concurring.
I join the opinion of the Court in its entirety, and deem it appropriate to add a few words.
The trial judge was prejudiced. We know this because of his own statements, which are of record. Ordinarily no further inquiry would be necessary. Nothing else would matter. If due process means anything, it means a trial before an unbiased judge and jury. That is why I made the statement at the oral argument, quoted by my Brother Bright, post, p. 1252, that “if I had been on this court when this case came up before, I would have voted to grant the writ.” 1
*1250In fact, of course, I was not on the Court 15 years ago, when the denial of the first habeas petition was unanimously affirmed. Only one judge who sat on the 1969 Walker panel is still an active Circuit Judge. Only three judges who are active Members of this Court today were on the Court then. The point is this: a judge who is today presented with a fresh issue is asked to solve a far different problem from a judge who is asked to repudiate the opinion of his own Court on the very question now before him. In the latter instance, the question is not simply, “Did James Dean Walker get a fair trial?,” but rather, “Did the United States Court of Appeals for the Eighth Circuit, when it held, on the fullest consideration and after an evidentiary hearing, that James Dean Walker got a fair trial, err so egregiously that the ends of justice require the same Court, 15 years later, to come to the opposite conclusion?” It is not a question, then, of simple agreement or disagreement with this Court’s prior position. (I suspect that, but for the fact that I have other work to do, I could leaf through any volume of the Federal Reporter and find at least one opinion of this Court with which I disagree.) Something more than mere disagreement must be shown to justify a successive habeas petition.
What is that something more? Sanders v. United States, 373 U.S. 1, 83 S.Ct. 1068, 10 L.Ed.2d 148 (1963), gives some examples, but none of them applies here. There is no new evidence, unrevealed at the time of the first habeas proceeding. There is no change in the law. There is of course no claim of taint in the first post-conviction proceeding. There is simply the claim that this Court in 1969 misapplied the law to the facts. I know of no case that grants a successive habeas petition in that situation. (United States v. Brown, 539 F.2d 467 (5th Cir.1976), relied on by the dissenting opinion, post, pp. 1258-59, was a first, not a successive, petition for post-conviction relief.) No doubt there may be such a case some day, but this one should not be it, for several reasons.
First, I am not persuaded that the trial judge’s bias did Walker any actual harm that would not have occurred if the case had been tried by another judge. The judge did deny a short recess, desired by defense counsel to enable them to try to show that two witnesses “could be found nightly in Little Rock nightclubs,” post, p. 1256. But this incident occurred near the beginning of the State’s case. If in fact the availability of the witnesses Ford and Roberts could have been shown in such short order, would not defense counsel have gone out that very night, found witnesses who knew Ford and Roberts were in town, and presented them in chambers the next morning in support of a motion to strike or for a mistrial? If in fact Ford was available, and if her testimony in person would have differed from the transcript of the first trial, would counsel not have found her and demolished the prosecution by calling her as a defense witness? Perhaps not, but the likelihood is sufficiently great to persuade me that the denial of the 30-minute continuance was not in fact so crucial in the second trial as the dissenting opinion argues.
As to the defense ballistics report, I simply have no fault to find with the trial judge’s ruling. It was hearsay and not admissible on any theory. The trial court’s order that the report be made “part of the record,” whatever it may mean, simply cannot reasonably be read as a stipulation by the prosecution that the report could be read to the jury. No prosecutor in his right mind would agree that an ex parte examination of an expert hired by the defense be admitted into evidence, at least without some substantial concession in return.
Finally, I return to the dispositive standard, “the ends of justice.” Certainly justice to the petitioner is crucial in our system. But he is not the only party to the case. The State also has a right to fairness, and we have to consider as well the effect on the administration of justice as a whole of a decision to grant this successive application. Petitioner has, after all, been convicted by a jury, on evidence that meets the Jackson v. Virginia standard. He does not claim that the jury was biased. It would be difficult, if not impossible, for the State to retry him at this point. An error may be sufficiently grave to cause a reversal on direct appeal, or on the first habeas application, but not sufficiently grave, when weighed in the balance with all the other constituent elements of “justice,” to justify relief on a successive habeas application. In my view, that.is the category in which the error committed here fits.
Accordingly, although there is much to admire in the dissenting opinion, denial of the writ seems to me less unjust than granting it. I vote to affirm the judgment *1251of the District Court, and I join Judge John R. Gibson’s cogent opinion for this Court en banc.
. My further remark that "there ought to be a reprieve,” also quoted, post, p. 1252, was improvident. Executive clemency is not our business. The Governor has his constitutional responsibilities, just as we have ours. Having said that, I remain of the view that this case does little credit to the State of Arkansas,