Christopher Guillory was convicted in Louisiana of three counts of first degree murder. In this federal habeas petition he attacks the process for selecting the foreperson of the grand jury that indicted him in Calcasieu Parish. The system of selection that Guillory complains of here was at issue in Campbell v. Louisiana1 in 1998, discontinued the next year by an amendment to the state criminal code and condemned in State of Louisiana v. Ricky Langley,2 in reliance upon Campbell.

The contention is that before the change in the criminal code, the judge presiding over the grand jury selected a foreperson from the general venire summoned for grand jury service and not from those randomly chosen for service from that venire. The argument continues that this unguided discretion, left to presiding judges, formerly white males, had produced a historical pattern of selecting white males over blacks and females sufficient to create a prima facie case of both racial and gender-based discrimination.

Both the state courts of Louisiana and the federal district court refused Guillory's request for relief. The United States District Court concluded after conducting an evidentiary hearing that the state had overcome Guillory's prima facie case of discriminatory selection with the testimony of the state trial judge who selected the foreperson in his case. It then issued a certificate of appealability limited to "whether the indictment should have been quashed due to discrimination in the selection of the Grand Jury foreman".

* Guillory was indicted on May 6, 1993, on three counts of murder and found guilty by a jury on May 10, 1996. He was sentenced to three life terms. The Third Circuit Court of Appeals affirmed his conviction and sentence on March 11, 1998, and the Supreme Court of Louisiana denied certiorari on October 9, 1998, in an unpublished opinion. On direct appeal Guillory raised 20 points of error, including in one assignment that the indictment should have been quashed because there was racial and sexual discrimination in the selection of the foreperson of the grand jury in Calcasieu Parish. The state court of appeals rejected this contention, finding that Guillory had failed to make a prima facie case. It faulted the absence of a statistical comparison of the race and sex of the selected foreperson with the venire from which they were drawn, applying State v. Young's3 teaching that drawing a proportion with general population figures was not meaningful.

Guillory then filed on October 8, 1999, a federal petition for writ of habeas corpus under 28 U.S.C. 2254, the petition now before us. He asserted eleven claims, but only the claim that the selection of the foreperson was tainted by discrimination remains.

A federal magistrate judge, after reviewing affidavits submitted by the parties at his direction, conducted an evidentiary hearing. Guillory's counsel had developed an extensive study of the patterns of selection of forepersons of grand juries in Calcasieu Parish, as counsel in the Langley case. By agreement of counsel, this study, as well as the testimony of Dr. Joel Devine who testified in Langley as an expert in statistics, was admitted before the magistrate. The state relied upon the testimony of Judge Wilford Carter, the presiding judge who selected the foreperson, and Dr. Nola McDaniel, the state's expert in statistics.

The state did not contend that census figures could not serve as a base in establishing a prima facie case of discrimination. Rather, its strategy was to accept that a prima facie case was established by the same record developed in Langley and then carry the burden of responding to it.4 In short, events in this case overran the limited circumstances under which a federal court can grant an evidentiary hearing in federal habeas review of a state conviction.5 Our question is then whether the finding of no intentional discrimination by the magistrate judge was clearly erroneous.

The state first contends that the historical record of the appointing process is not relevant when the trial judge who made the appointment at issue explains why the selection was made. We agree in part. But the argument misses the point that it was the practice of requiring the presiding judge to select a foreperson from the venire and not from randomly selected members of the grand jury that opened the door to discrimination. And a simple denial that race or sex had nothing to do with a selection that followed this selection practice, in the face of the statistically established prima facie case it produced, is not adequate.6 This does not mean that the selecting judge cannot offer objective and nondiscriminatory reasons for the selection which, if found to be credible by the trier of fact, will defeat the prima facie case. But that very prima facie case places the burden of offering such explanations upon the official.

The state's related argument is that since Judge Carter was black, a new judge making his first selection, and at the time of the federal hearing had only made three selections, there was an inadequate basis for inferring discrimination. This ignores the force of the prima facie case created by the unchallenged evidence of the historical record that this system produced: the burden is now upon the state to offer nondiscriminatory reasons. Judge Carter's short tenure as a judge at the time of the now-contested selection, his first, and a total of only three selections will not alone support an inference of intentional discrimination. That is plainly so, as Dr. McDaniel observed, but that reality also cuts against the state's efforts to draw upon it in defense. So we return to our required inquiry of whether the state has offered credited evidence of nondiscriminatory purpose.

We review findings of fact by the standard of clearly erroneous and questions of law de novo. The magistrate judge made findings regarding the question that controls this case, whether there was intentional discrimination. This finding of fact in the context of this case a fortiori overcomes the prima facie case. There is no legal question here regarding the structure or mechanics of the prima facie case and the state's burden of proof. We have only a pure question of fact, and we review only for clear error.7 We start with the findings of fact that underpin the finding of no intentional discrimination.

Judge Carter testified that his primary aim in selecting a foreperson was to choose a person "who would be fair and independent" and "not necessarily go along" with the government. He denied that race or gender was a factor in his selection of Edwin Eisen, a white male administrator at McNeese State University, to be foreperson of the grand jury that indicted Guillory. He explained that as an alumnus of the University, he knew Eisen and was familiar with his education and reputation for not being "a go along person". This, he said, was important because he wanted a person who would stand up to the district attorney. At the same time, Judge Carter made plain that "to some extent" women and minorities should be given some preference, given their historic under-representation.

If this were the sum of Judge Carter's testimony, it would offer strong support to the finding by the magistrate judge of no discrimination. The complication in this otherwise straightforward case rises from additional comments he made, largely volunteered. On the matter of race, Judge Carter expressed his concern that blacks would tend to go along with the district attorney. On the matter of gender, while expressing no reservations in selecting a female as foreperson, he observed that women made better secretaries than men because they were more careful with detail; that he had never had a male secretary.

These and other stereotypical observations cut against the magistrate's finding of no discrimination and give us pause. They may well be read to confess forms of both racial and gender bias. Accepting them as such, the question remains, however, whether they animated Judge Carter's decision to select Eisen. The magistrate judge, with the benefit of hearing and observing this witness, an experienced lawyer and sitting judge, concluded that the selection of Eisen, a person known by Judge Carter to possess traits that are plainly desirable, was not the product of intentional discrimination. Ultimately, we are persuaded that the magistrate's findings that Judge Carter selected Eisen because of his education and reputation and not because of race or gender are not clearly erroneous.

The presence of identified objective criteria known in advance to the appointing judges would have mitigated the difficulties of the selection system then in place. The process was flawed, and that seeded the statistical pattern underlying the prima facie case. It does not follow however that every selection of a grand jury foreperson in Calsiesu Parish before the criminal code was amended was the product of discrimination. Not only was Judge Carter's selection of Eisen supported by nondiscriminatory reasons which he articulated, there is no evidence that any other member of the venire was better qualified. Of course, without more, there would remain the inference that black or female candidates faced the headwind of Judge Carter's bias and were passed over; that the absence of a selection process that considered every member of the venire by objective standards favored white males.

It is the case that the flawed system adversely impacted blacks and females. That is the prima facie case. But that doesn't answer the ultimate question of whether this selection was the product of intentional discrimination. The selection of blacks by Judge Carter in the two selections that followed the appointment at issue, while offering meager statistical fodder, is direct evidence that any assertion that Judge Carter's expressed concern that blacks would not be sufficiently independent of the district attorney did not find expression in his appointments. Taken through the venire list by Guillory's counsel, Judge Carter identified two black males he knew and thought qualified, but explained that both were in law enforcement and in his view should not serve as a foreperson. He did not pass over them because he thought blacks would not be sufficiently independent of the prosecutor. As for gender discrimination, Judge Carter testified that he would have been "glad to get a woman he was as comfortable with as Eisen," but did not know one on the venire that was better qualified or that he knew as well. Fairly read, his statement regarding his being "comfortable with" his appointment speaks to relative qualifications-education and leadership experience. Critically, these circumstances and reasons for selection were credited by the trier of fact. There is record evidence to support them, and we must then affirm the ultimate legal conclusion that in this case the flawed system did not produce a selection that was the product of intentional discrimination.

To accept the magistrate's findings of fact while rejecting the magistrate's conclusion of law would translate to a conclusion that the process of selection itself compelled a finding of intentional discrimination. This confuses the prima facie case with the elements of an entire case.

Nor is our decision in Guice v. Fortenberry, 722 F.2d 276 (1984), to the contrary. There the same selection process was at issue, and the majority of the panel held that the denial of discrimination by the presiding judge was insufficient to overcome the petitioner's prima facie case. That proof included evidence that no black person had ever served as the foreperson of a grand jury in the parish. The judge making the selection at issue there was a white male and had served as judge since October of 1963. He had selected twenty eight of the thirty two forepersons selected during the time from his appointment to the impaneling of the grand jury that indicted the petitioners Guice and Claxton. All were white. The panel stressed the absence of objective selection criteria and viewed his testimony "with a great deal of judicial scrutiny." It found clear error in crediting his testimony. Given his past record, his explanation for his selection of a local banker was found to be business as usual.

We do not read Guice to hold that the sole means of rebutting a prima facie case is proof that racially neutral procedures have been independently adopted. That may well be the case when the only rebuttal of the prima facie case is a denial from the official whose decisions created it. Here we have a new judge who was selecting a foreperson for the first time. He had not been a part of the regime that produced the monochromatic all white male result and the prima facie case. Indeed, his later selections were both black. The prima facie case was comprised of decisions made before he became a judge, and he articulated criteria that went beyond not knowing only members of the same race and sex-criteria that led him to select black forepersons for other grand juries.

It is true that Judge Carter's testimony read in transcript is at times rambling and convoluted, but that is common to direct transcription of oral testimony. As we observed, his studiously open and politically incorrect phrasings leave his testimony vulnerable to being read as proof that he harbored both racial and sexist views. And a forceful argument is made that they are. The argument goes to the question of whether the reasons he gave for his decision were to be believed. If credited, they are adequate to rebut the prima facie case. They need not have been credited, but they were. We are persuaded that this was the call of the magistrate judge and we cannot say that it was clearly erroneous, however we think we may have viewed it, if we had been presiding at the evidentiary hearing.

AFFIRMED.