Petitioner Paul House appeals from the district court's denial of a writ of habeas corpus, 28 U.S.C. § 2254. A Tennessee jury found House guilty of the murder of a neighbor, Carolyn Muncey, and sentenced him to death.

This court granted a certificate of appealability as to all issues. However, House has limited his brief to a discussion of only two claims: 1) Whether the manner in which the Tennessee courts applied the state law doctrine of waiver during House's post-conviction proceedings constitutes an adequate and independent state procedural bar to his ineffective assistance of counsel claims; and 2) assuming that the Tennessee courts properly deemed House's claims to be waived, whether that waiver should be excused on the grounds that House has established his actual innocence under Schlup v. Delo, 513 U.S. 298, 115 S. Ct. 851, 130 L. Ed. 2d 808 (1995). After the Tennessee Supreme Court declined a request by an en banc panel of this court to answer certified questions relating to issues of state law, House v. Bell, 311 F.3d 767 (6th Cir.2002), this court is again faced with the same claims.

Having considered the arguments of the parties regarding the two claims that are before us, we affirm the district court's denial of the writ for the reasons set forth below.

This court reviews a district court's legal conclusions in a habeas proceeding de novo and its factual findings for clear error. See Lucas v. O'Dea, 179 F.3d 412, 416 (6th Cir.1999). House initiated this habeas action on September 30, 1996; the petition was amended on September 16, 1997. Consequently, this court's review of the state court's decision is governed by the standards set forth in the Antiterrorism and Effective Death Penalty Act of 1996, Pub. L. No. 104-132, 110 Stat. 1214 (1996) ("AEDPA"). See Lindh v. Murphy, 521 U.S. 320, 336, 117 S. Ct. 2059, 138 L. Ed. 2d 481 (1997); Harpster v. Ohio, 128 F.3d 322, 326 (6th Cir.1997).

Because factual determinations by state courts are entitled to a presumption of correctness, 28 U.S.C. § 2254(e)(1), we will describe the factual circumstances surrounding the murder for which House was convicted by quoting from the Tennessee Supreme Court's opinion denying him relief in his direct appeal:

The victim of the homicide was Mrs. Carolyn Muncey, who lived with her husband and two young children on Ridgecrest Road in rural Union County, Tennessee. Mrs. Muncey was in her late twenties, and her children were about eight and ten years old at the time of her death on July 13, 1985.

In March 1985 appellant Paul Gregory House was released from a prison in Utah and moved to the rural community in which the Muncey family lived. There he resided with his mother and step-father for several weeks, but in June he moved into a trailer occupied by his girl friend, Donna Turner, which was located about two miles from the Muncey home. Appellant did not own an automobile; but he was permitted to drive his mother's car from time to time, and he also drove Ms. Turner's car on some occasions.

Other than doing occasional farm work for his stepfather, appellant does not appear to have been regularly employed. He did not testify at trial at either the guilt phase or the sentencing hearing. He was shown to have had one prior conviction for aggravated sexual assault, a charge to which he pled guilty on March 16, 1981 in Salt Lake County, Utah. Apparently he was placed on parole in that state, and supervision of his parole was transferred to Tennessee when he returned to this state. He was approximately twenty-three years old at the time of the homicide in this case.

Mrs. Muncey disappeared from her home in the late evening of Saturday, July 13, 1985. Her badly beaten body was found on the following afternoon at about 3 p.m., lying partially concealed in a brush pile about 100 yards from her home. Apparently the husband of the victim was not at home during the early part of the evening of July 13. Mrs. Muncey and her children visited a neighbor and left at about 9:30 p.m. to return to their home. Later the older child, Laura, awoke. She testified that she heard a voice which sounded like her grandfather making inquiry about her father. She also heard someone tell her mother that her father had been in a wreck near the creek. She heard her mother sobbing or crying as she left the house. When her mother did not return, the two children went to look for her at neighboring homes. Not finding her, they returned home and waited until their father arrived. Discovering that his wife was missing, he took the children back to the home of the neighbor where they had visited earlier in the evening and then called for members of his family to look for his wife.

When the body of Mrs. Muncey was discovered the next afternoon, she was dressed in her nightgown, housecoat and underclothing. Her body was badly bruised, and there were abrasions and blood giving every evidence that she had been in a fierce struggle. Apparently a severe blow to her left forehead had caused her death. It appeared, however, that she had also been partially strangled. A pathologist testified that the blow to her left forehead caused a concussion and hemorrhage to the right side of the brain from which she died, probably one or two hours after being struck. He testified that she probably would have been unconscious after having been struck. He estimated the time of her death at between 9 p.m. to 11 p.m. on Saturday, July 13, but emphasized that this was at best a rough estimate.

Appellant never confessed to any part in the homicide, and the testimony linking him to it was circumstantial. There was evidence showing that he knew Mr. and Mrs. Muncey and had been with them socially on a few occasions. Through defense proof there was testimony that Mrs. Muncey and her husband had been having marital difficulties and that she had been contemplating leaving him. There was no evidence to indicate that the appellant was aware of that situation, however, or that there had been any previous romantic or sexual relationship between him and the victim.

On the afternoon of Sunday, July 14, 1985, two witnesses saw the appellant emerge from a creek bank at the side of Ridgecrest Road at the site where Mrs. Muncey's body was later found concealed in the underbrush. He was wiping his hands with a dark cloth and was walking toward a white Plymouth automobile, parked on the opposite side of the road, belonging to his girl friend Donna Turner. The two witnesses spoke briefly to appellant, all of them discussing the fact that Mrs. Muncey had disappeared. Later the two witnesses became suspicious of what they had observed and returned to the point where they had seen appellant emerge from the embankment. Looking down the bank, they found the partially concealed body of Mrs. Muncey. They promptly notified the sheriff.

Appellant later admitted that he had been in the area but denied that he had seen the body of Mrs. Muncey or had any knowledge of its presence. The dark rag which he had been using when first seen was never produced. It was the theory of the State, however, that this was a dark "tank top" or jersey which appellant was shown to have been wearing on the previous evening, July 13.

Appellant gave two statements to investigating officers in which he denied being involved in the homicide. In both of these statements he stated that he had been at Ms. Turner's trailer the entire evening of July 13 and that he had not left until the next afternoon when he went to look for Hubert Muncey after learning of the disappearance of the latter's wife.

On Sunday afternoon various witnesses observed that appellant had numerous scratches and bruises on his arms, hands and body, there being an especially significant bruise on the knuckle of his right ring finger. Appellant explained that these injuries had been sustained innocently earlier during the week, but when Ms. Turner was called as a witness, she said that she had not observed them prior to the evening of July 13. Appellant also told investigators that he was wearing the same clothes on Sunday, July 14 as he had been wearing the previous evening. It was later discovered, however, that a pair of blue jeans which he had been wearing on the night of the murder was concealed in the bottom of the clothes hamper at Ms. Turner's trailer. These trousers were bloodstained, and scientific evidence revealed that the stains were human blood having characteristics consistent with the blood of Mrs. Muncey and inconsistent with appellant's own blood. Scientific tests also showed that fibers from these trousers were consistent with fibers found on the clothing of the victim. There were also found on her nightgown and underclothing some spots of semen stain from a male secretor of the same general type as appellant.

Some of the most damaging evidence against appellant was given by his girl friend, Ms. Turner. She at first told investigators that he had not left the trailer during the course of the evening of July 13. Later, however, she modified this testimony to state that he had been in the trailer until about 10:45 p.m. at which time he left to take a walk. When he returned an hour or so later, he was panting, hot and exhausted. He was no longer wearing either his blue jersey or his tennis shoes. The shoes were later found in an area different from the place where appellant told her he had lost them.

Appellant told Ms. Turner that he had thrown away the navy blue tank top because it had been torn when he was assaulted by some persons who tried to kill him. It was after the appellant's return to the trailer that Ms. Turner first noticed the bruises and abrasions on his hands referred to previously.

Appellant's mother testified that he had not used her automobile on Saturday evening. She testified that during Saturday and Sunday she had been planning to separate from appellant's stepfather and that appellant had been assisting her in her preparations for moving.

At the sentencing hearing the State proved appellant's prior conviction for aggravated sexual assault. Appellant's parents testified that he came from a broken home and had been subjected to stress as a result of that experience. Appellant's mother also testified that in the interval between the guilt trial and the sentencing hearing appellant had attempted suicide. She read into evidence a letter which he had written to her denying his involvement in the homicide. Apparently he had cut his wrists while in the jail awaiting the sentencing hearing, but the degree and extent of the injuries were not detailed in evidence. They do not appear to have been serious and did not prevent his attending the sentencing hearing.

Although the evidence against appellant was circumstantial, it was quite strong. Particularly incriminating was the testimony that he had emerged from an embankment where the body was found, wiping his hands on a dark cloth, without disclosing to anyone the presence of the body. Damaging also were the discovery of his bloodstained trousers and the testimony of Ms. Turner, which a trier of fact could have found sufficient to demolish his alibi and to demonstrate that he had been in a heavy struggle near the time when the homicide must have occurred.1 A classic case for determination by a jury was presented, and the evidence clearly is sufficient to support the conviction.

Following the sentencing hearing, the jury imposed the death penalty. In their verdict they found that the State had established three aggravating circumstances, these being: (1) appellant had previously been convicted of a felony involving the use or threat of violence to the person; (2) the homicide was especially heinous, atrocious, or cruel; and (3) it was committed while appellant was committing or attempting to commit or fleeing from the commission of rape or kidnapping. T.C.A. §§ 39-2-203(i)(2), (5), and (7).

There was ample evidence to support all of these findings and to support the conclusion of the jury that no mitigating circumstances had been established which would outweigh the aggravating circumstances. Certainly the sentence of death was not disproportionate to that imposed in other cases in view of the violent and brutal nature of the homicide shown in this record.

State v. House, 743 S.W.2d 141, 142-44 (Tenn.1987) (footnote in original).

After his conviction and sentence were handed down in February 1986, House took a direct appeal to the Tennessee Supreme Court. The Tennessee Supreme Court affirmed in the opinion just cited. House failed to seek a writ of certiorari in a timely manner.

Post-conviction proceedings began in February 1988 with the filing of a pro se petition for post-conviction relief in the trial court. This petition, amended after appointment of counsel, was denied by the trial court on November 29, 1988. Although claims of ineffective assistance of trial counsel were submitted to the court, they were not argued, nor was trial counsel called as a witness at the hearing. Only a single issue was taken to the Tennessee Court of Criminal Appeals: whether a jury instruction improperly misled jurors into believing that they must unanimously find mitigating circumstances. The Court of Criminal Appeals affirmed the conviction on December 15, 1989. The Tennessee Supreme Court denied leave to appeal and the United States Supreme Court denied certiorari.

A second petition for post-conviction relief was filed on December 14, 1990. After conducting hearings, the trial court denied relief on the ground urged by the State: that all the issues presented had either been previously determined in the first petition or, if not, had been waived.2

The Court of Criminal Appeals affirmed the trial court's decision on September 2, 1992. However, the Tennessee Supreme Court remanded for reconsideration in light of one of its recent opinions, which it later withdrew. The Court of Criminal Appeals remanded the matter to the trial court for further consideration of the waiver issue. This remand was averted, however, when the Tennessee Supreme Court re-instated the trial court's initial denial of the second petition for post-conviction relief. House v. State, 911 S.W.2d 705 (Tenn.1995). Critical for this habeas action is the Court's holding concerning waiver:

We conclude that a "full and fair hearing" sufficient to support a finding of previous determination occurs if a petitioner is given the opportunity to present proof and argument on the petition for post-conviction relief. We further conclude that the rebuttable presumption of waiver is not overcome by an allegation that the petitioner did not personally and therefore, "knowingly and understandingly," waive a ground for relief. Instead, waiver is to be determined by an objective standard under which a petitioner is bound by the action or inaction of his attorney. Finally, we conclude that there is no right to effective assistance of counsel in post-conviction proceedings, and therefore, an allegation of ineffective assistance of prior post-conviction counsel does not preclude application of the defenses of waiver and previous determination.

Id. at 714. Because of its holding, the Tennessee Supreme Court re-instated the trial court's original denial of relief.

House filed a pro se habeas petition on September 30, 1996, which was eventually amended after the district court granted in forma pauperis status and appointed counsel. The district court granted the State's motion for summary judgment on the majority of claims in an order entered June 25, 1998. In February 1999, it conducted an evidentiary hearing on House's claim that the procedural default of his ineffective assistance of counsel claims was excused because he could establish his actual innocence. After considering post-hearing briefs of counsel, the district court denied habeas relief. It also denied a certificate of appealability, a denial which was later superseded by this court's grant of a certificate as to all issues.

Although a certificate of appealability had been granted as to all issues, House has raised only two issues in his briefs to this court: 1) whether the Tennessee state courts applied the state waiver rule in such a way that it constituted an adequate and independent state ground that procedurally defaulted his ineffective assistance of counsel claims; and 2) whether, if his ineffectiveness claims were procedurally defaulted, he had made a sufficient showing of his actual innocence of the crime, permitting him to revive them. On March 11, 2002, a three-judge panel of this court issued an opinion rejecting House's claims and affirming the district court's denial of the writ. House v. Bell, 283 F.3d 737 (6th Cir.2002) (withdrawn). That opinion was vacated when a majority of the active judges of the court voted to rehear the case en banc. On November 22, 2002, following oral argument, the en banc court issued an opinion certifying questions of state law to the Tennessee Supreme Court. House v. Bell, 311 F.3d 767 (6th Cir.2002) (en banc), cert. denied, 539 U.S. 937, 123 S. Ct. 2575, 156 L. Ed. 2d 622 (2003). The Tennessee Supreme Court declined to answer the questions. House v. Bell, No. M2003-01952-SC-S23-CQ (Tenn. Nov. 24, 2003). Following the issuance of that order, we requested additional briefing and argument from the parties prior to rendering this decision.

The first issue House has raised on appeal involves his contention that both the Tennessee courts and the district court erred when they concluded that his claims of ineffective assistance of counsel were procedurally barred.

Initially, the district court deferred granting summary judgment to the State on the claims of ineffective assistance of counsel because "these claims are intertwined with the petitioner's claim of actual innocence." Memorandum Opinion, June 25, 1998, at 41. It went on to provide the following rationale:

[T]he court will be unable to determine whether the claims should be barred on the ground of procedural default until the conclusion of the evidentiary hearing. Bousley v. United States, 523 U.S. 614, 118 S. Ct. 1604, 1611, 140 L. Ed. 2d 828 (1998) ("Where a defendant has procedurally defaulted a claim by failing to raise it on direct review, the claim may be raised in habeas only if the defendant can first demonstrate either `cause' and actual `prejudice,' or that he is `actually innocent.'") Accordingly, in addition to petitioner's claim of actual innocence, the court will consider [the ineffective assistance of counsel claims].

Id. at 41-42. However, after conducting an evidentiary hearing and determining that House had failed to establish actual innocence, the district court held that House's ineffective assistance of counsel claims were "barred on the ground of procedural default." Memorandum Opinion, February 16, 2000, at 46-47.

House concedes that an adequate and independent state-law ground can procedurally bar subsequent habeas claims. This circuit has developed a four-part analysis to determine whether a claim is barred:

Under Maupin v. Smith, 785 F.2d 135, 138 (6th Cir.1986), this circuit utilizes the following four-part analysis when the state argues that a federal habeas claim has been procedurally defaulted in state court: (1) whether there is a procedural rule that is applicable to the petitioner's claim and whether the petitioner failed to follow the rule; (2) whether the state courts actually enforced the state procedural rule; (3) whether the state procedural rule is an adequate and independent state ground to foreclose federal relief; and if so (4) whether the petitioner has established cause for his failure to follow the rule and prejudice by the alleged constitutional error.

White v. Schotten, 201 F.3d 743, 749 (6th Cir.2000).

House points to the tortured path that his case took in the courts of Tennessee and argues that no hard and fast rule existed to procedurally bar his pro se ineffective assistance of counsel claims. Specifically, he points to the following paragraph of the opinion of the Tennessee Supreme Court:

Our research has revealed no reported Tennessee case dealing directly with the issue of the appropriate standard to apply when determining whether an issue has been waived. Courts in other states have split on whether to apply a subjective or objective standard and provide us little assistance because their decisions were based largely on the particular state statutory procedure. However, Tennessee cases dealing generally with the concept of waiver in the post-conviction context apply an objective standard and impute the conduct of counsel to their clients. See, e.g., Caruthers v. State, 814 S.W.2d 64, 70 (Tenn.Crim. App.1991); State v. Bishop, 731 S.W.2d 552 (Tenn.Crim.App.1986).

House, 911 S.W.2d at 713 (emphasis added) (footnotes omitted). In House's view, this opinion illustrates that no rule had been clearly established.

We do not write on a clean slate with respect to this issue. See Cone v. Bell, 243 F.3d 961 (6th Cir.2001), rev'd on other grounds, 535 U.S. 685, 122 S. Ct. 1843, 152 L. Ed. 2d 914 (2002); see also Coe v. Bell, 161 F.3d 320 (6th Cir.1998). In Cone, this court made clear that the House opinion that House relies upon merely explained existing state law. The petitioner in that case contended that he had not personally waived a challenge to a jury instruction. In rejecting his claim, we explained:

First, we are aware of two cases in which courts have considered whether a petitioner is bound by his attorney's waiver of a constitutional claim, Coe, 161 F.3d 320, and House v. State, 911 S.W.2d 705 (Tenn.1995). The House court stated that "[w]aiver in the post-conviction context is to be determined by an objective standard under which a petitioner is bound by the action or inaction of his attorney." House, 911 S.W.2d at 714. House does not appear to announce a new standard, as Cone suggests. Rather, it seems merely to affirm Tennessee's standard of review.

In Coe, as we explained earlier, this court held that the petitioner had procedurally defaulted his state claim that the trial court failed to give a correct malice instruction. He presented the claim for the first time in his second petition for post-conviction relief rather than his first petition, as a consequence of which the Tennessee Court of Criminal Appeals found it had been procedurally waived. Coe, 161 F.3d at 329-31. This court cited House when determining that Coe had defaulted his claim under an "objective" standard of waiver. However, the petition upon which the court relied in finding the default was filed before House was decided. Thus, concerning defaults that occurred before House was decided, the Tennessee courts have strictly and regularly applied the traditional standard of waiver, whether the waiver is made by counsel or the petitioner personally.

....

Last, we are not persuaded that Cone is correct in his claim that Tennessee law was in a state of confusion on whether an "objective" or "subjective" standard of waiver is appropriate. It is not clear from the Tennessee cases that procedural default may not be charged to a petitioner who has not himself "knowingly and understandingly" waived timely assertion of a federal constitutional claim when his attorney has done so. We are satisfied that Tennessee follows the traditional rule that a petitioner is chargeable with his attorney's failure to timely assert a claim and with the consequences of failing to do so.

Cone, 243 F.3d at 974. We find this reasoning dispositive of the issue before us and accordingly conclude that the district court was correct when it determined that House's ineffective assistance of counsel claims had been procedurally defaulted.

We now turn to House's other claim. House argues that even if his ineffective assistance of counsel claims have been procedurally defaulted, he has established his actual innocence of the crime for which he was convicted, a showing which, if made, revives his ineffectiveness claims. In Schlup v. Delo, 513 U.S. 298, 115 S. Ct. 851, 130 L. Ed. 2d 808 (1995), the Supreme Court held that a petitioner must show either cause and prejudice or a miscarriage of justice in order to obtain habeas review of an otherwise procedurally defaulted claim. House seeks to invoke the miscarriage of justice exception here. With respect to a miscarriage of justice, a petitioner must demonstrate that "a constitutional violation has probably resulted in the conviction of one who is actually innocent of the crime." Schlup, 513 U.S. at 324, 115 S. Ct. 851. The Court explained the nature of this exception as follows:

Claims of actual innocence pose less of a threat to scarce judicial resources and to principles of finality and comity than do claims that focus solely on the erroneous imposition of the death penalty. Though challenges to the propriety of imposing a sentence of death are routinely asserted in capital cases, experience has taught us that a substantial claim that constitutional error has caused the conviction of an innocent person is extremely rare. To be credible, such a claim requires petitioner to support his allegations of constitutional error with new reliable evidence — whether it be exculpatory scientific evidence, trustworthy eyewitness accounts, or critical physical evidence — that was not presented at trial. Because such evidence is obviously unavailable in the vast majority of cases, claims of actual innocence are rarely successful. Even under the pre-Sawyer regime, "in virtually every case, the allegation of actual innocence has been summarily rejected." The threat to judicial resources, finality, and comity posed by claims of actual innocence is thus significantly less than that posed by claims relating only to sentencing.

Of greater importance, the individual interest in avoiding injustice is most compelling in the context of actual innocence. The quintessential miscarriage of justice is the execution of a person who is entirely innocent.

Id., 513 U.S. at 324-25, 115 S. Ct. 851 (footnote and citation omitted). The Court cautioned that this exception is rare and should be applied only in the extraordinary case, concluding that, "[t]o establish the requisite probability [that a petitioner is actually innocent], the petitioner must show that it is more likely than not that no reasonable juror would have convicted him in the light of the new evidence." Id., 513 U.S. at 327, 115 S. Ct. 851.

Because the district court conducted an evidentiary hearing on this issue, there is testimony about events beyond that which was presented during the original trial. House contends that this new evidence is sufficient to establish his actual innocence. We will summarize that evidence and the district court's response to it before explaining why, in our view, House has failed to show that it is more likely than not that no reasonable juror would have convicted him.

House testified for the first time at the evidentiary hearing. He offered this version of the night of murder: "I went for a walk. I got jumped, ran around, came back." He went on to explain that the terrain was hilly and that it was dark:

I had only been walking about maybe 20 minutes at the most it seems like. A truck pulled up behind me with, I remember it as being like a 4-wheel drive. It sat up high, you know. Headlights were on. It had lights across the roof of the cab and they were on. I couldn't see anything other than that about the truck.... I turned around and I kept walking. I believe there were at least two guys in the truck. I know the driver got out on his side, one guy got out on the passenger side. I couldn't really discern it, but I think there was one other guy in the cab. The driver came up. I can remember he said something, but I don't even know if I heard him correctly at the time. He grabbed me by the arm.

He started to jerk me around. I turned around and threw him back with my left hand. I hit him. He let go. I started running. I ran kind of diagonally across the road into some trees, bushes, whatever it was. I heard a shot, at least one. There might have been two. I am not sure. I ran around through those woods for a while. I don't know how long. When I came back out — I believe at one point I ran to the right, once I got into the woods, and started heading back.... I went back across the road up to Donna's house. When I got, I stepped on something, a sharp rock or something. I knew I stepped on it. When I looked down I only had one shoe. I lost one of them while I was running. I took the other one off and threw it across the road.

....

I didn't even notice my shirt was gone until I got up to the trailer.

As the factual summary of the Tennessee Supreme Court attests, this version of events is relatively consistent with the one presented by Donna Turner during the trial. When asked why he initially lied to investigators by telling them he had not left the trailer at all, House responded, "I was on parole. I didn't want to draw attention to myself."

In short, House's testimony merely restates a scenario presented to the jury that had convicted him. Moreover, the district court, which had the opportunity to assess House's demeanor, found his testimony to be less than credible.

During the evidentiary hearing, House devoted considerable time to the trial testimony of Billy Ray Hensley, the witness who saw House near the spot where the body was discovered. Specifically, House introduced maps and photographs in order to show that Hensley could not have seen what he purported to see from the place where he claimed to have been.

At trial, Hensley testified that his wife received a telephone call at about 2 p.m. on July 14 that Mrs. Muncey was missing. He drove to the Muncey trailer and talked to some of the family members. He then "went to check on [his] tobacco." After visiting his fields, he ended up on Bear Hollow Road, which was near where the Munceys lived.

He then had an encounter with House, which he described in these terms:

[J]ust before I rounded the curve of Ridgecrest, whatever the name of that road is, I saw Mr. House come out from under a bank, wiping his hands on a black rag. And I went on down to Little Hube's3 driveway. I pulled up in the driveway where I could see up toward Little Hube's house and I seen Little Hube's car wasn't there, and I backed back out in the road, and come back towards to the Dump Road, that is what I call it. And that is when Mr. House flagged me down....

The two men had a short conversation about the fact that Mrs. Muncey was missing. House was driving his girlfriend's white Plymouth.

After leaving, Hensley became suspicious and, along with his friend Jackie Adkins, returned to the spot where he thought House had emerged: "I said — right along here is where I saw [House], and I got out and was looking off the bank, and [Adkins] got out and walked around the car and he said — oh, my God."

On cross-examination, Hensley conceded that he could not have seen House while he was actually "down in the embankment." He would have first seen him at the top of the bank. Defense counsel also tried to bring out some inconsistencies in Hensley's statements concerning precisely when and where he first saw House. On re-direct, the prosecution attempted to have Hensley clarify:

Q. Let me ask you if this is a true statement — "I traveled about 500 feet on Ridgecrest Road when I saw a '66 or '67 white Plymouth sitting on the left-hand side of Ridgecrest Road," is that true?

A. That's true.

Q. Is that where you saw the car?

A. Yes, sir.

Q. Is this true? "I saw a man later identified to me as Paul G. House enter the roadway from the right-hand side of the road?"

A. He was walking toward the road, yes.

Q. All right. "And he was coming up over the bank and he had a rag in his hand and he was wiping his hands," is that true?

A. That's true.

....

Q. The estimation [of the distance on the road] that you have given, that you were pressed for, is that an accurate measurement or is that an estimate on your part?

A. That is just an estimate.

The exhibits introduced by House during the evidentiary hearing were designed to show that Hensley could not have seen House coming up the embankment. However, even if we accept House's contention that Hensley could not have seen him until he emerged onto the road, it is undisputed that House was seen in the general vicinity of the body carrying a black rag. Moreover, trial counsel effectively cross-examined Hensley regarding his inconsistent statements about when and where he saw House. Thus, in our view, House's attack on Hensley's testimony advances his cause little, if at all.

In addition to presenting his own version of events while attempting to cast doubt on the accuracy of Hensley's testimony, House takes aim at the physical evidence that linked him to the crime.

Dr. Alex Carabia performed the autopsy of Mrs. Muncey's body. He testified at the trial that death was caused by a blow to the left side of her head. Mrs. Muncey died about an hour and a half after she was hit.

At trial, photographs of the bruises on House's body were entered into evidence and three witnesses testified about his physical condition. Prior to his arrest, House provided various accounts of their origin, attributing them to a mysterious fight on the night of the murder and to tearing down a shed a few days earlier. During closing argument, the prosecution emphasized these inconsistent statements.

Also at trial, FBI Agent Paul Bigbee testified that the blood samples taken from the victim were degraded. Nonetheless, the blood found on the jeans was consistent with that of the victim and not with that of House.

At the evidentiary hearing, House mounted a concerted challenge to this evidence.

Four vials of blood were taken from the victim during the autopsy. These were placed in a styrofoam container, which was sent from the Tennessee Bureau of Investigation (TBI) to the FBI. House referred to two demonstrative exhibits in the district court: photographs of the styrofoam container viewed from above and from the side. The container was sealed with tape by the TBI in both directions for shipping. The photograph of the side view shows that one of the seals was broken and then resealed by a second layer of tape. FBI Agent Bigbee placed his lab number on the second layer. The first layer of tape is incomplete; it only covers the lid of the container. Agent Bigbee conceded that it was possible that the first seal had been cut before the second seal had been placed over it. To support the theory that the container was opened between the time it left the TBI and arrived at the FBI, House points to the fact that the label on the container indicated that it held both blood and vaginal secretions. Yet, Agent Bigbee received the secretions separately in a manila envelope.

As mentioned above, four vials of blood were sent to the FBI. According to Agent Bigbee, he would have used one fourth of a vial in testing. House's trial serology expert, Howard Bragdon, took a photograph when he received the styrofoam container from the FBI. In House's view, the photograph shows that one of the vials was only one-half full and another was nearly empty. Furthermore, despite Agent Bigbee's testimony to the contrary, it appeared that some of the blood had spilled, although there is no evidence indicating that the spillage had occurred before the FBI received the blood.

At the evidentiary hearing before the district court, Dr. Cleland Blake, Assistant Chief Medical Examiner for the State, examined the results of the FBI tests of the blood found on the blue jeans and also the blood taken from Mrs. Muncey's body at the autopsy. He theorized, based upon the degree of the enzymatic degradation, that the blood on the blue jeans came from known samples, such as the blood contained in the vials, and not from Mrs. Muncey's body. When confronted with this conclusion, Agent Bigbee was doubtful, noting that the extent of enzymatic degradation could vary greatly from specimen to specimen taken from the same source depending upon the manner in which the specimens were handled after being extracted from the source, and upon other individual circumstances. To bolster Blake's conclusion, House asserted that the locations of the blood stains on the jeans were unlikely to have been caused by a struggle between House and the victim. The five stains were found on the outside left leg, on the inside left thigh, on the inside right pocket, outside the right pocket, and on the right cuff, respectively.

An expert on blood spatter analysis, Paulette Sutton, also testified at the evidentiary hearing. She testified, contradicting House's assertion, that the pattern of some of the blood spots on House's jeans was consistent with transfer stains resulting from blood being wiped onto them, that some of the stain patterns demarcated folds, and that the rest of the stains were consistent with spatter. She also noted, however, that some of the blood stains on the jeans were mixed with mud, although the photographs of the crime scene showed no mud present. Furthermore, National Weather Service records show that it had not rained for three days prior to the murder. Finally, there was no mud on the victim's nightgown.

Finally, House notes that no blood was found on the tennis shoes that he was wearing on the night of the murder. Charles Burks, House's trial attorney, also testified at the evidentiary hearing. The district court summarized his testimony on the issue of the tennis shoes as follows:

During the evidentiary hearing, Mr. Burks reviewed a report from the Forensic Services Department of the TBI, which referred to Mr. House's tennis shoes and indicated the absence of blood on the tennis shoes. Mr. Burks did not recall having seen that report before. That would have been relevant to Mr. Burks because, although there was blood on Mr. House's jeans near the cuff, there was no blood on the shoes he was wearing at the time of the offense. The shoes were found in the general area of Ms. Turner's trailer long after the murder.

Memorandum Opinion, February 16, 2000 at 25.

The court recounted the discovery of the blue jeans in these terms:

TBI Agent Charles Scott testified that he became involved in the investigation of Mrs. Muncey's murder, at the request of another agent, on the second day of the investigation. He took a statement from Mr. House and obtained consent to search from Ms. Turner. He went to Ms. Turner's home and seized a pair of blue jeans from the clothes hamper in the bathroom. The jeans had "reddish brown" stains that he suspected was [sic] blood on the upper part of the jeans and near the cuff; there was also some light colored mud that was not completely dry.

Mr. Scott did not thoroughly examine the jeans at that time but rather folded them and put them in a paper bag. Mr. Scott did not recall ever seeing the jeans in a plastic bag.

Memorandum Opinion, February 16, 2000, at 23.

The court then summarized the testimony of House's expert witnesses. Larry Johnson testified as an expert in crime scene investigation and opined that "the packaging of materials in the case did not meet professional standards" because items were not wrapped separately. DNA expert Lisa Calandro eliminated House as the donor of the semen found on Mrs. Muncey's underwear and nightgown.

Howard Bragdon testified for House as well. Bragdon was the manager of laboratory operations for DCI Laboratory in Nashville. This laboratory had performed the blood analysis for House at trial. Bragdon noted that he took possession of the blue jeans, the victim's clothes and fingernail scrapings, as well as blood from both the victim and House, on October 29, 1985. The next day, after transporting them to Nashville, he took pictures that showed dried blood around the upper left corner of the box in which the items had been contained. According to the district court, "Mr. Bragdon admitted that it was his custom to inspect the condition of serological evidence when he took possession and that there was no notation on the receipt of spillage. Mr. Bragdon also admitted that he had no way of knowing the condition of the blood samples at the time of the FBI's serological testing." Memorandum Opinion, February 16, 2000, at 35.

The court characterized Dr. Blake's testimony as follows:

Dr. Blake testified that there was no total chain of custody. Also, according to Dr. Blake, Dr. Carabia [the coroner] failed to refrigerate and preserve the blood in the tubes, failed to seal the tubes of blood, which could result in spillage, and failed to package the items individually.

Dr. Blake's testimony was based upon his review of photographs of the physical evidence, and was relevant to Dr. Blake's opinion that the blood on petitioner's blue jeans resulted from spillage of Mrs. Muncey's blood in the laboratory tubes...

....

Dr. Blake testified that if fresh blood had spilled on the blue jeans while Mrs. Muncey was alive, and then dried, the enzymes on the jeans would not have deteriorated to the same extent as the enzymes in the blood taken from Mrs. Muncey. From this, Dr. Blake concluded the blood was not spilled on the jeans but rather came from the spillage of the test tubes.

Dr. Blake also testified with respect to the age of Mr. House's bruises, based upon photographs taken from the state court record. Dr. Blake estimated some bruises at one to two days old; others at five to six days old. Also, in Dr. Blake's opinion, the bruising on Mr. House's right ring finger was an injury from being mashed; it was not consistent with striking someone.

Memorandum Opinion, February 16, 2000, at 37-38.

The district court also summarized the testimony of the blood spatter expert, Ms. Sutton, and noted that her testimony contradicted the theory that blood had spilled from the vials onto the blue jeans "because the blood and mud [with which it had mixed] would have had to have spilled at the same time." Memorandum Opinion, February 16, 2000, at 42.

With respect to the blood, the court determined:

Without question, one or more tubes of Mrs. Muncey's blood spilled at some time. It is likely the spillage occurred prior to the receipt of the evidence by [the] laboratory hired by Mr. House's trial attorney. Based upon the evidence introduced during the evidentiary hearing, however, the court concludes that the spillage occurred after the FBI crime laboratory received and tested the evidence.

... [T]he enzyme deterioration, as well as Mr. Muncey's alleged confession and the blood spillage, does not negate the fact that Agent Scott saw what appeared to be bloodstains on Mr. House's blue jeans when the jeans were removed from the laundry hamper at Ms. Turner's trailer and that the blood was in fact from Mrs. Muncey.

Memorandum Opinion, February 16, 2000, at 45-46.

As indicated in the passage above, House not only presented evidence to the district court that undermined the case against him, he also offered an alternative theory of the crime: that Mr. Muncey killed his wife.

At the evidentiary hearing, House produced witnesses who testified about Mr. Muncey's alcoholism and also his physical abuse of his wife. One acquaintance, Kathy Parker, testified that "[Mrs. Muncey] was constantly with black eyes and busted mouth." A friend, Hazel Miller, testified that Mr. Muncey told her that he was going to get rid of his wife a few months before her death. In the district court, Mr. Muncey acknowledged that he "smacked" his wife at least once.

As for the day of the murder, Mr. Muncey was supposed to have helped to dig a grave. He had gone over to his father's place, helped to work on some cars, and then dug the grave. However, rather than go home, he decided to attend the weekly dance at the C & C Recreation Center, where, according to his testimony, he stayed until midnight. When he arrived home, he found his wife missing.

Kathy Parker told the district court that Mr. Muncey visited her on a Friday in 1985 after the murder. Friends were sitting around drinking when Mr. Muncey "started crying and going on and rambling off." According to Parker, he was "[t]alking about what happened to his wife and how it happened and he didn't mean to do it, but I don't know exactly what all was said." She went on, "[H]e said they had been into an argument and he slapped her and she fell and hit her head and it killed her and he didn't mean for it to happen." According to Parker, Mr. Muncey was drunk when he made this confession. After hearing it, Parker claimed, "I freaked out and run him off." The next day Parker's mother took her to the courthouse to tell someone about the confession. However, she "never did really get to talk to anybody." When the district court asked her about the motivation behind her testimony in the evidentiary hearing, she replied, "An innocent man is in jail."

On cross-examination, Parker testified that she had tried to come forward but no one seemed interested. She had had seven or eight beers on the night of the confession.

Parker's sister, Penny Letner, also testified to having heard such a confession from Little Hube. Once again, she recalled that he was "pretty well blistered." According to Letner, Mr. Muncey confessed to killing his wife when he returned home:

He said he didn't mean to do it. That she was "bitching him out" because he didn't take her fishing that night, that he went to the dance instead. He said when he come home that she was still on him pretty heavily "bitching him out" again and that he smacked her and that she fell and hit her head. He said I didn't mean to do it, but I had to get rid of her, because I didn't want to be charged with murder.

Letner had not been drinking. She was frightened by the talk and left the party. As a young mother of 19, she testified that she had been too scared to report the confession.

Based upon the statements of Letner and Parker, House posits the following scenario:

When Mr. Muncey got home, he and his wife resumed their fight. He hit her at least once and she fell. When he checked, he found that he had killed her. He took her body down by a creek running near their home and hid it with some brush and branches.

Whether Mr. Muncey ever went back to the dance is uncertain. Constable Wallace, who was providing security at the dance, testified that he never saw Mr. Muncey return after he left around 10:30 p.m. Mr. Muncey claimed during the hearing that he never left the dance until it broke up some two or more hours later.

Petitioner's Brief at 33-34. House also points out that Dennis Wallace4 did not think that Mr. Muncey seemed upset when he reported his wife's disappearance or when the body was recovered. Also, the morning after the murder, Mr. Muncey asked a neighbor, Artie Lawson, to tell people that he was at the dance. Since she had not attended it herself, Ms. Lawson declined. Her daughter, Mary Atkins, testified that she not only saw Little Hube at the dance, but that she saw him hit his wife.

The district court discounted the testimony of Letner and Parker, finding their testimony lacking in credibility:

The court is not impressed with the allegations of individuals who wait over ten years to come forward with their evidence. This is especially true when there was no physical evidence in the Munceys' kitchen to corroborate his alleged confession that he killed her there.

Memorandum Opinion, February 16, 2000, at 45. Instead, the court credited the testimony of Laura Muncey Tharp, the victim's daughter, who testified at both the trial and evidentiary hearing:

While in bed, she heard a deep voice saying her dad had been involved in a wreck next to the creek. Sometime later she and her brother got up and went looking for their mother. They went to the neighbors and looked up and down the driveway. She did not see anything out of the ordinary in the house; nothing was out of place and there was no sign of a struggle.

According to Ms. Tharp, her parents got along fine. They argued, but she did not recall any physical pushing or hitting. If they argued, she could hear them if she was in her bedroom. The family did not have air conditioning in the home. She did not hear any arguments that night.

The court found Ms. Tharp a very credible witness. She had no reason to lie. Her testimony during the evidentiary hearing was consistent with her trial testimony.

Memorandum Opinion, February 16, 2000, at 12-13.

As the preceding recitation makes clear, House has mounted a concerted attack on his conviction. Indeed, it is fair to say that he has presented a colorable claim of actual innocence. However, as the Supreme Court has made clear, that is not the standard that we are bound to apply. To prevail, House must do more than raise questions about the reliability of portions of trial testimony or the manner in which physical evidence was handled or analyzed; he must show "that it is more likely than not that no reasonable juror would have convicted him in the light of the new evidence." Schlup, 513 U.S. at 327, 115 S. Ct. 851. Moreover, in weighing the new evidence we review the factual findings of the district court for clear error. Campbell v. Coyle, 260 F.3d 531, 539 (6th Cir.2001).

The following facts that implicate House are undisputed: he lied to investigators about his whereabouts on the night of the murder; he gave inconsistent versions of the origins of the scratches and bruises on his hands and arms; he was seen near where the body was discovered on the day after the murder; he lied about what he was wearing on the night of the murder; blue jeans belonging to House, spattered with blood mixed with mud, were found at the bottom of Ms. Turner's laundry hamper; House has a deep voice and Laura Muncey testified that the man who came to the trailer on the night of the murder had a deep voice; and, according to Ms. Sutton, the blood and mud found together on House's blue jeans had been mixed together, which "certainly eliminates the possibility of any stains being created by contamination in an evidence container." We note that the fact that mud may not have been present at the crime scene, and may have been scarce in the surrounding area, cannot be taken as proof that there was no mud anywhere on the route between Ms. Turner's trailer and the scene of the crime.

With respect to House's theory that Mr. Muncey committed the murder, we defer to the finding of the district court that Ms. Letner and Ms. Parker, who allegedly heard Mr. Muncey's confession, were not credible. Furthermore, the content of Ms. Letner's testimony, indicating that Mr. Muncey killed his wife upon returning to the trailer, is belied by the presence of the children in the trailer, who heard no such confrontation, and the lack of any signs of a struggle. House's theory that a deep laceration cutting across Mrs. Muncey's head was caused when she fell and hit her head is inconsistent with the testimony of Dr. Carabia, who indicated that the laceration could only have resulted from a violent blow. The fact that Mr. Muncey may have asked his neighbor to say that she saw him at the dance during the time of the murder is insufficient to tip the balance in favor of House's theory.

Regarding House's attacks on the scientific evidence that incriminated him, he has succeeded in showing that the semen attributed to him during the trial was that of Mr. Muncey and that, at some point, the blood evidence appears to have been mishandled, resulting in spillage. However, the fact that the semen found on the victim's clothing came from her husband and not from House does not contradict the evidence that tends to demonstrate that he killed her after journeying to her home and luring her from her trailer, nor does the lack of any physical evidence of sexual contact contradict the notion that the murderer lured Mrs. Muncey from her home with a sexual motive. As for the mishandling of the blood evidence, the theory that the blood on House's jeans came from the vials of blood gathered at Mrs. Muncey's autopsy is based upon a speculative theory regarding enzyme degradation that was contradicted by other testimony in the record, and an analysis of the blood stain pattern does not demonstrate that the stains could not have resulted from Mrs. Muncey's murder. The lack of any blood spatter on House's shoes is inconclusive as well, because it is not clear when House took his shoes off. Finally, the district court's conclusion that "the spillage occurred after the FBI crime laboratory received and tested the evidence" cannot be characterized as clearly erroneous. The only unchallenged blood evidence, the testimony indicating that the blood and mud on the jeans were mixed, tends to support the conclusion that House committed the murder.

Despite his best efforts, the case against House remains strong. We therefore conclude that he has fallen short of showing, as he must, that it is more likely than not that no reasonable juror would have convicted him in light of the new evidence.

All of the issues before us having been decided, the judgment of the district court is affirmed.

MERRITT, Circuit Judge, dissenting.

I regard this as the rare or extraordinary case in which the petitioner through newly discovered evidence has established his actual innocence of both the death sentence and underlying homicide. The Court's opinion, like the Attorney General's argument for the State, regards as "undisputed" old evidence and inferences that are now contradicted by other evidence in the case. It fails to describe adequately the persuasive case of actual innocence that the petitioner's newly discovered evidence raises. Nor does it adequately describe the legal standards to be applied.

This dissent will first describe the constitutional standards applicable in "actual innocence" cases such as this one. There are four actual innocence theories applicable in this case based on the body of newly discovered evidence presented at the federal habeas hearing: (1) a "free-standing," substantive, so-called "Herrera" actual innocence claim, (2) a procedural, or "gateway," so-called "Schlup" actual innocence claim, (3) the more limited "gateway" claim that the petitioner is "actually innocent of the death penalty" because the new body of evidence shows that petitioner is not now eligible for the death penalty because the rape aggravator has been disproved, and (4) a free-standing, substantive "actual innocent of the death penalty" claim. It will then outline the evidence in detail and apply the standards in four parts, as follows:

First, at the state trial in Maynardville, Union County, Tennessee, in 1986, and in its argument to uphold the death verdict in the Tennessee Supreme Court, the State relied on rape as the motive for the kidnapping and the murder of Carolyn Muncey. There was no other motive offered. It relied on a semen specimen on her nightgown as proof that House tried to rape her. Newly discovered DNA evidence now conclusively establishes that the semen was her husband's. There is now absolutely no evidence of sexual assault. The new evidence disproves the motive the jury accepted as the basis for the kidnapping and murder and the aggravating circumstances the jury found as its basis for the death penalty.

Second, besides the semen evidence, the State introduced at the trial one other piece of highly incriminating scientific evidence: evidence of Carolyn Muncey's blood on House's blue jeans worn on the night of the murder. At the 1999 federal habeas hearing, the State's case was undermined by the State's own medical examiner, Dr. Cleland Blake. As "Consultant in Forensic Pathology" for the Tennessee Bureau of Investigation for 22 years, Dr. Blake has testified for the prosecution in the past in hundreds of cases. Four vials of blood were extracted at the time Carolyn Muncey was autopsied. Dr. Blake, the State's medical examiner, testified at length that he had no doubt that the blood on House's pants was spilled from one of these four vials of blood shipped to the lab by local law enforcement agents — spilled either accidentally or intentionally. There is no explanation besides spillage for the fact that one of the four vials of blood was empty. The new body of evidence shows conclusively that the vials of blood were not properly handled and shipped by law enforcement and that the blood that spilled from the vials cannot otherwise be accounted for.

Third, testimony from five new witnesses offered at the habeas hearing implicates Mr. Muncey in his wife's murder. The new evidence discloses that Mr. Muncey, with a flood of tears, confessed to two women friends after the murder that he had killed his wife. He told a third woman that he was going to "get rid" of Carolyn a few weeks before the murder. He asked a fourth woman to provide him with an alibi on the night of the murder and gave testimony about his whereabouts that night at the time of the murder that has now been contradicted by a local law enforcement officer. The State offered no evidence that any of these witnesses was biased in favor of House or prejudiced against Mr. Muncey.

Fourth, the evidence completely undermines the reliability of the testimony of Billy Ray Hensley, the witness who said that on Sunday afternoon before the victim's body was found, he saw House coming up the embankment on Ridgecrest Road where the body was later found that day. Based on his own testimony and an examination of the record, it would have been impossible for Hensley to see House as he claimed.

Union County, Tennessee, is a small rural county in the hill country of East Tennessee. It has a population of 12,000. Maynardville, population 1,000, is the county seat where Carolyn Muncey's murder was investigated and tried. Local law enforcement officials from the Sheriff's office testified that immediately after the murder they had two suspects, House and the victim's husband, Hubert Muncey, called "Little Hube." Mr. Muncey grew up and was well-known in the local community. Although "Little Hube" had a history of severe domestic abuse, the local police chose House as the murder suspect when he told them two days after the body was found that he had just recently moved into the local Luttrell community and that he had a sexual assault conviction in Utah and after they had developed other incriminating evidence.

In its opinion in 1987, affirming House's conviction, the Supreme Court of Tennessee noted the semen evidence and rape as the motive for the homicide. It noted that House had "never confessed to any part in the homicide, and the testimony linking him to it was circumstantial." The Court also observed that the Munceys "had been having marital difficulties and that she had been contemplating leaving him." State v. House, 743 S.W.2d 141, 143 (Tenn.1987). The case comes down to the question of whether the newly discovered evidence undermining the case against House and incriminating Mr. Muncey is sufficiently strong — despite the uncertainties that remain — to preclude a rational juror from finding guilt beyond a reasonable doubt and to make the execution of House "constitutionally intolerable."

Schlup v. Delo, 513 U.S. 298, 115 S. Ct. 851, 130 L. Ed. 2d 808 (1995), is the only Supreme Court case that has discussed and compared the standards to be applied in three of the four different types of actual innocence claims that may be asserted in habeas: (1) "free standing," (2) "gateway" and (3) "innocent of the death penalty" gateway claims. There, in an opinion for six members of the Court, Justice Stevens wrote that it is "firmly established in our legal system, that the line between innocence and guilt is drawn with reference to a reasonable doubt," and that "the analysis [of actual innocence claims] must incorporate the understanding that proof beyond a reasonable doubt marks the boundary between guilt and innocence." Id. at 328, 115 S. Ct. 851. This starting point is a major factor for all types of actual innocence claims. For such claims, this factual analysis is always a "probabilistic determination" about the behavior of a reasonably instructed juror. Id. at 329, 115 S. Ct. 851.

1. Gateway, Actual Innocence Claims. — In Schlup, 513 U.S. at 316, 115 S. Ct. 851, the Court stated that for substantive, free-standing claims where the "conviction was the product of a fair trial," the "evidence of innocence would have had to be strong enough to make an execution `constitutionally intolerable.'" Yet, when a claim of innocence is coupled with an assertion of constitutional error at trial, the "conviction may not be entitled to the same degree of respect." Id. Thus, the "evidence of innocence need carry less of a burden." Id. In Schlup, as in House's gateway claim in the instant case, the claim of constitutional error at the original trial was ineffective assistance of counsel; and the actual innocence claim was used as a "gateway" to overcome procedural default in order to reinstate the ineffective assistance claim and render it again cognizable. For such gateway or procedural claims, the Schlup opinion holds that a petitioner must demonstrate that "it is more likely than not that no reasonable juror would have found petitioner guilty beyond a reasonable doubt." Id. at 327, 115 S. Ct. 851.

2. Gateway Innocent of the Death Penalty Claims. — The Court in Schlup discussed and left intact the standard for claims of "innocent of the death penalty" (claims that the evidence of aggravating circumstances at the sentencing phase of the case is insufficient to render the defendant eligible for the death penalty) — a standard announced in the previous case of Sawyer v. Whitley, 505 U.S. 333, 112 S. Ct. 2514, 120 L. Ed. 2d 269 (1992). In such "innocent of the death penalty" claims governed by Sawyer, the question is always whether the petitioner's proof of innocence is sufficient to overcome a procedural default and render a defaulted constitutional claim again cognizable. Quoting Sawyer, the Court said that claims of actual innocence of the death penalty at the sentencing phase "must focus on those elements which render the defendant guilty of the death penalty" and show "by clear and convincing evidence" that "no reasonable juror would have found the petitioner eligible for the death penalty" beyond a reasonable doubt. Schlup, 513 U.S. at 323, 115 S. Ct. 851. This is the standard applicable to the question of innocence of the rape-homicide aggravator found by the jury at House's trial. If found, it would make House's ineffective assistance of counsel claim at the penalty phase of the case cognizable.

3. Free-standing Actual Innocence Claims. — In Herrera v. Collins, 506 U.S. 390, 113 S. Ct. 853, 122 L. Ed. 2d 203 (1993), the Supreme Court did not spell out a standard for free-standing substantive claims of actual innocence in cases which presuppose a fair trial. Instead, it merely noted that the required "threshold showing" would be "extraordinarily high." 506 U.S. at 417, 113 S. Ct. 853. In a concurring opinion, Justice White sets out a very demanding standard:

I assume that a persuasive showing of "actual innocence" made after trial, even though made after the expiration of the time provided by the law for the presentation of newly discovered evidence would render unconstitutional the execution of petitioner in this case. To be entitled to relief, however, petitioner would at the very least be required to show that based on proffered newly discovered evidence and the entire record before the jury that convicted him, "no rational trier of fact could [find] proof of guilt beyond a reasonable doubt." Id. at 429, 113 S. Ct. 853 (White, J., concurring) (quoting Jackson v. Virginia, 443 U.S. 307, 324, 99 S. Ct. 2781, 61 L. Ed. 2d 560 (1979)) (emphasis added).1 As laid out below, the White standard is appropriate to apply in free-standing actual innocence claims. Justice White's proposed standard borrows language from Jackson, in which the Court established the test governing habeas review of claims of insufficient evidence. In Jackson, the Court found that due process guaranteed by the Fourteenth Amendment requires "that no person shall be made to suffer the onus of a criminal conviction except upon sufficient proof — defined as evidence necessary to convince a trier of fact beyond a reasonable doubt of the existence of the element of the offense," 443 U.S. at 316, 99 S. Ct. 2781. When petitioners make Jackson claims, they assert that the trial leading to their conviction was flawed in that the evidence of guilt presented in the trial record was insufficient to support the verdict.

With free-standing actual innocence claims, petitioners are not claiming that their trial violated constitutional due process requirements due to insufficient evidence. Instead Herrera claims are to provide relief to one who faces the death penalty and can make a truly persuasive case of innocence. In order to prevent what the Supreme Court describes as "constitutionally intolerable," Schlup, 513 U.S. at 316, 115 S. Ct. 851, courts must consider newly discovered evidence in addition to the trial record. The proof established at the original trial must be examined in light of what is newly proffered to determine if all the available evidence would support conviction. The White standard is appropriate for use in free-standing actual innocence cases as it is a very high standard that will only be met in exceptional cases where new evidence has unquestionably undermined the case against the petitioner. It also forces courts to make an objective determination regarding what a fact finder could find rather than what a hypothetical juror would find.

Utilizing Justice White's standard would permit courts to adequately address the rare case of actual innocence without becoming bogged down by non-meritorious claims. Justice O'Connor in Herrera noted the Court's concern that if the standard for free-standing actual innocence cases is set too low, "federal courts will be deluged with frivolous claims of actual innocence." 506 U.S. at 426, 113 S. Ct. 853 (O'Connor, J., concurring). The White standard borrowed from Jackson addresses this concern as new evidence would have to so undermine the case against the petitioner as to make the conviction untenable and the imposition of the death penalty constitutionally intolerable. The Court has noted that the Jackson standard is an incredibly high hurdle for a petitioner to jump. See Schlup, 513 U.S. at 323 n. 38, 115 S. Ct. 851 ("Even the high standard of proof set forth in Sawyer falls short of the Jackson standard governing habeas review of claims of insufficiency of evidence."). Under the proposed standard, "the mere existence of sufficient evidence to convict would be determinative of a petitioner's claim." Id. at 330, 99 S. Ct. 2781.

The inquiry as to whether any rational juror could convict also seeks to make the determination of actual innocence more objective. As the Supreme Court wrote, describing the standard, "the use of the word `could' focuses the inquiry on the power of the trier of fact to reach its conclusion," whereas "the use of the word `would' focuses the inquiry on the likely behavior of the trier of fact." Id. Whether a rational fact finder would have the ability to convict the petitioner in light of all available, reliable, and relevant evidence "requires a binary response." Id. Either the evidence is sufficient to support the conviction or it is not. The analysis thus interferes with the original trial determination only to the extent necessary to ensure that a prisoner is not executed when the available evidence cannot support his conviction. This Court should adopt Justice White's standard from Herrera, and the writ should issue if in light of newly discovered evidence, no rational trier of fact could find proof of guilt beyond a reasonable doubt.

4. Free-standing Innocent of the Death Penalty Claims. — Although not discussed in Schlup or other cases, for cases raising a substantive, non-gateway claim that the condemned prisoner is not eligible for the death penalty at sentencing, the standard should be the same as the White standard. The petitioner must demonstrate no rational fact finder examining the trial record and the new evidence could impose the death penalty. If he can disprove the aggravator for which he was sentenced to death, it is plainly unconstitutional for the sentence to stand.

In my view, as I will now explain, the petitioner has carried his burden of proof on all four theories. He has clearly shown both that a reasonable juror would have a substantive and serious doubt as to his guilt and that the cumulative evidence is insufficient to enable a rational juror to convict House or sentence him to death for the murder of Carolyn Muncey.

II. Rape as House's Motive — the Semen Evidence

In its opinion, the Tennessee Supreme Court in its factual statement recounted evidence of rape as the motive for the kidnapping and murder noting that "there were found on her nightgown and underclothing some spots of semen stain from a male secretor of the same general type as appellant," State v. House, 743 S.W.2d 141, 143 (Tenn.1987), and referring to inferences available to the jury from the fact that "the state proved appellant's prior conviction for aggravated sexual assault." Id. at 144.

In light of the evidence introduced at the federal habeas hearing, the State now concedes that the semen specimen found on the nightgown Carolyn was wearing when murdered was "Little Hube's," not House's; and there is no other evidence of rape or sexual assault. The State's concession is based upon DNA analysis introduced at the habeas proceeding by counsel for House. But the State now seems to deny that the semen evidence was introduced at the state trial in 1986 to show that House attempted to rape Carolyn or that rape was the motive offered to the jury for this kidnapping and murder and the basis of the jury's verdict. This argument is simply not consistent with the facts, and the majority opinion does not acknowledge or deal with this problem. The State has no other explanation or evidentiary basis for introducing the semen evidence at the trial other than to show attempted rape as the motive. Because the Court's opinion does not deal with the problem and the State now denies what it strongly claimed when it tried the case and obtained the death verdict, we will itemize in some detail the evidence that contradicts the State's present argument that the prosecution did not rely on rape as the motive for the murder and for the death penalty aggravator.

From the beginning in July 1985, the police and prosecutors viewed this as a rape murder case and so advised the FBI lab and then the jury. In the various lab reports from the FBI to the local prosecutors, the correspondence is always designated as "rape-homicide." The first lab report described the crime as "rape-homicide" two weeks after the murder:

                                                           July 29, 1985

To: Mr. Williams Paul Phillips                             FEDERAL EXPRESS

    District Attorney General
    Post Office Box 10
    Huntsville, Tennessee 37756        FBI File No.

                                       LAB NO.        50717040 S XI VB VJ
                                                      50722020 S XI VB
                                       YOUR NO.

Re: PAUL GREGORY HOUSE—SUSPECT;
    CAROLYN MUNCEY—VICTIM;
    RAPE—HOMICIDE

Examination requested by:  Addressee and Sheriff of Union County,
                           Maynardville, Tennessee

Reference:                 Letter from District Attorney General dated
                           July 16, 1985, and letter from Sheriff of Union
                           County dated July 19, 1985

Examination requested:     Chemical Analyses—Microscopic Analyses—
                           Mineralogy

FBI File, Criminal Trial Exhibits, Addendum No. 2, Exhs. 31 & 32, admitted into evidence, Addendum No. 4, p. 903.2

Four such lab reports designating House as the "suspect" and "rape-homicide" as the nature of the case were shown to the jury at House's trial on February 6, 1986, after being admitted into evidence as State's exhibit 32. The jury read a letter from Union County District Attorney William Paul Phillips dated July 16, 1985, to the FBI lab in Washington stating that this was a "sexually motivated attack." Id. at p. 6. Then the jury saw four FBI lab reports like the one above referring to House as the "suspect," "Carolyn Muncey—victim," and the nature of the case described as "rape-homicide." Id. at pp. 4, 10, 12, 34. The FBI lab reports analyze the semen based on the presupposition that it was "House's" semen. On each of the lab reports placed before the jury the words "FEDERAL BUREAU OF INVESTIGATION, UNITED STATES DEPARTMENT OF JUSTICE" are prominently displayed at the top of the page just before "rape-homicide" is identified as the nature of the case and House as the "suspect." Id.

The semen evidence referred to in these reports was placed before the jury and came in at the state trial through the testimony of long-time FBI expert Agent Paul D. Bigbee, who specialized in analyzing "physical evidence in criminal cases for the presence of blood and other body fluids." Testimony of Paul Bigbee, Trial Transcript, Addendum No. 4, Vol. VI, p. 883. He testified at length about the semen evidence. Here are a few examples of his testimony on this subject:

Q All right, did you find semen on any of her items of clothing?

A Yes sir, I did.

Q Where did you find the semen?

A I found semen in the panties and on the gown.

....

Q Now, can you tell me, can you detect semen stains at times in determining things about the blood type of the person who emitted or the blood type of the person that semen is from?

A Yes, sir. If the person who deposited the semen is a secretor, which means eighty (80) percent of the population secrete the ABO blood group substances in other body fluids, such as semen and saliva, if that person is a secretor, then the ABO blood type can be determined from the semen.

Q How about Mr. House, is he a secretor?

A Yes sir, he is.

....

Q Can you tell me where, can you hold the gown up and show the jury where on the gown you found semen?

(WITNESS COMPLIES WITH REQUEST)

A If you look on the front of the gown, at the very bottom here, you can see one cutting I took that has the number and letter, 1S, that is the location where I found the semen.

Q On the front of the gown?

A Front bottom portion.

....

Q You may put that back now. What can you tell us about the blood type, if you can tell us anything, of the person who that semen came from, the male that that semen came from?

A The person who deposited that semen was a blood type A.

Q Now, what blood type is Mr. House?

A He is blood type A.

....

Q [T]he semen on the nightgown could be from Mr. Muncey?

A It could be, however, I was not able to determine his secretor status.

Q And he is not definitely a secretor?

A I don't know whether he is or not. I could not determine that.

Q But you know that Mr. House definitely is a secretor?

A Yes, sir.

Q And only if a person is a secretor, can you determine the blood type from the semen, is that right?

A That is correct.

Id. at pp. 896-900.

In the State's argument to the jury at the end of the guilt phase of the trial, the District Attorney told the jury that rape was House's motive in the following words:

Now you may have an idea why he did it. The evidence at the scene which seemed to suggest that he was subjecting the lady to some kind of indignity, why would you get a lady out of her house, late at night, in her night clothes, under the trick that her husband has had a wreck down by the creek? Why is it that you want to get her down by the creek?.... It is either to keep her from telling what you have done to her, or it is that you are trying to get her to do something that she nor any mother on that road would want to do with Mr. House .... and you kill her because of her resistance.

Closing Argument, Trial Transcript, Addendum No. 4, Vol. IX, pp. 1302-03.

After the jury returned its verdict at the guilt phase of the trial, the District Attorney advised the court in preparation for the sentencing phase of the trial that the State's theory requesting a death sentence is based on "the fact that there was semen on the outer garment, that is the nightgown," and "we say that the jury could well conclude that this murder occurred in the process of another crime, that being in the process of either rape or attempted rape." Trial Transcript, Addendum No. 4, Vol. X, p. 1381. The prosecutor opened his argument at the sentencing phase by telling the jury:

The defendant, the proof indicates, has in the past been convicted of aggravated sexual assault which is something that the judge will instruct you that you should consider.... We also think the proof shows strong evidence of attempted sexual molestation of the victim to accompany the taking away and murdering her.

Id. at pp. 1410-11. The prosecutor then tied the prior conviction for sexual assault and this attempted rape together arguing in favor of the death penalty because House has "been through the process before and having been convicted of a crime involving ... aggravated sexual assault," he "cannot benefit from the type of rehabilitation that correction departments can provide." Id. at p. 1413. Then at the end of his impassioned argument at the sentencing phase, the District Attorney returned again to this theme asking the question, "did he commit this crime while he was engaged in an attempt at rape?" Id. at 1444. And then he answered that question in the affirmative.

As soon as the District Attorney sat down, the court charged the jury that the defendant "was previously convicted of" sexual assault and then defined rape: "I will charge you that rape is the unlawful, carnal knowledge of a woman forcefully and against her will. Carnal knowledge is accomplished by the commencement of the sexual connection and proof of emission is not required." Id. at p. 1447. The court again for a second time said in charging the jury that they may sentence the defendant to death if "the murder was committed while the defendant was engaged in committing or was attempting to commit or was fleeing after committing or attempting to commit rape," and defined rape again as "the unlawful carnal knowledge of a woman, forcibly and against her will." Id. at p. 1448. The jury then returned its verdict of death saying that its reason for sentencing the defendant to death was that "the murder was committed while the defendant was engaged in committing or was attempting to commit or was fleeing after committing or attempting to commit rape or kidnapping." Id. at p. 1455.

When House appealed to the Tennessee Supreme Court, the State in its brief was not hesitant to argue rape as the motive for the kidnapping and the murder and the basis of the jury's death sentence. In its brief, the State argued that the murder was committed while the defendant was engaged in attempting to commit rape and that "[s]emen was found on the victim's gown and panties and was consistent with the defendant's blood type." Brief of the State of Tennessee, S.Ct. No. 2, Union County, filed Apr. 29, 1987, Addendum No. 6, p. 49. It is against this background that the Tennessee Supreme Court cited the prosecution's claim that the evidence proved that "on her nightgown and underclothing some spots of semen stain from a male secretor of the same general type as appellant." State v. House, 743 S.W.2d at 143. The Tennessee Supreme Court upheld House's conviction and the imposition of the death penalty based upon rape as the motive for the kidnapping and murder. Id. at 142. Even at the federal habeas corpus hearing 13 years later, Assistant Attorney General Glenn Pruden for the State, said that at the trial "an aggravating factor was found that this murder was committed in the attempt to commit, to perpetrate a rape and the perpetration of the kidnapping." House v. Bell, No. 3-96-CV-883, Transcript of Habeas Hearing, p. 91 (E.D.Tenn. Feb. 1, 1999).3 There can be no doubt that the State claimed rape as the motive from the beginning and throughout the trial, and that the jury so found and that the Tennessee Supreme Court approved the verdict on that basis.

The newly discovered evidence conclusively removes rape as the motive and eliminates rape as the aggravating circumstance that made the defendant eligible for the death penalty. Without any evidence of rape, the State has lost its motive, its theory of the case and the aggravating circumstance on which the State and the jury relied for its death verdict. House has met the standard for actual innocence of the death penalty by conclusively disproving the aggravator that made him eligible for the death penalty. No rational, reasonable juror could now find that House raped or attempted to rape Mrs. Muncey. Without any evidence of rape, the court would now have to direct a verdict in favor of House on this issue.

The State introduced at trial one other piece of highly incriminating scientific evidence: evidence of Carolyn Muncey's blood on the blue jeans worn by House on the night of the murder. Agent Paul Bigbee, the agent from the FBI laboratory that did the actual testing on the blue jeans in July 1985, shortly after the murder, testified that his testing demonstrated a clear match between the blood on the jeans and Mrs. Muncey's blood. At the 1999 federal habeas hearing, however, the State's case was undermined by its own medical examiner, Dr. Cleland Blake. Dr. Blake, the Assistant State Chief Medical Examiner, has been a consultant in forensic pathology for the Tennessee Bureau of Investigation for 22 years and has testified for the State in hundreds of cases. Dr. Blake testified at length that, based on enzymatic testing results performed by the FBI, he had no doubt that the blood on House's blue jeans was spilled, either accidentally or intentionally, from the vials of blood taken from Mrs. Muncey's body during her autopsy:

Dr. Blake: I conclude that the deteriorated blood which was in the jeans came out of the test tube [because] both [the blood on the jeans and the blood in the vials] had lost the enzyme.

...

The Court: So in your opinion then the blood on the jeans and the blood in the tube are one in the same.

Dr. Blake: That is my opinion.

Testimony of Dr. Cleland Blake, Transcript of Habeas Hearing, pp. 116-17, Feb. 2, 1999 (emphasis added). When we add to this testimony the fact that almost all of the blood in one of the vials was spilled and cannot be accounted for, as explained below, Dr. Blake's conclusion that the blood evidence does not now support the jury verdict leaves little room for debate.

Dr. Blake arrived at his conclusion based on his theory of what constitutes normal or expected enzymatic degradation of blood in different media; that is, whether the blood is preserved in liquid form in test tubes or vials or whether it is dried on cloth or other material. Depending on the medium on which the blood is preserved, and the condition or state of the blood at the time it is preserved, the blood will degrade at different rates. Dr. Blake found an unexpected consistency in enzyme degradation between the liquid sample taken from Mrs. Muncey's body after her death and the five samples of what the government contends is fresh blood that stained House's blue jeans as he allegedly murdered Mrs. Muncey. Simplistically, the blood enzymes at issue here are proteins that are able to function properly only within the living human body. Once outside the body, or after the body is no longer functioning, the enzymes start to break down and lose their structure and shape and no longer function; they degrade and are said to become "denatured." Generally, "fresh" blood from a live victim that comes in contact with dry cloth is "preserved" in a different, and generally better, way than blood that is stored in a test tube.4

The blood tests performed by Agent Bigbee of the FBI on the victim's blood, stored in vials after being taken from her body during the autopsy, and the five samples from the defendant's jeans, preserved on the cloth of the blue jeans, looked for the presence of various enzymes (to see whether they had denatured or not) in the blood. Of the ten enzymes tested, six had conclusive results. All six enzymes matched, making it reasonable for Dr. Blake to conclude that the enzymes in the blood on the jeans degraded at the same rate as the blood in the sample taken from the victim's body and stored in glass vials or test tubes. Because they degraded at the same rate despite being "preserved" in two different media—test tube vials and cotton cloth of blue jeans—it appears that the enzymes came from the same sample, that is, the blood on the defendant's jeans came from the sample taken from the autopsy of the victim, not the victim herself when alive.

There is additional strong evidence that the sample in the vials of blood were mishandled or possibly even tampered with and intentionally spilled on the jeans. Four vials of blood were taken from the victim during her autopsy. The vials, along with other evidence, were transported by law enforcement officials between Union County, Tennessee, and the FBI Laboratory in Washington, D.C. and back to Union County. In reaching his conclusion that the blood on House's blue jeans came from the vials of blood taken from Mrs. Muncey after her death, Dr. Blake also noted that the blood evidence had been mishandled in this matter. Although not an expert on the packaging and handling of evidence, Dr. Blake's work requires him to examine evidence that has been packaged and handled by others on a regular basis. He determined in this case that the evidence had been mishandled and that blood had spilled from the vials, causing the contamination of the blue jeans. Testimony of Cleland Blake, Transcript of Habeas Hearing, pp. 50-51, Feb. 2, 1999. Dr. Blake testified that if tubes containing blood are not properly sealed, the stoppers on the tubes may come out if the blood samples become too warm. Testimony of Cleland Blake, Transcript of Habeas Hearing, pp. 51, Feb. 2, 1999.

Detective Larry Johnson, of the Knox County Sheriff's Department, also testified at the habeas hearing as to the packaging and handling of the evidence in this case. Detective Johnson is an expert witness in the area of crime scene investigation, having examined hundreds of homicides, robberies, rapes and other crimes over 30 years. Detective Johnson testified after reviewing reports and photographs of the packaging of the evidence in this case that "the packaging of the materials in the case did not meet professional standards." He testified that although each piece of evidence was separately wrapped, it was placed in one box for transport or shipment, which is not the preferred manner of packaging. He testified that the best way is to put the victim's clothing in one container, the suspect's clothing in one container and biological evidence, like blood, in a different container to avoid transfer of potential evidence, such as fiber coming in contact with fiber. All evidence should be separated, labeled and sealed. Testimony of Larry Johnson, Transcript of Habeas Hearing, pp. 105-07, Feb. 1, 1999.

In this case, the testimony of TBI agents Murray and Breeding, who transported the evidence from Tennessee to the FBI in Washington, D.C., reveals that although the evidence from Mrs. Muncey's body may have been packaged separately, it was all transported to the FBI in Washington, D.C. in one box, contrary to good forensic practice. Testimony of Larry Johnson, Transcript of Habeas Hearing, pp. 104, Feb. 1, 1999. Moreover, it was the height of summer on the July day when the officers transported the material in the trunk of their car from Union County to Washington, D.C. There is no testimony that the trunk was air conditioned or that the blood samples from Mrs. Muncey were otherwise kept cool during the long trip from Tennessee to Washington, D.C. As noted above by Dr. Blake, the stoppers can blow off of blood vials if the sample becomes too warm.

Another witness, Howard Bragdon, worked for DiaClin, the private company that did the blood testing on the evidence at the request of House's attorneys in October of 1985 after it had been in the possession of the government for over three months.5 Mr. Bragdon transported the evidence, including the Styrofoam box containing the vials of blood, from Union County to DiaClin's offices in Nashville. Testimony of Howard Bragdon, Transcript of Habeas Hearing, pp. 1-6, Feb. 2, 1999. He testified that there was dried blood on the Styrofoam box when he took possession of it. Id. at pp. 6-7. The photographs taken by DiaClin in October 1985 and introduced at the habeas hearing clearly show blood staining the Styrofoam container used to transport the blood.

Mr. Bragdon also testified that one of the four vials was only partially full and another was nearly empty. Yet, Agent Bigbee testified that he used only one-fourth of a vial in performing his tests at the FBI lab in Washington, D.C. Mr. Bragdon also testified that two of the vials appeared to be leaking or had leaked when he picked them up in Union County.

The documentation, or lack thereof, concerning the chain of custody of the blue jeans and Mrs. Muncey's blood samples is particularly troubling due to the evidence of possible mishandling of the evidence. The purpose of the chain of custody requirement is to "demonstrate that there has been no tampering, loss, substitution, or mistake with respect to the evidence." State v. Braden, 867 S.W.2d 750, 759 (Tenn.Crim.App.1993). While the state is not required to establish facts which exclude every possibility of tampering, the circumstances established must reasonably assure the identity of the evidence and its integrity. State v. Ferguson, 741 S.W.2d 125, 127 (Tenn.Crim.App.1987).

The necessary "integrity" is lacking here. One exhibit introduced at the habeas hearing shows the plastic bag labeled by the TBI that contained the blue jeans. Petitioner's Ex. 10-6, Habeas Hearing, Feb. 1999. However, Agent Scott of the TBI testified that he put the jeans in a paper bag after retrieving them from the clothes hamper at Donna Turner's house after the murder. Testimony of Charles Scott, Transcript of Habeas Hearing at p. 115, Feb. 3, 1999.6 In addition, the FBI list of evidence it received from the TBI notes that the jeans were received in a sealed paper bag. FBI File, Addendum No. 2, Criminal Trial Exhibits, Ex. 31, pp. 36, 49. The plastic bag bore a label with the name "Charles Scott" written on it, but that label covers part of the markings made on the bag by the FBI, indicating that the label was put on after the FBI did its analysis. This sequence of events makes no sense because Charles Scott was the first agent to handle the jeans when he retrieved them from Ms. Turner's house after the murder. Given the evidence that the jeans were originally placed in a paper bag, Agent Scott's label likely was moved from another place—probably the original paper bag—and put on the plastic bag at a later time. The documentation showing the chain of custody for the blue jeans is of no help in sorting out the sequence of events. I can find no subsequent records showing when or by whom the jeans were transferred from the paper bag to the plastic bag.

Finally, adding to the likelihood that the blood samples were spilled on the jeans at some point after they were removed from Ms. Turner's home, the plastic bag holding the blue jeans appeared to have blood stains on it when received by Lisa Calandro, an expert in forensic DNA analysis and biological evidence analysis hired by House. See Photograph of plastic bag, Petitioner's ex. 10-6, Habeas Hearing, Feb. 1999; Respondent's Ex. 21, Habeas Hearing, Feb. 1999 (same). Ms. Calandro testified that the outside of the bag tested positive for the presence of blood, but analysis to further identify its source was unsuccessful. Testimony of Lisa Calandro, Transcript of Habeas Hearing, p. 76, Feb. 1, 1999.

This mystery of how the blue jeans got from the paper bag in Tennessee to a plastic bag sometime later, the confusion in the labeling of the bag, the evidence of blood spilled inside the box containing the evidence, and the acknowledgment of missing blood demonstrate that critical evidence in this case was possibly tampered with, or, at the very least, not handled according to recognized protocol and accidentally spilled.

The testimony of Pauline Sutton also raises questions as to whether the jeans were tampered with. She noted that some of the blood stains were mixed with mud. The National Weather Service records show that it had not rained for three days prior to the murder, and photographs of the crime scene showed no mud present. There was no mud on the victim's clothing. This evidence is inconsistent with the State's contention that the mud and blood got on the jeans at the crime scene.

While the State acknowledges, and even stipulates on the record, that some of the blood taken from Mrs. Muncey's body after her death cannot be accounted for and was likely spilled, Statement of Assistant Attorney General Glenn Pruden, Transcript of Habeas Hearing, p. 115, Feb. 1, 1999, as shown by the evidence introduced at the habeas hearing, it has offered no explanation nor even shown any interest in determining what happened to the missing blood.

Unlike many cases where accusations about chain of custody problems or mishandling of evidence are made, this is not a case where an inference of mishandling of the evidence has been made with no evidence to support such an inference or a case where there is simply a short gap in time where custody of the evidence cannot be determined. Here, both parties agree that blood was spilled or leaked from the vials. There is a blood stain on the container that held the blood samples and a blood stain on the plastic bag that contained House's blue jeans. The vials themselves clearly show that significant leakage occurred and blood is missing from at least one of the vials that has not been accounted for through testing. What has not been established is when this spillage or leakage happened or how it happened. This inability to account for the blood through either the chain of custody or testimony from an individual along that chain, and the undisputed existence of blood stains on the container that transported the blood vials eliminates the probative value of the blood evidence. Surely the facts demonstrating the lack of care in handling the evidence, when viewed in conjunction with Dr. Blake's expert testimony that the blood on the blue jeans came from the vials taken from Mrs. Muncey's body after her death, not when she was alive, would lead a reasonable juror, hearing this evidence, to have a substantial and serious doubt as to House's guilt. The Blake testimony together with the spillage creates a high probability that the blood on the House jeans came from a vial of blood.

IV. Newly Discovered Evidence That Mr. Muncey Killed His Wife

In addition to the newly discovered scientific evidence concerning the semen and blood stains, there is now strong independent evidence against Mr. Muncey as the perpetrator of the crime. There is no reason to believe that a jury would reject the testimony of the two life long residents of the Luttrell community who describe Mr. Muncey's weeping confession to them that he killed his wife accidentally while beating her. The women have no reason to fabricate, no axe to grind. Nor is there a reasonable basis for rejecting the highly incriminating testimony of two other women and the security guard at the Saturday night dance. Because the Court and the prosecution question the credibility of these witnesses and the inferences to be drawn from their testimony, we will quote their testimony in detail so that readers can make judgments for themselves.

Penny Angela Letner, a life long native of the Luttrell community, testified that there was a small gathering at the house of her sister, Kathy Parker, near the time of House's trial in 1986. At that time Mr. Muncey confessed that he killed his wife under the following circumstances:

Q. Okay. If you would, just tell His Honor, what happened at that party.

A. I came in—and we call him Bubba—he was sitting there and he was pretty well blistered. I could tell by the way he was talking.

Q. Who do you call Bubba?

A. Hubert Junior. We call him Bubba. We growed up around him. That was just a nickname for us. He was sitting there and he went to crying and was talking about his wife and her death and he was saying that he didn't mean to do it.

MS. LUSTRE: Objection.

THE COURT: Same ruling. Go ahead.

A. He said he didn't mean to do it. That she was "bitching him out" because he didn't take her fishing that night, that he went to the dance instead. He said when he come home that she was still on him pretty heavily "bitching him out" again and that he smacked her and that she fell and hit her head. He said I didn't mean to do it, but I had to get rid of her, because I didn't want to be charged with murder.

Q. What did you do, when you heard this?

A. When he said that he had to get rid of her, it scared me quite badly. I was 19 year old with a small child. I got out of there immediately. After he made those statements, I was ready to leave there.

Q. Had Little Hube been drinking that evening?

A. From the way he was acting, quite a bit. He was talking kind of stiff tongued like, but it wasn't to the point that I couldn't understand what he was saying.

Q. Was he emotional when he was telling the story?

A. He was crying. He was very upset.

Testimony of Penny Letner, Habeas Hearing Transcript, pp. 28-29, Feb. 1, 1999.

On cross-examination, she explained her reason for coming forward in 1999 to give this testimony:

Q. Did you report this to the police?

A. No. I didn't. I was 19 year old. I was kind of scared. I was frightened, you know. I didn't know how to take it. I figured me being 19 year old they wouldn't listen to anything I had to say.

Q. How is it that you came to give a statement now?

A. They was some people came and started asking questions and I answered their questions truthfully.

Q. What people?

A. It was an investigation that was on-going for Mr. House.

Q. Do you remember who it was that came to talk to you?

A. Michael Lee was the first gentleman that I spoke with. He was the first one that questioned me.

....

Q. Why did you not tell Mr. Lee to get lost, essentially?

A. Because it's not right. The gentleman stood right there in my face in front of several people and owned up to doing it. It's not right for a man to be in jail for something he didn't do.

Q. What made it not right 13 years later but okay for the intervening 13 years?

A. There wouldn't nobody listen to me. Finally there was someone willing to talk to me.

Id. at pp. 31-32.

Kathy Parker, Mrs. Letner's sister, also testified as follows about Little Hube's confession and her reasons for coming forward:

Q. Do you have any recollection about anything that he said?

A. Oh, yeah, he said they had been into an argument and he slapped her and she fell and hit her head and it killed her and he didn't mean for it to happen.

Q. Was he intoxicated?

A. He was drinking real heavily, yeah.

Q. Was he emotional?

A. Very.

Q. All right. How very is very?

A. Well, he was crying and just all to pieces.

Q. All right. How long had he been there before he told you about this incident?

A. Maybe 10 or 15 minutes, not real long.

Q. Did he say what they were arguing about?

A. He had wanted to go to a dance or something or another and was wanting to go somewhere else. That is what they got into an argument over.

Q. What did you do when you heard Little Hube say he hit his wife and she died?

A. I freaked out and run him off.

Q. You freaked out?

A. I freaked out and ran him off.

Q. Okay. After the party did you tell anybody about this?

A. Not that night. The next day I went to Union County and tried to talk to some law people and—

Q. Would they listen to you?

A. Went to Union County to the Sheriff's Department. I tried to speak to the Sheriff but he was real busy. He sent me to a deputy. The deputy told me to go upstairs to the courtroom and talk to this guy. I can't remember his name. I never did really get to talk to anybody.

Q. Tried to tell them?

A. Yeah.

Q. Did you talk to your mother about it?

A. A little later on there wasn't a whole lot said about it, but she was the one that took me to the courthouse.

Q. Your mother went with you to the courthouse that day?

A. She drove me. I didn't have a vehicle.

Q. Did you know Carolyn Muncey?

A. Yes, sir.

Q. Are you aware of whether or not Little Hube had ever abused her or beat on her?

A. She was constantly with black eyes and busted mouth.

....

Q. Let me ask you this. You know Little Hube?

A. Yes, sir.

Q. How long you been knowing him?

A. I dated him when I was 14 year old. I will be 42 in June.

Q. You have known him a long time?

A. Yes.

Q. Well, have you had any disagreement with him lately, any arguments, fights?

A. No.

Q. Are you mad at him?

A. No.

Q. Why are you here today?

A. Because I don't think what has happened is right. It needs to be taken care of. An innocent man is in jail.

Q. Do you know the petitioner, Paul House?

A. No, sir.

Q. Ever met him?

A. No, sir.

Testimony of Kathy Parker, Habeas Hearing Transcript, pp. 37-39, Feb. 1, 1999.

Around two months before the murder of Carolyn Muncey, Mr. Muncey also told Hazel Miller, also a life long resident of Luttrell, that he was going "to get rid of that woman one way or the other":

Q: All right. Back prior to the death of Mrs. Carolyn Muncey, did you have occasion to see Little Hube Muncey in your residence?

A. Yes, sir, sure did.

Q. About how long before Mrs. Carolyn Muncey's death was Little Hube in your house?

A. Two, three months, something like that.

Q. What was he there for, if you know?

A. He come tried to get my daughter to go out with him.

....

Q. What did Little Hube say?

A. He was upset with his wife, that they had had an argument and he said he was going to get rid of that woman one way or the other.

Q. What did you say?

A. Told him he ought to go home to his wife and kids and take care of them.

Q. Did he say anything about his kids?

A. He said his mother told him she was going to help him take care of them.

....

Q. Now, I take it you have known Little Hube for a while?

A. Yes. Basically about all his life.

....

Q. Have you had any disagreements or arguments or fights with Little Hube or any members of his family?

A. No, he has always treated me decent.

Q. Are you all presently on the outs? Are you mad at him or anything?

A. No.

....

THE COURT: All right. Where did this conversation take place?

A. At my kitchen table.

THE COURT: You said he ought to go home to his wife and children?

A. Wife and kids, yes. I thought he should.

THE COURT: What did he say to you? A. He was going to get rid of her one way or the other. I presume he meant he was going to get a divorce.

Testimony of Hazel Miller, Habeas Hearing Transcript, pp. 47-50, Feb. 1, 1999.

In addition to the confession to Letner and Parker and the "get rid of" statement to Miller, there was additional evidence incriminating Muncey at the habeas hearing. Hubert Muncey insisted he was at the dance until it ended on Saturday night and then went home after midnight to find his wife missing. This testimony is inconsistent with the testimony of Dennis Wallace, the security guard on duty at the dance. He testified:

Q. All right. During that period of time did you see Little Hube Muncey at the dance?

A. Yes, sir, I did.

Q. And during that period of time before the dance was over did you see Little Hube Muncey leave the dance?

A. Yes, sir, I did.

Q. All right. Then subsequent to that, did you see Little Hube Muncey return to the dance?

A. Not that I can recall.

....

Q. Do you remember approximately what time Little Hube Muncey left the dance that evening? When you say saw him leave, before it was over?

A. I would say around 10:00, 10:30, 9:30 to 10:30. I don't know exactly.

Q. You don't recollect now having seen him come back to the dance?

A. No.

Testimony of Dennis Wallace, Habeas Hearing Transcript, pp. 56-57, Feb. 1, 1999.

Mrs. Artie Lawson testified that she is "a good friend" of Little Hube's and that he came to see her Sunday morning to get her to help him establish an alibi for Saturday evening. This conversation on Sunday morning took place several hours before the body was found and before it became clear that Mrs. Muncey was murdered, not just missing. Mrs. Lawson testified:

A. Well, this Little Hube, he come to my house that morning on Sunday morning and he asked me if anybody come to say anything, you know, talk to me, to tell them that he was there at six o'clock.

....

A. He asked me if anyone, you know, come and asked me anything to tell him that he had, that I was at the dance that Saturday night—but I was not. He said that he had eat breakfast down there [at my house] at six o'clock that Sunday morning and he did not.

Testimony of Artie Lawson, Habeas Hearing Transcript, pp. 20-22, February 1, 1999.

Hours before his wife's body was found, Muncey went to Mrs. Lawson and tried to establish an alibi for Saturday night. He in fact had no alibi. He left the dance an hour and a half early and is unable to account for his whereabouts at the very time the county coroner gave as his best estimate as the time of death. Then early the next morning before the body was discovered, he went to Mrs. Lawson in an effort to enlist her help in establishing an alibi.

The evidence of what happened at the Muncey home on Saturday evening further undermines the case against House and further incriminates Muncey. Pamela Luttrell and her husband lived across the road from the Munceys. Mrs. Luttrell testified that Carolyn Muncey came to her house sometime around 8:00 P.M. on Saturday evening with her children and "stayed an hour or an hour and a half" and "said that Little Hube had gone to dig a grave and he hadn't come back." Testimony of Pamela Luttrell, Addendum No. 4, Vol. II, pp. 632-33. She testified that between 10:00 and 11:00 "within an hour or so after she left" with her children to go back home, Mrs. Luttrell believed she heard Little Hube's car coming back home:

Q. Now, after she left, did you see or hear anything else that night that drew your attention to outside directly in the vicinity of Carolyn Muncey's home?

A. I heard a car rev its motor as it went down the road.

Q. And you had heard that sound before or you thought you had heard that sound before, hadn't you?

A. Well, I have heard a sound like it.

....

Q. And that was the impression you got that night, wasn't it, that that was Little Hube's car revving up?

A. Yeah, just going by.

Q. And when he'd get to the house there, he would rev up. Is that correct?

A. And then keep going.

Q. Now that would have been somewhere around—after Carolyn had left and before you went to bed. You went to bed within an hour or so after she left?

A. Uh, huh (meaning yes by sound), I was on my way to bed when I heard it.

Id. at pp. 641-42.

Mrs. Luttrell testified that the car was coming from the direction in which Muncey's car would have traveled coming from the dance, from Kitts Road toward Tazewell Pike and "revved up" when he got to the house, as he usually did. Mrs. Luttrell's testimony is consistent with the testimony of Constable Wallace that Muncey left the dance well before it ended. It is inconsistent with Muncey's testimony that he stayed at the dance until it ended. Together with the testimony of Wallace, Mrs. Luttrell's testimony places Muncey at home between 10:00 and 11:00 that evening—the approximate time that the coroner testified that the murder occurred.

In addition, Pamela Luttrell's testimony about what happened later that evening is also completely inconsistent with the theory that House came to the Muncey home on foot during the evening and abducted Carolyn and murdered her. Luttrell testified that Little Hube, along with the two children, Laura and Matt, came to her home at about 1:00 A.M. that evening and told her Carolyn was missing. At that time, Luttrell questioned Laura in detail about what had happened that evening after she got home from her earlier visit to Luttrell's house. Laura said that she and her younger brother had gone to sleep:

Q. Mrs. Luttrell, Laura talked with you after she discovered her mother missing?

A. Yes.

Q. Did she tell you whether or not she woke up one time or two times, as far as different people coming to the house asking questions?

A. Well, she said she heard a horn blow, she thought she heard a horn blow, and somebody asked if Bubbie was home and her mama, you know, told them—no. And then she said she didn't know if she went back to sleep or not, but then she heard her mama going down the steps crying and I am not sure if that is when she told me that she heard her mama say—Oh, God, no, not me, or if she told me that the next day, but I do know that she said she heard her mother going down the steps crying.

Id. at 658.

Luttrell further testified that Laura told her that the man who asked for "Bubbie" after the horn blew sounded like her grandfather, whom she called "PawPaw." Luttrell said that "Bubbie" was a family name not normally used by others in the community who called Muncey "Little Hube." Laura said nothing in answer to Pamela Luttrell's questions that evening about anyone saying during the evening that Little Hube had been involved in an automobile accident. This evidence seems completely inconsistent with the State's theory that House walked almost two miles at around 10:30 on Saturday evening to the Muncey home where he abducted and murdered Carolyn Muncey. It is clear that House did not have a car that evening, did not blow a car horn, would not have called Little Hube "Bubbie," did not enter the Muncey home, wake up the Muncey children or talk to Carolyn Muncey within the ear shot of the children. In the contemporaneous, immediate conversation between Mrs. Luttrell and Laura, Laura said that she heard a horn blow, someone ask for "Bubbie," who was not there and that person "sounded like PawPaw," who called Little Hube by his family nickname, "Bubbie."

Although it is true, and highly incriminating, that House left his home on foot during a similar time on Saturday evening and returned with bruises and in a disheveled condition, Pam Luttrell's testimony that Little Hube came home around 10:30 and Laura's testimony about the hornblowing and the "PawPaw" voice leaves no room for the State's theory that House walked two miles to the Muncey house and murdered Carolyn Muncey and then walked two miles home within a relatively short period of time. The inference that he murdered Carolyn Muncey from the fact that he left home on foot on Saturday evening and returned with bruises and in a disheveled condition cannot be reconciled with Mrs. Luttrell's testimony about when Little Hube came home and Laura's answers to her questions about what happened that evening. Mrs. Luttrell's testimony is consistent with the testimony of Constable Wallace that Mr. Muncey left the dance at about 10:30 P.M. as well as Muncey's confession, his previous threats and brutal treatment of his wife and his attempt to establish an alibi through Mrs. Lawson on Sunday morning.

This particular highly incriminating evidence about House's returning home in a disheveled condition on Saturday night with "blood on his pants," and his semen on the victim's night gown, however, does go a long way in explaining the jury's capital conviction of House, a stranger in a community with a record of sexual assault, especially when it did not have any of the newly discovered evidence undermining the prosecution's theory of the case. In the absence of the newly discovered evidence, the State's theory seemed convincing. In light of the new evidence, it is no longer tenable. In my view, it is more probable than not that House is innocent of the homicide and that Mr. Muncey killed his wife, as he confessed. Six witnesses— Letner, Parker, Miller, Wallace, Lawson, and Luttrell—now provide either direct or circumstantial evidence that Mr. Muncey killed his wife. He had a motive. The motive attributed to House no longer exists. NOTE: OPINION CONTAINING TABLE OR OTHER DATA THAT IS NOT VIEWABLE

V. The Unreliability of the Hensley Testimony

The preceding map of the area where Carolyn Muncey's body was found is necessary to understand the weakness of the testimony of the one remaining witness, Billy Ray Hensley, who gave highly incriminating evidence at the 1986 trial. The opinion of our court accurately characterizes Hensley's testimony as a series of "inconsistent statements about when and where he saw House on Ridgecrest Road" supposedly "carrying a black rag." The prosecution claimed at the trial that the "black rag" was House's blue shirt which he had lost the night before and had just retrieved after climbing down the embankment where he had left Carolyn Muncey's body.

House explained his presence near Muncey's house at the habeas hearing. He testified that he regarded "Little Hube" and Carolyn Muncey as friends and he heard on Sunday morning that she was missing. He borrowed the car of his girl friend, Donna Turner, and drove down to the area on Ridgecrest Road where he knew they lived. This testimony is corroborated by other witnesses. He missed the driveway which was hidden in trees and undergrowth and turned around and parked the car near Merritt's barn and walked back up about 90 feet and found the driveway and saw that Muncey's car was not at home. He walked back down Ridgecrest Road to his car:

Q. You went back to your car. Did you get in?

A. Yes.

Q. And start driving?

A. I got in, pulled out. I was going to—between Little Hube's and Tazwell Pike there is another dirt road like a dead end.

Q. Somewhere off the chart?

A. Yeah, it is on his side where he lived. It is on that side of the road. It is just a dirt road, dead end. Bill Silvey had told me that people went to drink down there. Donna had told me that people went to drink down there. Donna had told me she had seen him drinking. I thought why not look there. As I was going up there, then Bill Hensley, I guess that is the name, was in the car. I saw him and he was coming towards me. I was going toward him. I waved him over. He stopped.

Q. He was in the little blue car?

A. Right. I asked him if he knew Little Hube. He said he did. Asked him if he had seen him, he said he hadn't, didn't see him around there anywhere.

Testimony of Paul House, Transcript of Habeas Hearing, p. 91, Feb. 3, 1999.

Hensley had also testified at the 1986 trial that House while in his car waved him to stop at that location on Ridgecrest Road and told him he was looking for Little Hube.

The incriminating part of Hensley's testimony at the trial was that as he was coming down Bear Hollow Road approximately 100-150 feet from its intersection with Ridgecrest Road he saw House coming up an embankment where the victim's body was later found:

Q. Now, you say that as you were coming down, we will call this Bear Hollow Road, just so the jury and everybody will know the roads we are talking about. Before you got to this intersection, before you turned left [onto Ridgecrest Road] to go to Tazewell Pike, you say you saw my client, Mr. House?

A. Yes, sir.

Q. All right. You weren't at the intersection, were you?

A. No, sir.

Q. In fact, you have previously told us you were somewhere between 100 and 150 up from the intersection, weren't you—100 to 150 feet up from the intersection?

A. Yes, sir.

Testimony of Billy Hensley, Addendum No. 4, Vol. V, p. 717.

After Hensley turned onto Ridgecrest road, he did not see House, according to his testimony:

Q. From the time that you made your left-hand turn, what was the first thing you saw in relationship to Mr. House or his vehicle?

A. I saw his vehicle.

Q. And at no time after you made your left-hand turn could you say where Mr. House was in relationship to the bank, the side, the road or anywhere in there, is that correct?

A. No sir, I could not.

Id. at 719-20.

There are two serious problems with this testimony. It is directly contrary to the statement Hensley gave to the police on the Sunday evening after the victim's body was found. In that statement he said that he turned left at the intersection and he:

traveled about 500 feet on Ridgecrest Road when I saw about a '66 or '67 white Plymouth sitting on the left side of Ridgecrest Road and I saw a man, later identified to me as Paul Gregory House enter the roadway from the right-hand side of the road. He was coming up over the bank, and he had a black rag in his hands and he was wiping his hands. I drove on by the man and went to Little Hube's driveway ...

Statement of Billy Ray Hensley taken by TBI on July 14, 1985, Addendum No. 1, p. 139; see also Testimony of Billy Hensley, Addendum No. 4, Vol. V, p. 729.

Hensley testified on cross-examination about the statement:

Q. Did you tell the police that on Sunday at 10:45 p.m., did you talk to them Sunday night, the same night—

A. Oh, yes.

Q. And you gave them that statement, did you not?

A. Yes, sir.

Testimony of Billy Hensley, Addendum No. 4, Vol. V, p. 730.

This statement that Hensley was at the embankment and saw House "coming up" is clearly contrary and unreconcilable with his trial testimony that he saw House 100-150 feet up Bear Hollow Road and did not see him after he made the turn onto Ridgecrest Road. The cross-examination of Hensley revealed not only these contradictions in his statements about where he was when he claimed to see House, as our court's opinion recognizes, but also it revealed the fact that as Hensley drove down Bear Hollow Road that various obstructions like Merritt's barn and trees would have prevented him from seeing anyone in the vicinity of the embankment where he claimed to have seen House. The map itself, the testimony and photographs, show conclusively that Hensley could not have seen House from that location on Bear Hollow Road.

Hensley's contradictory and implausible testimony is the last incriminating straw in the case against House. Like the semen evidence and the blood evidence it is no longer probative. It is a slender reed indeed on which to hang House's conviction and death sentence.

This is a death penalty case based purely on circumstantial evidence in which the prisoner has maintained his innocence during 20 years of incarceration. I would hold that he has easily met the requirements necessary to travel through the gateways of actual innocence of both the death penalty and the homicide by completely disproving the motive for the crime and the aggravator (rape) and through persuasive evidence undermining the main circumstances that gave rise to his conviction (the blood evidence and the Hensley testimony). This body of evidence alone meets both the Sawyer standard, that petitioner show by clear and convincing evidence that but for constitutional error at trial, no reasonable juror would have found him eligible for the death penalty and the Schlup standard, that petitioner show that it is more likely than not that no reasonable juror would have found him guilty beyond a reasonable doubt. The Court stated in Schlup that since gateway claims assert constitutional error at trial, these "convictions may not be entitled to the same degree of respect" as convictions that are the product of error free trials. Schlup, 513 U.S. at 316, 115 S. Ct. 851. At a minimum, based on the newly presented evidence we should allow House to pass through the gateway so that his constitutional claims of ineffective assistance of counsel at the guilt and penalty phases of the case become cognizable again. The remedy for passing the gateway test is a remand for further proceedings on the underlying constitutional claim.

In this case, however, I would go further and issue the writ of habeas corpus because the prisoner has affirmatively established a free-standing case of actual innocence. This is that rare and extraordinary case where petitioner has provided "a truly persuasive demonstration of `actual innocence'" that should free the prisoner immediately. Herrera, 506 U.S. at 417, 113 S. Ct. 853. Through extremely persuasive and affirmative evidence that Mr. Muncey killed his wife, House has shown that it is highly probable that he is completely innocent of any wrongdoing whatever. There is no reasonable basis for disbelieving the six witnesses who now incriminate Mr. Muncey as the perpetrator of the crime. The most compelling part of this new testimony involves his confession to the murder in front of two witnesses who have no connection to House and no bias against Mr. Muncey. Furthermore, before his wife's body was even located, he solicited a neighbor to fabricate an alibi on his behalf. He was heard returning home around the time of the murder. And considering his history of domestic violence and his wife's plans to leave him he had a motive to kill. In contrast, there is no evidence of a motive for House. All of the state's physical evidence, both blood and semen, allegedly tying House to the murder, has been effectively rebutted. The new body of evidence as a whole so completely undermines the case against House and establishes a persuasive case against Muncey that, had it been presented at trial, no rational juror could have found evidence sufficient for conviction. The new evidence so completely turns the case around that the proof is no longer constitutionally sufficient to warrant a conviction or imposition of the death penalty. Thus House should be immediately released.

Justice Scalia has referred to the question before us of actual innocence as death penalty's most "embarrassing question," a question he hoped "with any luck we [the Supreme Court] shall avoid ever having to face" in a "convincing" case. Herrera v. Collins, 506 U.S. at 428, 113 S. Ct. 853. Justice O'Connor has referred to this "embarrassing question" as a serious current problem: "If statistics are any indication, the system may well be allowing some innocent defendants to be executed." Speech to Women Lawyers in Minnesota, July 2, 2001.

This case and the few empirical studies that we have reinforce Justice O'Connor's view that the system is allowing some innocent defendants to be executed. See Radelet & Bedau, Miscarriages of Justice in Potentially Capital Cases, 40 Stan. L. Rev. 21-179 (1987) (includes tables and summary description of numerous capital cases in which innocent defendants were probably executed); Zimming, The Contradictions of American Capital Punishment 149-178 (2003) (summary of the studies, including study by Death Penalty Information Center describing 98 prisoners fully exonerated after time on death row between 1973 and 2001).

High on the list of the causes for mistakes are the kinds of errors we see in this case: the misinterpretation or abuse of scientific evidence, the adverse inferences drawn from the prior record of a defendant, particularly one who is a stranger in the local community, the failure of counsel to uncover (until it is too late) witnesses who could exonerate the defendant, and the existence of one or more other suspects with a motive to commit the offense. Once the initial trial and appeal have occurred, it is clear from the studies that the state, and its officials who have prosecuted, sentenced and reviewed the case, are inclined to persevere in the belief that the state was right all along. They tend to close ranks and resist admission of error. Intelligent citizens who strongly believe in the reliability of the capital sanction are also inclined to persevere in the belief that a case raising the "embarrassing question" will never really arise and close ranks with the state in resisting the admission of error. This case is a good example of how these errors can lead to the execution of a defendant who is actually innocent.

GILMAN, Circuit Judge, dissenting.

After reading Judge Norris's majority opinion and Judge Merritt's dissent, I am convinced that we are faced with a real-life murder mystery, an authentic "who-done-it" where the wrong man may be executed. Was Carolyn Muncey killed by her down-the-road neighbor Paul House, or by her husband Hubert Muncey? The evidence at House's state-court trial clearly pointed to him as the perpetrator, highlighted by the physical evidence of the semen and the blood. But the evidence at House's habeas corpus hearing before the district court just as clearly pointed to Hubert Muncey as the guilty party, highlighted by Muncey's confession of guilt to two female acquaintances, the uncontroverted fact that the semen found on his wife's clothing turned out to be his own, and the considerable doubt cast on how the victim's blood came to appear on House's blue jeans.

At the end of the day, I am in grave doubt as to which of the above two suspects murdered Carolyn Muncey. I am also puzzled as to why more of my colleagues are not similarly in doubt after evaluating the well-written but diametrically opposed opinions by Judges Norris and Merritt. Be that as it may, the question becomes what is a federal judge to do when faced with such grave doubt? The United States Supreme Court, fortunately, has provided an answer in the case of O'Neal v. McAninch, 513 U.S. 432, 115 S. Ct. 992, 130 L. Ed. 2d 947 (1995), a case that incidentally arose out of the Sixth Circuit. Its key holding is as follows:

We repeat our conclusion: When a federal judge in a habeas proceeding is in grave doubt about whether a trial error of federal law had "substantial and injurious effect or influence in determining the jury's verdict," that error is not harmless. And, the petitioner must win.

Id. at 436, 115 S. Ct. 992.

The Court in O'Neal also defined the term "grave doubt," stating that "[b]y `grave doubt' we mean that, in the judge's mind, the matter is so evenly balanced that he feels himself in virtual equipoise as to the harmlessness of the error." Id. at 435, 115 S. Ct. 992. Finally, the Court noted that

our conclusion is consistent with the basic purposes underlying the writ of habeas corpus. As we have said, we are dealing here with an error of constitutional dimension—the sort that risks an unreliable trial outcome and the consequent conviction of an innocent person.

Id. at 442, 115 S. Ct. 992.

The risk of an unreliable trial outcome is precisely what we are dealing with in the case before us. For the reasons well-documented in Judge Merritt's dissenting opinion, I believe that House has passed through the Schlup "actual innocence" gateway and is entitled to our consideration of his constitutional claims on their merits. On the other hand, I disagree with Judge Merritt's conclusion that the proof in favor of House is so strong that we should issue an unconditional writ of habeas corpus.

The proper disposition of this case, in my opinion, is to issue a conditional writ that would free House unless he is provided a new trial by the state of Tennessee. See Castleberry v. Brigano, 349 F.3d 286, 294 (6th Cir.2003) (granting a conditional writ of habeas corpus to vacate the defendant's conviction and sentence unless the state commenced a new trial against the defendant within 90 days). A new trial would allow the jury to assess House's guilt or innocence free from the erroneous introduction of the semen evidence, with full knowledge of the controversy surrounding the blood evidence, and with the benefit of the testimony implicating Hubert Muncey. Under circumstances where we face the execution of a man who might well be innocent, I believe that our system of justice demands no less.