- so that the ultimate solution could be based on experience rather than on prophecy.
The order of the District Court is,
Affirmed. IN AGREEMENT
LEVENTHAL, Circuit Judge (concurring in result):
I have a different view from that of my brethren as to the rule appropriate in implementing Jackson v. Denno, 378 U.S. 368, 84 S. Ct. 1774, 12 L. Ed. 2d 908 (1964), concerning the standard that properly governs the trial judge when making a determination as to the voluntariness of defendant's confession before permitting its presentation in evidence in a criminal case tried to a jury.
1. As to all cases tried subsequent to Jackson v. Denno, the standards and procedures that are applicable, in my view, are those stated in United States v. Inman, 352 F.2d 954, 956 (4th Cir. 1965), where Judge Bryan succinctly outlined the procedure prescribed for the district courts in our neighboring circuit:
To assure the accused complete protection, the procedure should substantially be this. On proffer of the confession, even though there be no objection, the court should let the jury withdraw, and then take evidence upon the confession and its factual setting. On this voir dire the defendant may testify without prejudice to his privilege not to take the stand before the jury, but he may be examined or cross-examined only with regard to the origin and character of the confession, not upon his innocence or guilt. The court will thereupon independently determine whether the confession is admissible.
In this determination the District Judge will evaluate the evidence to ascertain whether, after resolving any conflicts therein, it convinces him beyond a reasonable doubt that the confession was voluntary. Unless the judge is so persuaded, the confession may not be admitted. If, however, he is satisfied beyond a reasonable doubt of its voluntariness, he should explicitly make and include in the record a finding of that fact. (Emphasis added.)
Following the voir dire hearing, the trial will proceed with or without the confession, according to whether it was received or rejected. If admitted, the court should instruct the jury, whether requested or not, upon the law governing the use of a confession. Stevenson v. Boles, 331 F.2d 939 (4 Cir. 1964), aff'd per curiam, 379 U.S. 43, 85 S. Ct. 174, 13 L. Ed. 2d 109. Included would be a forthright caution that before giving any weight to the confession, the jury must be satisfied beyond a reasonable doubt that it had been made by the accused uninfluenced by promise of reward, threat of injury or diminution of his rights.
Applying a reasonable doubt standard would not be a strange or exotic task for a judge. Every time he tries a criminal case without a jury he must find all the elements of a crime established beyond a reasonable doubt. The determination by a judge of the voluntariness of a confession is so significant in the final assessment of guilt or innocence, and relates to such basic values in our system of jurisprudence, that it must be governed by the same standards as apply to the elements of the crime, and is not to be relegated to a lower plane on the ground that it is a mere ruling on a point of evidence.
We are not dealing with an ordinary ruling on evidence. The exclusion of confessions not shown to be voluntary inheres in the constitutional privilege against self-incrimination. Bram v. United States, 168 U.S. 532, 542, 18 S. Ct. 183, 42 L. Ed. 568 (1897); Malloy v. Hogan, 378 U.S. 1, 6-7, 84 S. Ct. 1489, 12 L. Ed. 2d 653 (1964); Miranda v. State of Arizona, 384 U.S. 436, 461, 86 S. Ct. 1602, 16 L. Ed. 2d 694 (1966). The historical root is the maxim nemo tenetur seipsum accusare, "no one shall be held to accuse himself." "[This] maxim, which in England was a mere rule of evidence, became clothed in this country with the impregnability of constitutional enactment." Brown v. Walker, 161 U.S. 591, 596-597, 16 S. Ct. 644, 647, 40 L. Ed. 819 (1896), quoted in Miranda (supra) 384 U.S. at 442-443, 86 S. Ct. 1602.
In Bram Justice White, later the Chief Justice, explored the "quantum of proof" necessary to show the voluntariness of a confession. The current vitality of his approach is attested by the recent and extensive reliance on his opinion.1 He expressly declared that in the court's determination whether to admit a confession "any doubt as to whether the confession was voluntary must be determined in favor of the accused." 168 U.S. at 565, 18 S. Ct. at 195. In Harrold v. Territory of Oklahoma, 169 F. 47, 53-54 (8th Cir. 1909), Judge Sanborn cited Bram and spelled out its requirement:
The burden was upon the prosecutor to prove to the court that the confession was voluntary, that it was not influenced by compulsion, hope, fear, or other inducement of any sort, and, if the evidence failed to establish that fact beyond a reasonable doubt, it was the duty of the court to reject the confession.
Other decisions that applied the same rule are cited in the footnote.2 This aspect of Bram was not applied in the District of Columbia, where "voluntariness" was for determination by the jury under the reasonable doubt standard, and local "law and practice" contemplated submission of the confession to the jury without a finding of voluntariness by the court. Wright v. United States, 102 U.S.App.D.C. 36, 45, 250 F.2d 4, 13 (1957). Prior to Jackson v. Denno there was no ruling that the Constitution required a judicial determination of voluntariness.
That the court making the determination of voluntariness required of the court by Jackson v. Denno should be governed by the standard of being convinced beyond a reasonable doubt, and not by the analytically lower standard of proof to the "satisfaction of the court" which governs findings of fact involved in ordinary rulings on evidence, is attributable to the distinctive nature of the voluntariness determination. It has the deepest roots in our Constitution and system of jurisprudence. And it relates to a matter which is usually the key item in the proof of guilt, and certainly one of overpowering weight with the jury. The very introduction of an involuntary confession is a denial of constitutional rights so prejudicial as to vitiate the conviction irrespective of the quantum of other untainted evidence demonstrating the guilt of the accused.3
Plainly the jury must disregard a confession unless they find it voluntary beyond a reasonable doubt. That was recognized by both federal and state courts in both the pre- Jackson era4 and in the post- Jackson cases wherein this issue was re-submitted to the jury.5
In Jackson the Supreme Court held that procedures in the trial court must "be fully adequate to insure a reliable and clear-cut determination of the voluntariness of the confession, including the resolution of disputed facts upon which the voluntariness issue may depend." 378 U.S. at 391, 84 S. Ct. at 1788. So important is the need for reliability that the basic and primary determination must be made by the court. This is not to derogate from the jury as such, but to recognize the prejudice inhering in the admixture of a determination of voluntariness together with the jury's inescapable consideration of reliability of the confession, and indeed ultimate guilt or innocence.
Certainly the need for a finding beyond a reasonable doubt must govern the judge operating under the orthodox, or Wigmore, rule, where the issue of voluntariness is not presented as such to the jury. And Jackson makes clear that the Supreme Court's approval of the Massachusetts rule rests on the premise that it provides as much protection as the orthodox rule in regard to the screening by the judge, which remains the basic determination safeguarding the accused even though the voluntariness is re-submitted to the jury. See 378 U.S. at 378 n. 8, 84 S. Ct. at 1781.
It is true that in Jackson v. Denno the Supreme Court did not expressly address itself to the question before us. But to me a reasonable doubt standard for the judge's determination of voluntariness ensues from the Court's reasoning as surely as Euclid's corollaries unfolded from his theorems. The decisive ruling in Jackson v. Denno is that it is the determination of voluntariness by the judge which is crucial to our constitutional liberties. Any redetermination by the jury may be "ultimate" in time but not in constitutional significance.
Practical considerations of proof underscore the importance of the court's use of the reasonable-doubt standard, notwithstanding a later submission to the jury. There may be more evidence before the court than before the jury, for example the testimony of a defendant who dare not risk presentation of prior convictions to the jury. It would be unjust if such evidence raised a reasonable doubt only to find that this doubt was ignored by the judge, under the rule proclaimed by the majority, and was never available to the jury.
Finally attention should be called to the weight of the post- Jackson rulings that the judge must determine voluntariness of the confession beyond a reasonable doubt. States that applied this standard prior to Jackson have continued it.6 Rulings to this effect appear not only in jurisdictions applying the orthodox rule,7 but also in jurisdictions, like New York, which resubmit the issue of voluntariness to the jury.8 This standard has been adopted even where the court expressly took into account that it was dealing with a rule of evidence.9
The view I hold is the result reached not only by the Fourth Circuit but also, so far as I am aware, by every other court that has analyzed Jackson in its opinion. And it is not without significance that the majority have not adduced any post- Jackson ruling discarding the reasonable doubt standard for the trial judge which they can champion as authoritative.10
2. Assuming, however, that the reasonable doubt standard for the judge is not a constitutional prerequisite in a jurisdiction where the issue is resubmitted to the jury, the reasonable doubt standard for the judge should be enunciated in the exercise of our responsibility to exercise supervisory powers over the administration of justice in the District of Columbia. Cf. Tate v. United States, 123 U.S.App.D.C. 261, 359 F.2d 245, 252 (1966), and cases cited. I see no reason for our rejection of a rule approved by the Supreme Court in Bram and recently adopted by both federal and state courts (see notes 7-9 supra) as an incident of Jackson v. Denno.
3. However, the requirement that a reasonable doubt standard govern the judge's determination of voluntariness should not, in my view, be applied retroactively to a conviction which became final some sixteen years before Jackson was decided.
Jackson v. Denno is plainly retroactive in the sense that it applies to cases even though judgment has already become final, so as to require a judicial determination of voluntariness when there was no judicial screening of the issue prior to its presentation to the jury in a setting likely to prejudice a defendant.11 This does not govern the question whether similar retroactivity is requisite in a case where there was a preliminary judicial determination but it is contended that although the judge was satisfied of voluntariness he was not convinced beyond a reasonable doubt.
There are substantial considerations for withholding retroactivity as to this aspect of Jackson v. Denno. The fact that the judge was independently "satisfied" of voluntariness removes the most serious prejudice, the prejudice that appears in the case where "some evidence" was enough to bring the confession into the juryroom. There is only a relatively narrow zone of cases in which a judge in a criminal matter will be "satisfied" of voluntariness though he harbors a reasonable doubt.
On the other side of the coin lie considerations of judicial administration. Whether consideration is to be by the same or another judge, it becomes extraordinarily difficult of ascertainment whether a judge who was convinced of voluntariness was or was not convinced beyond a reasonable doubt. The extra margin of convincement assured by the reasonable-doubt standard is an elusive matter difficult to define. Compare McGill v. United States, 121 U.S.App.D.C. 179, 185, 348 F.2d 791, 797 (1965). Applying this standard is difficult enough at trial. The difficulty is enormously increased if the trial has long since been concluded and there is injected in the equation such imponderables as the lack of demeanor evidence of witnesses not now before the court, failure of recollection with the passage of time, inability to reconstruct the atmosphere of the trial. And, of course, a judge at a pre- Jackson trial would have had no occasion to make a particular note of the depth of his conclusion.
In considering whether this aspect of Jackson v. Denno should be applied retroactively, we cannot ignore the fact of the course of decisions in the District of Columbia which did not follow the Bram forebear. Cf. Linkletter v. Walker, supra note 11, at 636 of 381 U.S., at 1741 of 85 S. Ct. It does not matter whether this was right or wrong under the non-constitutional doctrines then deemed applicable; whether it was due to the fact that Bram was overlooked, or was then regarded as a poorly considered opinion, whether perhaps the courts felt free to fashion a different common law rule for the District of Columbia treated as a state than the rule applicable in other federal courts. The fact remains that Bram's reasonable doubt standard for judges was dormant at best and was not revitalized until Jackson v. Denno.
As the Supreme Court's decisions make clear, the retroactivity issue is not a matter of absolutes. A balance must be struck in terms of the degree of unfairness of what was done and the practical quagmires of correction. As the Court noted in Johnson v. State of New Jersey, supra note 11, at 729, of 384 U.S., 86 S. Ct. 1772, it is essentially concerned with a "matter of degree," a "question of probabilities."
In my view the overall interest of justice is achieved by ruling that this aspect of Jackson v. Denno -- that the judge must be convinced of voluntariness beyond a reasonable doubt even where the issue is also submitted to the jury -- is requisite for subsequent trials, but need not be accorded retrospective application.12