II. ANALYSIS

A. Ineffective Assistance of Counsel on Direct Appeal4

A defendant has a right to counsel to aid in the direct appeal of his or her criminal conviction. See Evitts v. Lucey, 469 U.S. 387, 105 S.Ct. 830, 83 L.Ed.2d 821 (1985). This right to counsel is violated when appellate counsel is ineffective. Id.; Alvord v. Wainwright, 725 F.2d 1282 (11th Cir.1984). This Circuit has applied the Supreme Court's test for ineffective assistance at trial, see Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984), to guide its analysis of ineffective assistance of appellate counsel claims. See Orazio v. Dugger, 876 F.2d 1508 (11th Cir.1989). Therefore, Heath must show that his appellate counsel's performance was deficient and that this performance prejudiced the defense. Strickland, 466 U.S. at 687, 104 S.Ct. at 2064. While Heath can demonstrate that his appellate counsel's performance was deficient, Heath is unable to show that this deficient performance prejudiced his defense.

1. The Performance Prong

Strickland held that in evaluating whether the appellate counsel's performance was deficient, counsel's performance must be evaluated for "reasonableness under prevailing professional norms." Strickland, 466 U.S. at 688, 104 S.Ct. at 2065. Courts should presume effectiveness and should avoid second-guessing with the benefit of hindsight. Id. at 689, 104 S.Ct. at 2065. Specifically, Strickland encouraged reviewing courts to allow attorneys broad discretion to represent their clients by pursuing their own strategy. However, the Court realized that merely invoking the word "strategy" to explain errors was insufficient since "particular decision[s] must be directly assessed for reasonableness [in light of] all the circumstances." Id. at 691, 104 S.Ct. at 2066.

The Supreme Court, on at least two occasions, has had an opportunity to explain the parameters of what constitutes a reasonable strategy for appellate advocates. In Jones v. Barnes, 463 U.S. 745, 103 S.Ct. 3308, 77 L.Ed.2d 987 (1983), the Court held that the Sixth Amendment does not require appellate advocates to raise every non-frivolous issue. The Court suggested that effective advocates "winnow out" weaker arguments even though the weaker arguments may be meritorious. Id. at 751-52, 103 S.Ct. at 3313. The Court in Smith v. Murray, 477 U.S. 527, 106 S.Ct. 2661, 91 L.Ed.2d 434 (1986), held that an appellate advocate who reviewed the entire record, thought about various claims, and then chose to pursue thirteen claims on appeal had furnished effective appellate assistance. The Court recognized that even though a non-appealed issue might have been successful, the appellate advocacy had to be judged in its entirety.

In the case at bar, appellate counsel's performance was deficient. Heath's attorney selected only one issue to appeal, the double jeopardy claim.5 The argument section of his brief to the Court of Criminal Appeals was only six pages long. The argument section of his subsequent brief to the Alabama Supreme Court was only one page long. While we are loath to encourage attorneys to file numerous claims merely for the sake of filing claims, we also realize that in a capital appeal the attorney must raise many issues in a timely fashion or else the defendant may be procedurally barred from ever raising those issues. We also note that the quality of the one claim that was briefed was unreasonably deficient. Heath has provided affidavits from four criminal appellate attorneys with expertise in capital appeals who testified that his appellate counsel's decision to rely on only one claim was below "prevailing professional norms." Strickland, 466 U.S. at 688, 104 S.Ct. at 2065. All four attorneys stated that, had they supervised this appeal, they would have forbidden the filing of these briefs. One attorney stated that in Kentucky since 1976 the average capital appeal has raised 34 issues and the brief has averaged 194 pages.

This is not a case where the attorney made a strategic decision to winnow out his less persuasive claims. Heath's attorney Roney, in his testimony during the state collateral review, did not advance any reasonable strategic reasons for raising only the double jeopardy claim. While he focused on the double jeopardy claim to the exclusion of other issues during the appeal, he admitted that even before the trial, he thought the double jeopardy issue was a losing issue. Furthermore, he explained that he did not bring the change of venue/pretrial publicity claim on appeal because he thought that Heath was guilty and that Heath would be found guilty wherever he was tried. Evidently, Roney was not aware of the fact that his client had a constitutional right to a fair trial regardless of his client's guilt. We are not willing to characterize this mistaken understanding of the law as a reasonable strategy.

The state argues that Heath's appellate counsel's performance was reasonable because he was able to convince the Supreme Court to grant certiorari. While the grant of certiorari is impressive and indicates that the issue that Heath's attorney selected is "an important question of federal law," Rules of the Supreme Court, Rule 10.1(c), this alone is not dispositive. First, we note that Heath's original appellate counsel, Roney, did not draw up the petition for certiorari. Heath obtained Professor Allen, a professor from Northwestern University School of Law, to pursue the certiorari petition after his appeal in the Supreme Court of Alabama had failed. Second, the grant of certiorari does not necessarily indicate that the position advocated by Heath has any merit, only that it is an important question. We are reluctant to conclude that the selection of an important question of federal law, to the exclusion of other issues, which was ultimately unsuccessful, is sufficient to redeem an otherwise poor appellate performance.

2. The Prejudice Prong

While we conclude that Heath's appellate counsel's performance was deficient, we cannot conclude that Heath is able to show any prejudice from this poor performance. A petitioner has satisfied the prejudice prong of Strickland when he or she can show that the appellate counsel's performance was sufficiently deficient to deprive the defendant of "a trial [or an appeal] whose result [was] reliable." Strickland, 466 U.S. at 687, 104 S.Ct. at 2064. In the context of an ineffective assistance on appeal claim, this Court in Cross v. United States, 893 F.2d 1287 (11th Cir.1990), held that in order to determine prejudice the court must first perform "a review of the merits of the [omitted or poorly presented] claim." Id. at 1290. If the Court finds that the neglected claim would have a reasonable probability of success on appeal, then according to Cross it is necessary to find "appellate counsel's performance prejudicial because it affected the outcome of the appeal." Id.

With the legal standards for prejudice in mind, Heath claims that his appellate counsel would have been successful if he had raised two claims on appeal.

a. Denial of Fair and Impartial Jury

Heath first alleges that he was denied a fair and impartial jury due to extensive and highly prejudicial pretrial publicity.6 In support of this allegation, Heath notes that local TV and newspapers extensively covered the murder, the investigations, and the assorted trials. Heath's allegations raise three related claims: first, whether the trial court erred in not striking individual jurors for cause; second, whether the publicity created inherent prejudice; and third, whether the publicity caused actual prejudice.

Heath argues that the trial court erred in failing to strike some of the jurors for cause. A prospective juror must be removed for cause if his or her views " 'would prevent or substantially impair the performance of his [or her] duties as a juror.' " Wainwright v. Witt, 469 U.S. 412, 424, 105 S.Ct. 844, 852, 83 L.Ed.2d 841 (1985) (quoting Adams v. Texas, 448 U.S. 38, 45, 100 S.Ct. 2521, 2526, 65 L.Ed.2d 581 (1980)). This Circuit has held that the question of whether a juror was in fact biased is a question of fact. See Bundy v. Dugger, 850 F.2d 1402, 1426 (11th Cir.1988). A habeas petition will be granted for a state trial court's failure to strike a juror for cause only when there is not fair support in the record for the trial court's determination that the juror was unbiased. Id. Heath discusses the fact that, of the eighty-one veniremembers, he challenged sixty-seven of them for cause (including all twelve of the eventual jurors), and the trial court granted only eight of his challenges for cause.7 However, even if a veniremember should have been struck for cause, the Supreme Court in Ross v. Oklahoma, 487 U.S. 81, 108 S.Ct. 2273, 101 L.Ed.2d 80 (1988), held that there is no constitutional violation where the biased veniremember does not eventually sit on the jury. The Court in Ross held that a habeas petitioner's constitutional rights were not violated when he was forced to waste a peremptory challenge to remove a veniremember whom the court should have removed for cause. Therefore, in this habeas petition, Heath can raise only the trial court's denials of Heath's challenges for cause of those veniremembers who eventually sat on the jury.

After a close examination of the record, we conclude that there is at least "fair support" in the record for the trial court's decision to deny Heath's motions to strike each of the jurors for cause. We are unable to find a single member of the jury whose answers indicated that his or her prior knowledge or pre-dispositions would impair his or her performance as a juror. While all of the jurors had heard of the case in the media and several of them had heard that Heath had pled guilty in Georgia to the same crime,8 all of the jurors stated that they would be unbiased and that they would follow the judge's instructions. Moreover, from the skeletal nature9 of the questions and answers, we are hard pressed to find any indication of any bias against Mr. Heath. In the Bundy case, we found no error in a trial judge's refusal to strike for cause a juror who had a negative impression about the defendant, might be influenced by the defendant's lack of testimony, and was uncertain if he or she had a fixed opinion about the defendant's guilt. Bundy, 850 F.2d at 1428 (Juror No. 3). In that case, we focused on the juror's promise to follow the judge's instructions and to judge the evidence fairly. Id. In the case at hand, the jurors made the same promises, and we are unable to find anything in the record suggesting that the jurors did not intend to keep these promises.

Second, Heath argues that he received an unfair trial because the news media closely followed the crime, investigation, and trials and published distorted and prejudicial information. Heath has compiled over 100 newspaper articles and four videotapes of television newscasts. He argues that the reports resulted in an unfair trial because the articles focused on the sensational aspects of the murder. Heath claims that many of the articles were melodramatic. The articles, Heath alleges, focused on the fact that Mrs. Heath was nine months pregnant when she was murdered. The articles referred to the dead fetus by the name that Heath had said he and his wife had agreed upon, and they called the murder a double homicide. In one article, the Georgia prosecutors lamented their inability to bring charges for the death of Mrs. Heath's fetus due to the lack of a feticide statute.10 Other articles referred to Heath as a "con man." His parents and brother told the press that they thought he was possessed by the devil. Heath claims the coverage became more prejudicial following his plea of guilty. The parents of the victim told how they were upset that Heath was not sentenced to death. Several letters to the editor attacked the plea agreement. The public's reaction to the plea agreement eventually forced the prosecutor and the judge to make statements defending their actions. In the course of this outcry against the plea agreement, Alabama indicted Heath and sought his extradition for capital murder. The media reported that the Alabama trial was necessary because of Georgia's failure to seek the death penalty.

In addition, Heath complains of several sideshows that were extensively covered in the media. First, he notes that the Troup County, Georgia family court held open hearings in the custody battle between his parents and his in-laws over his son and again in the later fight to strip Heath of all of his parental rights. Both family court hearings quickly degenerated into mud slinging. Second, he notes the extensive attention that was focused on Denise Lambert's award in a beauty pageant, her reputed family connections and wealth, and the judicial system's supposed lenient treatment of her.

Heath argues that this pretrial publicity infected the venire and the jury. He notes that 93% of the venire members (75 of 81) heard about the case, 70% of the venire members (57 of 81) knew that Heath had entered a plea of guilty or was convicted in Georgia. He also notes that 83% of the actual jurors (10 of 12) knew, prior to being summoned for jury duty, that Heath's guilt had been determined in Georgia. And, Heath notes, the rest of the jurors were told of the guilty plea during voir dire.

Heath argues that the pretrial publicity deprived him of a fair trial under both an inherent prejudice and an actual prejudice analysis. Inherent prejudice occurs when the pretrial publicity "is sufficiently prejudicial and inflammatory and the prejudicial pretrial publicity saturated the community where the trials were held." Coleman v. Kemp, 778 F.2d 1487, 1490 (11th Cir.1985). Inherent prejudice rarely occurs. Id. Our Circuit has treated the inherent prejudice analysis as a mixed question of fact and law, Coleman, 778 F.2d at 1537 n. 17, but we review the record cognizant of the fact that inherent prejudice is rarely found. Id. Even if Heath cannot show inherent prejudice, he may be able to prove that the pretrial publicity resulted in an unfair trial if he is able to show actual prejudice. Actual prejudice occurs when the prejudice actually enters the jury box and affects the jurors. Coleman v. Kemp, 778 F.2d 1487, 1490 (11th Cir.1985). This Court should look at the totality of circumstances to determine the extent of the prejudice. Id. The petitioner cannot establish actual prejudice without proving that at least one juror should have been dismissed for cause. Bundy v. Dugger, 850 F.2d 1402, 1427 (11th Cir.1988).

Heath claims that he can demonstrate inherent prejudice because he claims that he meets both the inflammatory and the saturation requirements of the inherent prejudice analysis. We disagree and hold that he has not met either prong of the inherent prejudice analysis.

We are guided by some standards to evaluate whether news coverage is allegedly inflammatory. For instance, the courts have distinguished, and deemed acceptable, any pretrial publicity which is purely factual in nature, see Patton v. Yount, 467 U.S. 1025, 104 S.Ct. 2885, 81 L.Ed.2d 847 (1984), from pretrial publicity which includes prejudicial or inflammatory commentary. See Coleman, supra. The courts are also extremely wary when the media saturates a town with highly prejudicial evidence such as inadmissible evidence, see Coleman, supra, or the defendant's confession, see Rideau v. Louisiana, 373 U.S. 723, 83 S.Ct. 1417, 10 L.Ed.2d 663 (1963). This highly prejudicial evidence increases the risk of a trial by the media to the exclusion of a fair trial where the rules of evidence and cross-examination govern the admission and weight of the evidence that is ultimately placed before the jury. As we recently noted in a related context, the Fourteenth Amendment incorporates the Sixth Amendment right of a defendant to be tried in "a fair trial in which the jury reaches its verdict based only on the evidence subjected to the crucible of the adversarial process." Woods v. Dugger, 923 F.2d 1454 (11th Cir.1991).

We find that the record does not support Heath's claim that the news coverage was inflammatory. The vast majority of the articles were purely factual presentations of the news about the killing. In addition, Heath is unable to cite any articles in which the media discussed highly prejudicial evidence in a manner that raises concerns about potential jurors being influenced by information not subjected to the "crucible of the adversarial process." While Heath is able to find several articles which are inflammatory (e.g., the articles about Heath being possessed by the devil) these articles are not typical of the vast majority of the articles. Moreover, Heath's claims that the press expressed outrage at his plea bargain and that the media later supported Alabama's later prosecution are not an accurate reflection of the record. Despite isolated letters-to-the-editor expressing outrage, the general tone of the media's reporting is quite restrained.11

We find against Heath for a second, independent reason. Heath has not demonstrated that the media coverage saturated the market. A petitioner must prove (1) that a substantial number of the people in the relevant community could have been exposed to some of the prejudicial media coverage, and (2) that the effects of the media saturation continued until the trial. We have, on at least one occasion, rejected an inherent prejudice claim because of a failure of proof. See Mayola v. Alabama, 623 F.2d 992 (5th Cir.1980) (holding that the record was unclear how widely read the Birmingham News and the Post-Herald were in Blount County, Alabama in 1962). Heath has met the first portion of the saturation requirement. Heath has produced evidence that Columbus, Georgia and Phenix City, Alabama are a single media market and that the relevant television stations and local newspapers cover the entire area. Heath has also produced evidence that each of the major local media sources covered the murder and ensuing investigation equally thoroughly. However, Heath has failed to meet the second portion of the saturation requirement. Courts have recognized that in certain limited circumstances, the effect of inflammatory pretrial publicity may be lessened if the media focuses on other news in the months prior to the trial. See Patton v. Yount, supra. But see Coleman, supra (declining to apply the "cooling off" rationale when there was evidence that the local emotions continued to run deep and the "cooling off" period was marred by intense publicity immediately before the trial). The record in this case supports such a cooling off rationale. The crime occurred on August 31, 1981 and Heath was arrested four days later. In the next several weeks, the story was placed on the front page of the two local papers nearly every day. Heath pled guilty in February of 1982, and his co-defendants went to trial during the spring. Throughout this period, the case seemed to command the local news media's attention. In May of 1982, days after the final trial finished in Georgia, Alabama brought charges against Heath. The case stayed in the news until the end of May. However, from June of 1982 until the trial began in February of 1983, the case received little or no attention in the news media. On average, there were only one or two articles in the local papers each month. Moreover, in the days prior to the trial, the media mentioned the upcoming trial in only one or two short articles. Therefore, in order for Heath to prevail on this claim, he must show that the articles published prior to June of 1982 resulted in an unfair trial held eight months later. Heath is unable to show that these articles, which were, for the most part, not inflammatory, marred his trial sufficiently for us to conclude that this is one of those extremely rare cases in which we should find inherent prejudice.

Finally, Heath alleges that the publicity created actual prejudice. He notes that a large number of veniremembers heard about the crime and his plea agreement. According to Patton v. Yount, however, the question is not whether a juror remembered the case from the media. Patton, 467 U.S. at 1033, 104 S.Ct. at 2889. Rather, the Patton Court held that after examining the voir dire, the court should determine whether the jurors had fixed opinions to the extent that they were unable to judge the defendant's guilt impartially and recommend an appropriate sentence. Similarly, this Court in Bundy, supra, recognized that seating of a juror by a state court over a challenge for cause will be affirmed as long as there is fair support in the record for the court's decision. As we have already noted, all of the seated jurors denied having a fixed opinion as to Heath's guilt and they all promised to follow the judge's instructions.

Therefore, under Cross, 893 F.2d at 1290, we do not find appellate counsel's failure to bring this claim on appeal prejudicial because the claim would not have been successful if it were brought on appeal.

b. The Violation of the Privilege Against Self Incrimination

During the Georgia trial of Owens and Lumpkin, the Georgia district attorney called Heath to the stand.12 Heath invoked his Fifth Amendment right not to testify. The district attorney argued that Heath was not entitled to the protection of the Fifth Amendment because he had already pled guilty to the murder and Georgia was not going to bring any other charges against him for this crime. Heath disagreed; he argued to the court that he was entitled to the protection of the Fifth Amendment because there was a possibility that Alabama might bring charges against him for the same crime. During Heath's Alabama trial, the district attorney, over a defense objection, asked Officer Malueg, one of the investigating officers who was present when Heath testified in Georgia, about Heath's testimony during Owens' and Lumpkin's trial in Georgia. Officer Malueg testified that Heath "had refused to testify based on the fact ... that there was an implication that other charges might be filed at a later time because of the fact that his wife was abducted from--and taken from Alabama." In his closing argument to the jury, the district attorney mentioned Officer Malueg's testimony in order to prove that the kidnapping had occurred in Alabama.

We do not need to determine whether Heath's Fifth Amendment rights were violated because it is clear that any error was harmless beyond a reasonable doubt. Both parties admit that Heath's testimony was relevant only to establish that the kidnapping started in Heath's home in Russell County, Alabama. We note that there is extensive evidence supporting the state's contention that the kidnapping started in Russell County. The most important piece of evidence was Heath's extensive confession to Officer Malueg on the night of his arrest during which he admitted that the kidnapping started in his home. Moreover, some circumstantial evidence confirmed that the kidnapping started in Heath's home. The investigating officers found a crushed cigarette on Heath's kitchen floor, and Heath admitted that neither he nor his wife smoked. Also, on the morning of the kidnapping, two neighbors noticed Heath speaking with some men in a car parked down the street from Heath's house. These men were seated in a car that was seen later that morning speeding away from the wrecked car containing Mrs. Heath's body. Furthermore, the police found tire tracks leading up to the Heath house, they found a door unlocked, they found a fan blowing in Mrs. Heath's bedroom, and they found that Mrs. Heath had not worn some jewelry that she customarily wore when she left the house. The prosecutor, in his closing, emphasized each of these pieces of evidence. In light of all this evidence, we conclude that the error was harmless beyond a reasonable doubt. Because the error was harmless, the Fifth Amendment claim would not have been successful on appeal. We therefore hold that appellate counsel's failure to raise this claim was not prejudicial.

B. Subject Matter Jurisdiction