Donald Wolff, Ir. Baris, St. Louis, Mo., Philip Resnick, Mark W. Peterson, Minneapolis, Minn., and Philip Miller, Des Moines, Iowa, for appellants.

Sam Rosenthal, Washington, D.C., for appellee.

Before LAY, Chief Judge, and HEANEY, BRIGHT, ROSS, McMILLIAN, ARNOLD, JOHN R. GIBSON, FAGG, and BOWMAN, Circuit Judges, en banc.

ARNOLD, Circuit Judge, with whom LAY, Chief Judge, and HEANEY, BRIGHT and ROSS, Circuit Judges, join.

On October 12, 1984, the Comprehensive Crime Control Act of 1984, Title II of Pub.L. No. 98-473, 98 Stat.1976, became law. Chapter I of this Act is known as the Bail Reform Act of 1984, and Section 203(a) of this chapter, 98 Stat.1976, 1981-82, enacts new standards for the admission to bail of convicted persons pending their direct appeal. The new provision, to be codified as 18 U.S.C. Sec. 3143(b), provides as follows:

(b) Release or Detention Pending Appeal by the Defendant.--

The judicial officer [usually a district judge, a circuit judge, or a court of appeals] shall order that a person who has been found guilty of an offense and sentenced to a term of imprisonment, and who has filed an appeal or a petition for writ of certiorari, be detained, unless the judicial officer finds--

(1) by clear and convincing evidence that the person is not likely to flee or pose a danger to the safety of any person or the community if released pursuant to section 3142(b) or (c); and

(2) that the appeal is not for purpose of delay and raises a substantial question of law or fact likely to result in reversal or an order for a new trial.

If the judicial officer makes such findings, he shall order the release of the person in accordance with the provisions of section 3142(b) or (c).

In each of the appeals now before us, the District Court found that the defendants were not likely to flee or pose a danger to the safety of any person or the community. Bail pending appeal was nevertheless denied, the Court being convinced that the requirements set forth in paragraph (2), quoted above, had not been met. In each case, the District Court held that the appeal did not raise "a substantial question of law or fact likely to result in reversal or an order for a new trial." Our first task is to interpret this phrase and to describe, as helpfully as possible, how it is to be applied. Next, we must consider whether the statute, as so interpreted, is constitutional. And finally, we shall explain how this standard has been applied in the individual cases before us.

The United States takes the position that the portion of the statute in question requires two separate determinations: (1) whether the appeal raises a substantial question, and (2) whether, if the defendant prevails on this question, reversal or an order for a new trial is likely. The government further argues that a question is "substantial" for this purpose if the defendant's argument on the question has a substantial chance or a substantial likelihood of prevailing on appeal. This requirement, according to the government, means that an argument must be more than simply nonfrivolous, but need not be so compelling as to require the conclusion that it is more likely that the defendant will win the argument than lose it. Rather, the government says, an argument is "substantial" for this purpose if the question is a close one or one that could very well go either way.

If a question presented by an appellant passes this part of the test, the government continues, it should then be asked whether, assuming the question is decided in favor of the defendant, it is more probable than not that reversal of the conviction or a new trial will be required. To make this determination, the Court must assume that the defendant's argument will prevail on appeal and assess the impact of the assumed error on the conviction in view of the entire record. If, for example, the strength of the prosecution's case makes clear that the assumed error had no effect on any substantial right of the defendant, or if the assumed error, even though not harmless, would affect fewer than all the counts on which defendant has been sentenced to imprisonment, release pending appeal would not be appropriate, despite the existence of a "substantial question."We hold that this interpretation of the statute is correct and that the new law, as so construed, is constitutional.

We have the benefit of opinions on this question from three other circuits, the Third, the Eleventh, and the Ninth.

United States v. Miller, 753 F.2d 19 (3d Cir.1985), was the first appellate opinion filed on the subject. The Court held that two separate showings must be made by a defendant to satisfy the language in question: first, that the question of law or fact presented by the appeal is "substantial; " and second, that if the defendant prevails on that question, either a reversal or an order for a new trial, of all counts on which imprisonment has been imposed, is likely. By "substantial," the Court explained, it meant a "significant question at issue ... which is either novel, which has not been decided by controlling precedent, or which is fairly doubtful." 753 F.2d at 23. The second part of the statutory standard, the one referring to likelihood of reversal, the Court read as requiring that the question presented, assuming its decision in appellant's favor, be "important to the merits," ibid., or "so integral to the merits of the conviction on which defendant is to be imprisoned that a contrary appellate holding is likely to require reversal of the conviction or a new trial," ibid. As examples of questions that would not qualify under this second element of the standard, the Court mentioned errors that are "harmless, ... have no prejudicial effect, or ... have been insufficiently preserved." Ibid. Miller thus requires appealing defendants (the burden rests upon them to establish their right to bail) to make two kinds of showings relevant in the present context: that the question is substantial, and that reversal or a new trial is likely if it goes in defendant's favor.

The next case decided by a court of appeals on this issue was United States v. Giancola, 754 F.2d 898 (11th Cir.1985) (per curiam). The Eleventh Circuit, referring to "the thoughtful analysis of ... Judge Sloviter" for the Third Circuit in Miller, 754 F.2d at 900, adopted the same basic two-part approach, subject to one important qualification with respect to the first part of the analysis. It defined a "substantial question" not simply as one on which there is no controlling precedent, but as "a 'close' question or one that very well could be decided the other way." 754 F.2d at 901. Such a question, the Court said, "is one of more substance than would be necessary to a finding that it was not frivolous." Ibid. "[A]n issue could well be insubstantial even though one could not point to controlling precedent.... [T]here might be no precedent in this circuit, but there may also be no real reason to believe that this circuit would depart from unanimous resolution of the issue by other circuits." Ibid.

The last appellate opinion (as of this writing) on the subject is United States v. Handy, 753 F.2d 1487 (9th Cir.1985) (per curiam). Like Giancola, Handy followed the two-part analytical framework first set out in Miller. The statute, Handy said,

should be interpreted to read that "substantial" defines the level of merit of the question presented and "likely to result in reversal or an order for a new trial" defines the type of question that must be presented.

753 F.2d at 1489 (emphasis in original). The Handy Court appeared to adopt Miller so far as the second part of the analysis is concerned. As for the meaning of "substantial," it offered the phrase " 'fairly debatable,' " 753 F.2d at 1490, as a definition.

These opinions make our task considerably easier. We need not repeat much of the history and analysis contained in them. Like the Ninth and Eleventh Circuits in Handy and Giancola, we are indebted to the Third Circuit's pathfinding effort in Miller, and we also adopt the basic two-part approach to the statute it put forward. We do so with certain observations and qualifications of our own, however. First, as to whether a question is "substantial," we choose to follow Giancola ("a 'close' question or one that very well could be decided the other way") rather than Miller ("novel," "not ... decided by controlling precedent," or "fairly doubtful") or Handy (" 'fairly debatable' "). We believe Giancola is more responsive to the announced purpose of Congress, which was, bluntly, that fewer convicted persons remain at large while pursuing their appeals.

Under prior law, release on bail pending appeal was the normal practice. It was the rule, not the exception, and there was a presumption in favor of release. So far as presently pertinent, the Bail Reform Act of 1966, Sec. 3(a), 80 Stat. 214, 215-16 (formerly codified as 18 U.S.C. Sec. 3148), required that bail be granted unless "it appears that an appeal is frivolous or taken for delay." Congress passed the new law because it was unhappy with the old one. "The change ... requires an affirmative finding that the chance for reversal is substantial. This gives recognition to the basic principle that a conviction is presumed to be correct." S.Rep. No. 98-225, 98th Cong., 1st Sess. 27 (1983), U.S.Code Cong. & Admin.News 1984, pp. 3182, 3210. Further, under the prior law the government had to show, if it wanted bail denied, that the appeal was frivolous or taken for delay. (See advisory notes to former Rule 9(c) of the Federal Rules of Appellate Procedure.) Rule 9(c) of the Federal Rules of Appellate Procedure was amended by Section 210 of Pub.L. No. 98-473, 98 Stat. 1987, to conform with the new Section 3143(b). But under the new law, "the burden of showing the merit of the appeal should now rest with the defendant." S.Rep. No. 98-225, supra, at 27 n. 86. Senator Thurmond, Chairman of the Senate Committee on the Judiciary and manager on the Senate floor of the bail-reform bill that was later incorporated into the Comprehensive Crime Control Act of 1984, expressed the same thought when he said that a major purpose of the bill is to "reverse the presumption in favor of releasing individuals who have been found guilty.... Once someone has been found guilty, there is no longer a presumption of innocence...." 130 Cong.Rec. S938 (daily ed. Feb. 3, 1984). See also p. 386 of the President's Message to the Congress proposing what was then called the Comprehensive Crime Control Act of 1983, which is a relevant piece of legislative history because the language that became law came from this proposal by the Executive Branch.

The Handy and Miller formulations of what "substantial question" means would not work much of a change in prior law. The authorities Handy cites in favor of its "fairly debatable" definition--opinions of Circuit Justices on bail applications--all predate by many years the Bail Reform Act of 1984 and reflect the previous view, long entrenched in our law, that bail is the rule rather than the exception, to be granted whenever a nonfrivolous question is raised. And under Miller the term "substantial question" seems to be defined without any regard for the probability of success on appeal. We think the Giancola formulation is more faithful to the purpose of Congress. We doubt that Congress would have gone to the trouble of passing a new statute to obtain no more change than is brought about by either Miller or Handy. In short, a judge considering the question of bail pending appeal need not hold (as one reading of the statute, discussed below, would require) that he or she has probably made a mistake. But bail can be granted only if the question is close, one that could go either way. The formulation is inexact--though probably less so than prior law--, but we think experienced judges and lawyers will find it reasonably easy to apply. Bail will be less frequent; it will be the exception, not the rule. That may not be wise policy. Certainly it is not what we are accustomed to. But it is the command of the sovereign, expressed by the people's elected representatives, and we are obliged to receive and apply it hospitably.

We turn next to the task of elaborating the meaning of the second part of the standard that a defendant must meet--"likely to result in reversal or an order for a new trial." The Third Circuit in Miller gave some examples of arguments that would not meet this standard--for example, a harmless error, or a question as to which the defendant had not sufficiently preserved his record. We agree completely with Miller as far as it goes in this regard, but we believe an additional comment is in order. The Miller approach tells us some kinds of errors do not qualify as "likely to result in reversal," but it does not tell us what sorts of errors do qualify, nor does it lay down a comprehensive standard for judging whether a given question falls in one group or the other. The issue here is the meaning of the word "likely" in the statute. We believe the word should be read in its ordinary sense, as referring to something that is more likely to happen than not.

If one asks whether a horse is likely to win a race, and the answer is yes, the person who asked the question naturally understands that the chances of the horse's winning are greater than those of its losing. He would not ordinarily believe that a "yes" answer meant only that the horse had a greater than negligible (in the legal context, nonfrivolous) chance of winning. We hold, therefore, that in order to satisfy this part of the test for bail pending appeal, a defendant will have to show that the substantial question presented (assuming that part of the standard has already been met) will more probably than not, if decided in defendant's favor, lead to a reversal or an order for new trial on all counts on which imprisonment has been imposed. (An argument that would produce a reversal of fewer than all such counts would be insufficient in this context, because if one count imposing imprisonment survives, the reason for allowing bail pending appeal, that a defendant should not be imprisoned under a legally erroneous sentence, disappears.)

We shall not leave the subject before responding to a question that may be in the mind of the reader. The statute says, "substantial question of law or fact likely to result in reversal...." Why not interpret it in what may be the most natural and immediately obvious sense, as requiring the defendant to show that the question presented will more likely than not result in reversal? To this question there are several answers. In the first place, none of the parties before us urges this position, not even the government, though it once did embrace it, only to abandon it later. Of course we are bound to interpret the statute in accordance with Congress's intention, no matter what positions the parties take before us, and no concession, even of the United States, is binding on the courts, but still it would be awkward to adopt a posture more zealous than that of the prosecution. Second, such a position, if adopted, would put us squarely in conflict with each of the three circuits that have already ruled on the question, and the conflict would be of the most fundamental sort, not like the relatively minor differences in statutory approach that we have outlined above in this opinion. And finally, to require a defendant to show that the particular question urged is more likely to produce a reversal than an affirmance would, as the Miller opinion properly observes, make the word "substantial" in the statute redundant, for Congress could have accomplished such a result simply by requiring that a defendant show that he or she is raising a "question of law or fact likely to result in reversal." No doubt Congress sometimes repeats itself for the sake of emphasis, just as we all do, but in the interpretation of statutes it is usually the safer course at least to try to attribute a separate and additional meaning to each word and phrase in the law. When we do so here, we are driven to the conclusion that the phrase "likely to result in reversal" must mean something beyond the word "substantial." For these reasons, we reject what may be the simpler interpretation of the law and adopt the two-part analysis first used in Miller and elaborated in this opinion.

To sum up: We hold that a defendant who wishes to be released on bail after the imposition of a sentence including a term of imprisonment must first show that the question presented by the appeal is substantial, in the sense that it is a close question or one that could go either way. It is not sufficient to show simply that reasonable judges could differ (presumably every judge who writes a dissenting opinion is still "reasonable") or that the issue is fairly debatable or not frivolous. On the other hand, the defendant does not have to show that it is likely or probable that he or she will prevail on the issue on appeal. If this part of the test is satisfied, the defendant must then show that the substantial question he or she seeks to present is so integral to the merits of the conviction that it is more probable than not that reversal or a new trial will occur if the question is decided in the defendant's favor. In deciding whether this part of the burden has been satisfied, the court or judge to whom application for bail is made must assume that the substantial question presented will go the other way on appeal and then assess the impact of such assumed error on the conviction. This standard will, we think, carry out the manifest purpose of Congress to reduce substantially the numbers of convicted persons released on bail pending appeal, without eliminating such release entirely or limiting it to a negligible number of appellants.

Is the statute as so interpreted constitutional? The constitutional objection most strongly urged before us on these appeals is that the statute, if interpreted to require a defendant to show that the particular question presented will more probably produce a reversal than an affirmance, so unreasonably restricts the right of bail pending appeal as to deny liberty without due process of law, in violation of the Fifth Amendment, or to constitute "excessive bail," in violation of the Eighth Amendment, or both. Since we have not adopted this construction, this constitutional objection disappears. We entertain no doubt that the interpretation of the statute we have adopted is fully consistent both with the Due Process Clause of the Fifth Amendment and with the Bail Clause of the Eighth Amendment. Under our interpretation, judges considering bail applications, trial or appellate, will apply an objective standard and will weigh the probabilities according to this standard in a manner long familiar to the bench, for example in connection with motions for injunctions or stays pending appeal.

A couple of other constitutional arguments, nonfrivolous but insubstantial, deserve brief mention. It is suggested that the new statute, as applied to defendants whose alleged crimes were committed before its effective date, October 12, 1984, violates the Ex Post Facto Clause. The Third Circuit in Miller has already answered this argument sufficiently. Admission to bail pending appeal is, for Ex Post Facto Clause purposes, "procedural." It does not increase the punishment for a crime already committed, but simply regulates the time at which imprisonment for that crime will begin after conviction.1

We are also told that the entire Bail Reform Act of 1984, and, presumably, the entire Comprehensive Crime Control Act of 1984, is unconstitutional because it went to the President not as a bill but as a joint resolution. Joint resolutions, to be sure, are usually employed, the parliamentary manuals tell us, for subsidiary, inferior, or temporary kinds of legislation. Indeed, in this very case the Comprehensive Crime Control Act of 1984, including the bail-reform provisions, has become law by way of amendment to a joint resolution making "continuing appropriations" for the operation of the federal government, appropriations, that is, to enable certain departments and agencies to continue their work temporarily until the regular appropriations bills for those departments and agencies could be passed. Such continuing resolutions have become a regular feature of the congressional landscape towards the end of almost every fiscal year and, since, in theory at least, the government, or part of it, will come to a halt if the continuing resolution is not signed into law, they are considered "veto proof," or nearly so, and members of Congress with cherished projects of their own strive to attach them as amendments to the continuing-appropriations joint resolution, secure in the knowledge that the President cannot veto part of a measure. Appellants complain that this kind of legislation tends to be hasty and ill-advised, and they may be right, but that is none of our affair. The fact that the words at the top of the first page of a law are "a bill" instead of "a joint resolution" is of significance only for internal congressional purposes. A joint resolution, once signed by the President, is every bit as much of a law as a bill similarly signed. Our task is simply to hold the Congress within the limits of the power given it by the Constitution, not to pass judgment on matters of legislative practice.

In short, we hold that new Section 3143(b) of Title 18, as enacted by the Bail Reform Act of 1984, is constitutional.

We have already announced our rulings on the various applications for bail presented by these cases. We shall now briefly explain how these rulings follow from the standards set forth in this opinion.2

In No. 84-2430, defendants Geary Powell, James Barfield, and William Barfield were convicted of illegally transporting aliens in violation of 8 U.S.C. Sec. 1324(a)(2) (1982). In their application for bail pending appeal, defendants list three principal issues for reversal of their convictions: (1) that the evidence fails to support the verdict; (2) that the prosecutor's personal interpretations of testimony given in Spanish by a witness were improper; and (3) that the court failed to take sufficient measures to protect them against the prejudicial effect of pretrial publicity. We of course have not given these arguments plenary consideration. In the context of applications for bail pending appeal, we must carefully and objectively assess the chances of success on the questions presented, as well as the probability that success on these questions will produce a reversal or an order for a new trial, as explained above. We necessarily cannot and do not rule finally on the merits of the appeal. That ruling will come after the appeal is submitted to a panel for decision in the usual course. Having considered the arguments that defendants seek to raise, we hold that they have not met the statutory standard. Although the questions are not frivolous, and although at least one of them (insufficiency of the evidence) would surely produce a reversal if we ultimately were to agree with it, they are not close enough to be "substantial" within the meaning of the statute. At the risk of repeating ourselves, we emphasize that this holding is tentative, in the sense that it is without prejudice to whatever the panel deciding the appeals may determine after full briefing, argument, and study.

In No. 84-2439, Charles Bruce Nabors was convicted of bank robbery and of conspiracy to rob a bank. He was sentenced to imprisonment on both counts, the terms to run concurrently. In support of his appeal, he urges that there was a variance between the copy of the indictment given him and the copy used by the prosecution at trial, in that the later version of the indictment contained an overt act omitted from the version that had been delivered to the defendant. He also argues that the prosecutor's final argument called attention to his failure to testify and impermissibly invited the jury to make an inference unfavorable to him from his silence. The first argument, even if it succeeded, would apparently go only to the conspiracy count, so on that score it would fail the "likely to result in reversal" element of the test. Furthermore, whatever error may have occurred may have been cured by the District Court's prompt admonition to the jury not to consider the additional overt act. As for the prosecution's closing argument, we have read the transcript, and, subject to whatever briefing and further study might develop, it seems that the argument was more an attack on counsel for the defense for failing to advance certain theories, than it was a reference to the defendant's personally failing to take the stand. The second argument, in other words, does not seem close, and the first one, even if it is "substantial," would, even if successful, produce a reversal on only one of the two counts on which imprisonment has been imposed.

In Nos. 84-5223 and 84-5225, Kent August Moeckly and William Joseph Coulombe were convicted of three counts of conspiracy to import and to possess with intent to distribute cocaine, in violation of 21 U.S.C. Secs. 841, 846, 952, and 963 (1982). In addition, Moeckly was convicted of two counts of perjury, in violation of 18 U.S.C. Sec. 1623 (1982). In their joint memorandum in support of their application for bail, defendants argue that they have a substantial chance of prevailing on appeal, but they do not offer specific issues that they claim will have this effect. While this sort of application might have been acceptable under the old law, it gives us nothing specific with which to work, and we cannot say that defendants are raising "substantial" questions within the meaning of the new statute.

In No. 84-2493, Bayard Spector was convicted on one count of conspiracy to distribute cocaine in violation of 21 U.S.C. Sec. 846 (1982) and one count of interstate travel in furtherance of the cocaine offense, in violation of 18 U.S.C. Sec. 1952 (1982). Defendant seeks to raise a number of issues on the appeal, but we need to discuss only one. He claims that a key witness for the government was induced to testify by a promise that he would receive favorable treatment on charges pending against him if his testimony led to "successfully solving and prosecuting crimes." This language, defendant says, can be taken to mean that Adams would not receive favorable treatment unless he, Spector, were convicted by the jury, and this sort of incentive, it is argued, creates too great a likelihood of perjury.

The argument is similar in many respects to a contention upheld by a panel of this Court in United States v. Waterman, 732 F.2d 1527 (8th Cir.1984). Rehearing en banc was granted on the government's petition in Waterman, an action which under our practice has the legal effect of vacating the panel opinion, but the subsequent history of the case compels a holding that the issue is "close," one that could go either way. For after argument to the Court en banc, the conviction was affirmed by an equally divided vote of four to four. 732 F.2d 1527, 1533 (8th Cir.1984) (order) (en banc), petition for cert. denied, --- U.S. ----, 105 S. Ct. 2138, 85 L. Ed. 2d 496 (1985). One judge did not participate because of illness at the time of the en banc argument, but he had been a member of the original panel and had voted to reverse the conviction at that time, so the vote to reverse might have been five to four had all active members of the Court participated in the en banc argument. However that may be, a question on which this Court is evenly divided must be the easiest kind of issue to label "close." There may be distinctions between the agreement with the government witness in this case and that at issue in Waterman, and no doubt these distinctions will be argued in the briefs in Spector's appeal, but we think the question is still clearly a "substantial" one for present purposes. And, if Spector prevailed on this question, which relates to the government's key witness, it would likely require reversal or a new trial. Again, our decision is for bail purposes only and without prejudice to whatever conclusion the panel which hears this appeal may reach after plenary consideration.

In No. 84-2449, Louis Kenneth Risken was convicted of four counts involving conspiracy to have a grand-jury witness murdered and making false statements to a second witness to influence his testimony before the grand jury. Violations of 18 U.S.C. Secs. 1503, 1512(a)(2)(A), 1502, and 1512(a)(1) (1982) were alleged. As to counts I and IV, the charges involving harassment and intimidation by force, defendant argues that he could be prosecuted only under Sec. 1512 and not under Sec. 1503, on the ground that when Sec. 1512, relating specifically to attempts to influence a witness by intimidation, physical force, threats, or misleading conduct, was enacted, Sec. 1503, which had covered generally unlawful efforts to influence or intimidate jurors, officers, and witnesses, was amended by striking any reference to witnesses. This argument, that threats against witnesses are henceforth to be dealt with only under Sec. 1512, has been accepted by at least one other circuit, United States v. Hernandez, 730 F.2d 895, 899 (2d Cir.1984), and there is no authority directly on the point in this circuit. We think this sort of argument clearly falls in the category of questions that could go either way. In addition, defendant claims that after the trial he learned that a key government witness had been paid $5,000, whereas at trial the witness testified that he had received only $500 and had not been guaranteed any additional payment by the government. On the record that was before us at the time we considered the bail application, we could not be certain whether there had been a misrepresentation by the government on this score. It is possible that the additional payment was made to the witness after he had testified, and that in fact no promises of additional payments had been made to him. Still, the additional payment, coming so soon after the conclusion of the trial, may have been reasonably forseeable by the government at the time of trial, and, if so, it seems to be the sort of thing that the government or its witness ought to have revealed. Whether in fact this argument turns out to be a strong one we cannot forecast with certainty at this point, but on the basis of the papers before us we believe it should be classified as a "close" question. Also, this question pertains to a key government witness, and if Risken won on this point, reversal or a new trial would be likely.

For the reasons given in this opinion, we deny the applications for bail pending appeal of the appellants Powell, James Barfield, William Barfield, Nabors, Moeckly, and Coulombe, and the decisions of the district courts denying their applications are affirmed. We grant the applications for bail pending appeal of the appellants Bayard Spector and Louis Risken. The decisions of the district courts denying their applications are reversed, and the causes remanded to the district courts with directions to enlarge them on bail upon such terms and conditions as may be reasonable, all in accordance with our order entered on January 22, 1985.

It is so ordered.

PER CURIAM.

The motion of appellant Bayard Spector for release on bail pending appeal is granted, subject to such reasonable terms and conditions as the District Court may fix in its discretion. The District Court is requested to set these terms and conditions forthwith.

ROSS, JOHN R. GIBSON, FAGG, and BOWMAN, Circuit Judges, would deny the motion for bail pending appeal at this time and remand for further consideration by the District Court.

The motion of Louis Kenneth Risken for release on bail pending appeal is granted, subject to such reasonable terms and conditions as the District Court may fix in its discretion. The District Court is requested to set such terms and conditions forthwith.

ROSS, FAGG, and BOWMAN, Circuit Judges, would deny the motion for release on bail pending appeal at this time and remand for further consideration by the District Court.

The motions of Kent August Moeckly, William Joseph Coulombe, Geary David Powell, James Barfield, Bill Barfield, and Charles Bruce Nabors for release on bail pending appeal are denied, and the orders of the District Courts denying bail pending appeal for these appellants are affirmed.

LAY, Chief Judge, HEANEY, BRIGHT, and McMILLIAN, Circuit Judges, would vacate the orders of the District Courts denying these appellants' motions for release on bail pending appeal and remand these cases for further consideration.

Opinions will be filed in due course expressing the views of the Court and of the various dissenting judges that have led to this order.

Let the mandates issue forthwith.

It is so ordered.

LAY, Chief Judge, concurring.

I concur in the court's interpretation of 18 U.S.C. Sec. 3143(b), but write separately to address Judge Gibson's dissent. The dissent finds fault with the majority's interpretation of the 18 U.S.C. Sec. 3143(b) requirement that a defendant must raise a "substantial question of law or fact likely to result in reversal or an order for new trial." Rather than follow a two-part objective analysis of the statutory phrase, the dissent would interpret the provision literally, and rely on the "objective detachment and conscience of a district judge * * * " to avoid the inherent subjectivity in the literal test. I respectfully must disagree. District judges have found a literal reading of the statute difficult to apply. See United States v. Miller, 753 F.2d 19, 22 (3rd Cir.1985) (quoting the district judge's observation that "the Act 'practically means that the district judge has to determine that he has probably made an error in the decision that he has rendered in the lower court * * *.' ") (citation omitted).

More significantly, a subjective interpretation of the statute creates serious due process and other constitutional questions which are obviated under this court's holding. I would find it extremely difficult to concur in the constitutionality of section 3143(b) if the dissenting view were the prevailing one.

McMILLIAN, Circuit Judge, concurring in part and dissenting in part.

I concur in most of the majority opinion's analysis. I fully agree with the majority opinion's interpretation of the second or "likely to result in reversal or an order for new trial" part of the two-part analysis. I believe that the majority opinion's interpretation of the second part is essentially the same one set forth by the Third Circuit in United States v. Miller, 753 F.2d 19, 23 (1985), although the majority opinion has helpfully clarified and elaborated upon the language in United States v. Miller. Where I disagree with the majority opinion is the first or "substantial question" part of the analysis. I would prefer to adopt the interpretation of the first or "substantial question" part set forth by the Third Circuit in United States v. Miller, id., instead of the narrower interpretation set forth by the Eleventh Circuit in United States v. Giancola, 754 F.2d 898, 901 (1985) (per curiam).

I also agree that the provision for release on bail of convicted persons pending direct appeal in the Bail Reform Act of 1984, as interpreted in the majority opinion, is constitutional.

JOHN R. GIBSON, Circuit Judge, with whom BOWMAN, Circuit Judge, joins, concurring in part and dissenting in part.

I concur in the actions of the court today, except that I would reject Bayard Spector's application for bail pending appeal. Further, I am in agreement with much of what the court says. However, because I believe the majority essentially has rewritten an act of Congress rather than given the act its natural, reasonable interpretation, I express my separate views and, particularly, my understanding of the phrase "likely to result in reversal or an order for a new trial."

I acknowledge the seductive appeal of the court's opinion. The logic of its exegesis, I suppose, may at first glance leave one with every reason not to oppose it. Nonetheless, I believe that to the question the majority poses--"Why not interpret it [i.e., the statutory language] in what may be the most natural and immediately obvious sense, as requiring the defendant to show that the question presented will more likely than not result in reversal?"--the proper answer is that such an interpretation is, in fact, correct. Such a construction not only adheres to the clear language of the statute but also bears out the intent and purpose of Congress.

I have no quarrel with the court's interpretation of the phrase "substantial question." To define it as "a close question" or "one that could go either way" is a more workable, practical test than those given in United States v. Miller, 753 F.2d at 23, although the tests there articulated, ("whether the question is "novel," "not decided by controlling precedent," or "fairly doubtful") may have some value in fleshing out the simplicity of the court's expressed formula.1

However, when the court reaches the second phase of its inquiry, it construes the phrase "likely to result in reversal or * * * a new trial" simply to mean that prejudicial error is required. The court sets out the only legislative history that bears at all on the meaning of this phrase: "The change [in the law] * * * requires an affirmative finding that the chance for reversal is substantial. This gives recognition to the basic principle that a conviction is presumed to be correct." S.Rep. No. 98-225, 98th Cong., 1st Sess. 27 (1983), U.S.Code Cong. & Admin.News p. 3210. Further, the court correctly observes that the word "likely" "should be read in its ordinary sense as referring to something that is more likely to happen than not."2 Supra at 1233.

However, the court then simply rewrites the statute in accord with its view of reasonableness by, in essence, inserting the phrase "if decided in defendant's favor" between the word "likely" and the phrase "to result in reversal or * * * a new trial." This reduces the meaning of the whole phrase, leaving it to require no more than that release pending appeal should not be granted if the question could be disposed of on a harmless error or a procedural default basis. The approach adopted by the court is simple to apply, relatively certain of result, and possesses the appeal of sweet reason. Its only fault is that it simply is not what Congress has enacted. Further, the approach of the court fails to carry out Congress's intent that there be "an affirmative finding that the chance for reversal is substantial."

While the court may be correct in suggesting that to read "likely to result in reversal * * * or a new trial" as I do renders the adjective "substantial" redundant, at least to the extent that questions likely to so result will also be substantial, I do not find this fatal to my interpretation. While it is generally true that interpretations leading to redundancies should be avoided, as the Supreme Court has also stated, the presence of "a clear legislative purpose" may compel such a reading. Singer v. United States, 323 U.S. 338, 344, 65 S. Ct. 282, 285, 89 L. Ed. 285 (1945). Here I believe the legislative history, scant though it is on this point, and "the most natural and immediately obvious sense" of the statutory language establish such a purpose.

I cannot agree with the statement in Miller that such a construction is capricious and would put the district court in the position of a "bookmaker" who "trade[s] on the probability of ultimate outcome." 753 F.2d at 23. Federal district judges currently make similar determinations in deciding whether a stay of judgment pending appeal should be granted under Fed.R.App.P. 8. We rely on the objective detachment and conscience of a district judge to grant a new trial when he is satisfied that trial error has occurred that will lead to reversal. At other times, however, issues may arise which the district judge may consider not to have such certainty, but which may still suggest the "likelihood of reversal or * * * a new trial." Congress made clear that the presumption is against release, and it is only the presence of this latter, limited class of issues that may permit the release of a defendant on bail pending appeal.

The court argues that such a position has not been urged by the United States and that we should not adopt "a posture more zealous than that of the prosecution." If there is zealousness present, it is that of Congress displayed in enacting the statute. The question before us is not that of our posture or that of the prosecution, but what Congress wrote and intended. We should not blithely add a phrase because the attorneys representing the United States urge a meaning of a statute at odds with the statutory language. While deference should be paid to the interpretation of a statute by the agency responsible for administering it, as the Supreme Court has stated, such constructions must be rejected when they "frustrate the policy that Congress sought to implement." Securities Industry Association v. Board of Governors of the Federal Reserve System, --- U.S. ----, 104 S. Ct. 2979, 2983, 82 L. Ed. 2d 107 (1984) (quoting Federal Election Commission v. Democratic Senatorial Campaign Committee, 454 U.S. 27, 32, 102 S. Ct. 38, 42, 70 L. Ed. 2d 23 (1981) ).

If one reads the statute straightforwardly, its immediate and apparent sense is simply that following conviction a person may be released on bail if, in addition to the other requirements of the statute, he has raised a substantial question of fact or law that is likely (i.e., more probable than not) to result in reversal or an order for a new trial. What legislative history there is supports this reading. The statute should be applied as written. The second phase of the analysis should consider the issue of the likelihood or probability of error--not just whether harmless, as opposed to prejudicial, error is present.3

Accordingly, I respectfully dissent from that portion of the court's opinion which, in essence, adds the phrase "if decided in defendant's favor" to the statute that Congress has enacted. If Congress had felt it desirable, it would have added the language. Congress wished to substantially restrict the number of persons released on bail pending appeal. We should not judicially legislate a meaning more permissive than that clearly apparent from the language of the statute.

FAGG, Circuit Judge, with whom BOWMAN, Circuit Judge, joins, concurring in part and dissenting in part.

I join in Judge Gibson's concurring and dissenting opinion, except I would also reject Risken's application for bail pending appeal.