240 F.3d 1169: Donald M. Paradis, Petitioner-appellee, v. A.j. Arave, Respondent-appellant
UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT - 240 F.3d 1169
Argued and Submitted December 1, 2000Filed March 5, 2001
COUNSEL: L. Lamont Anderson, Deputy Attorney General, Boise, Idaho, for the respondent-appellant.
Edwin S. Matthews, Jr., Coudert Brothers, New York, New York, for the petitioner-appellee.
Appeal from the United States District Court for the District of Idaho Edward J. Lodge, District Judge, Presiding. D.C. No.CV-95-0446-S-EJL
Before: William C. Canby, Jr. and A. Wallace Tashima, Circuit Judges, and Roslyn O. Silver, District Judge.*
TASHIMA, Circuit Judge:
Petitioner-Appellee Donald Paradis was convicted in Idaho state court of the murder of Kimberly Palmer and sentenced to death, but the State later commuted his sentence to life without parole. In November 1997, this court reversed in part the district court's dismissal of Paradis' second federal petition for writ of habeas corpus, and remanded the case for an evidentiary hearing on his claims that the prosecution breached its duties under Brady v. Maryland, 373 U.S. 83 (1963), by failing to disclose several sets of notes taken by prosecutor Marc Haws. See Paradis v. Arave, 130 F.3d 385 (9th Cir. 1997). On remand, the district court held a three-day evidentiary hearing and subsequently conditionally granted the writ, unless the State retries Paradis within 120 days. See Paradis v. Arave, No. CV 95-446-S-EJL, 2000 WL 307458 (D. Idaho Mar. 14, 2000). We have jurisdiction over this timely appeal pursuant to 28 U.S.C. S 1291, and we affirm.
On the morning of June 22, 1980, Palmer's body was discovered lying face down in a shallow creek at the bottom of a steep ravine. The body was found 70 to 80 feet below an overturned blue and white Volkswagen van near Mellick Road, a steep mountain road near Post Falls in Kootenai County, Idaho. She was dressed in jeans, but no underpants, and was naked from the waist up, with her shirt underneath her. There was a cut in the front of the jeans and, through it, a corresponding cut in her skin near the labia. The body of Scott Currier, a friend of Palmer's, was found stuffed into a sleeping bag 20 or 30 feet from the same Volkswagen van.
At approximately 6:30 a.m. on the previous morning, a blue and white Volkswagen van carrying two or three men was seen ascending Mellick Road. Thirty minutes later three men, later identified as Paradis, Thomas Gibson, and Larry Evans, were seen descending the same road on foot, and they were repeatedly observed at various locations in Post Falls during the next 30 minutes.
In the evening of June 22, 1980, Detective George Elliott and Deputy Wesley Krueger of the Kootenai County Sheriff's Office transported the bodies of Palmer and Currier to Portland, Oregon, for an autopsy by Dr. William Brady, the Oregon State Medical Examiner. Elliott and Krueger arrived in Portland at approximately 9:00 a.m. on June 23. Except for stepping outside a few times for some fresh air and possibly once for a phone call, Elliott and Krueger were present throughout the autopsies of both Palmer and Currier. They brought the bodies back to Idaho on the same day, arriving in Idaho late that night.
In the morning of June 24, 1980, law enforcement personnel from Kootenai County, Idaho, and Spokane County, Washington, met at the Spokane County Sherriff's Office regarding Palmer's and Currier's deaths. Kootenai County Deputy Prosecutor Marc Haws attended the meeting, as did Elliott, but Krueger did not. Elliott was the only person at the meeting who had also been present at the autopsies.
Authorities investigating the deaths concluded that Currier was killed at Paradis' home in Spokane, Washington, and that his body was moved, but they concluded that Palmer was killed in Idaho at or near the site where her body was found.
Paradis and Gibson were tried jointly in Washington for Currier's murder and were acquitted in September 1980. They were tried separately in Idaho for Palmer's murder. Gibson was convicted of first-degree murder and sentenced to death. His federal petition for writ of habeas corpus was later granted by the district court and, while the appeal of that decision was pending, the parties reached a plea agreement under which Gibson pled guilty to second-degree murder.
Following Gibson's trial, Paradis was also tried and convicted in Idaho of the first-degree murder of Palmer and was also sentenced to death. His conviction and sentence were affirmed by the Idaho Supreme Court, see State v. Paradis, 106 Idaho 117, 676 P.2d 31 (1983), which also affirmed the denial of state postconviction relief, see Paradis v. State, 110 Idaho 534, 716 P.2d 1306 (1986). Paradis then brought his first petition for writ of habeas corpus in federal court, which was denied after a six-day evidentiary hearing. See Paradis v. Arave, 667 F. Supp. 1361 (D. Idaho 1987), aff'd in part and rev'd in part, 954 F.2d 1483 (9th Cir. 1992), vacated and remanded, 507 U.S. 1026 (1993), aff'd on remand, 20 F.3d950 (9th Cir. 1994). The State subsequently commuted his sentence to life imprisonment without the possibility of parole.
In 1995, Paradis brought a second petition for writ of habeas corpus in federal district court. The court dismissed the petition as successive or abusive, Paradis appealed, and this court reversed in part and remanded. See Paradis, 130 F.3d 385. We held that the district court had erred by treating Paradis' Brady claim as a sufficiency of the evidence claim. Id. at 392-93. We further held that his Brady claim was not barred as successive or abusive because (1) Paradis had shown cause and prejudice, see id. at 393-95, and, in the alternative, (2) Paradis had made a sufficient showing of actual innocence under Schlup v. Delo, 513 U.S. 298 (1995). See Paradis, 130 F.3d at 396-99. We remanded the case for an evidentiary hearing on the Brady claim. After holding a three day evidentiary hearing, the district court granted the writ, and this timely appeal followed.
Paradis' Brady claim is based on three sets of handwritten notes (Notes I, II, and III) taken by Haws, each apparently recording the medical opinions of Dr. Brady regarding Palmer's death; opinions that Dr. Brady held shortly after he performed the autopsy. See Paradis, 130 F.3d at 392. Notes I contain the following phrases, among others: "no time of death either"; "spoonful water in lungs -siphon prob dead when went in water"; "not sexually assaulted." Notes III contain the following phrases, among others: "may have been alive when went"; "no gross evid. sex. molest."; "wound outside genitalia -tear enlarged after death -cut under tear on jeans right before death"; "cut -1" long evid no bleeding"; "blood ear nose could -have immediately followed strangulation." On remand, Paradis conceded that Notes II were taken by Haws at Paradis' own trial and, therefore, are not Brady material.
We found that Paradis was prejudiced by the failure of the prosecution to disclose the notes, and that he had made a sufficient showing of actual innocence, for the following reasons.1 First, the location of Palmer's death was important in at least two ways: (1) If Palmer did not die in Idaho, then the State of Idaho did not have jurisdiction to prosecute anyone for her murder; (2) Paradis claimed that both Palmer and Currier were murdered at Paradis' home in Spokane, Washington, at a time when Paradis was not there. Paradis, 130 F.3d at 392.
Second, Dr. Brady's assertion at Paradis' trial that Currier died before Palmer is undermined by his apparent inability to form such an opinion at the time of
the autopsy, as reflected by "no time ofdeath either" in Notes I. Paradis, 130 F.3d at 396-97.
Third, Dr. Brady expressed the opinion at Paradis' trial that, although the cause of Palmer's death was manual strangulation, Palmer aspirated some water from the creek with her final agonal breaths. This claim supported the prosecution's theory that she was strangled at the site where her body was found, because she would have to have been alive and still gasping for air when she was placed in the water. See id. at 394. Dr. Brady's opinion was based on his observation at the autopsy that Palmer's lungs were heavy and filled with fluid. Id. However, we noted testimony from other experts to the effect that heavy, fluid-filled lungs are common after manual strangulation and do not allow an inference that water was aspirated. Id. at 397. In addition, the statements of Dr. Brady that are reflected in Haws' notes, namely, "spoonful water in lungs" and "dead when went in water," contradict his trial testimony that she aspirated water and that he had observed that her lungs were filled with fluid. Id.
Fourth, there was the complex issue of the labial tear. Had the tear been inflicted before Palmer died or shortly thereafter, it would have bled profusely (because the area is highly vascularized); if it did not bleed, it is overwhelmingly likely that it was inflicted at least one-half hour after Palmer died. Id. at 395. Because Paradis, Gibson, and Evans were only at the Mellick Road site for approximately one-half hour, if the wound were postmortem, that would show that they could not have killed her there. A luminol test (a highly sensitive chemical test) detected no traces of blood on Palmer's jeans, strongly suggesting that the wound did not bleed. See id. at 398. Dr. Brady, however, expressed the opinion (at Evans' trial) that the blood must have been washed away by rain and by the water in the stream. Id. But the notes show that Dr. Brady found blood on Palmer's ear and nose, and that the blood "could have immediately followed strangulation." Id. Because Palmer's head was lower in the water than the remainder of her body, it is difficult, if not impossible, to understand how the stream could have washed away the blood from the labia wound so thoroughly that it could not be detected by a highly sensitive chemical test, while failing to wash away the blood from an equally (or more) exposed nose and ear. Id. at 398-99.
Dr. Brady did not testify regarding the labial tear at Paradis' trial and Paradis' trial counsel, William Brown, did not raise the issue on cross-examination. Id. at 394. Brown claimed that he avoided the issue because he was afraid it would "raise the specter of sexual abuse."2 Id. We concluded that this fear would likely have been allayed if Paradis had had access to Notes I, which unequivocally indicate that Palmer was "not sexually assaulted." Id. at 395.
D. Evidentiary hearing and decision on remand
At the evidentiary hearing, the district court heard testimony from Elliott, Krueger, Haws, Dr. Brady, and Dr. Glenn Faith, a medical expert who worked with Paradis' trial team. The court also received into evidence various exhibits, including prior testimony of Dr. Brady at both Paradis' and Gibson's trials and an affidavit from Dr. John Thornton, Ph.D., an expert on luminol testing. The court also received into evidence Elliott's original police report, which stated that Palmer "was sexual [sic] assaulted." Although it appeared that other copies of the report indicated that she was not sexually assaulted, the court found that the version stating that she was sexually assaulted was the one given to Brown.
In its decision, the district court included a summary of some of the testimony presented at the hearing. In particular, the court noted that Elliott testified that he was present when Dr. Brady cut open Palmer's lungs, and that a small amount of fluid came out. It also noted that Dr. Faith testified that because of the police report, Brown was worried about the suggestion of sexual assault, and that if Brown had known the report was in error, Brown would have cross-examined Dr. Brady regarding the labia wound.
The court noted that at Gibson's trial, Dr. Brady testified that the labia wound showed no evidence of bleeding or crusting. It also described Dr. Brady's testimony at the evidentiary hearing regarding his statement in the autopsy report that the labia wound showed "no evident vital reaction. " Dr. Brady testified that "no evident vital reaction" did not mean that the wound did not bleed--a "vital reaction" is a healing process, not bleeding. Dr. Brady further testified that he assumed that the wound had bled and that the blood had been washed away, but he also testified that he has no evidence that the wound did bleed. The court also noted that Dr. Brady testified that the photographs that appear to show blood on Palmer's nose and ear actually show only "skin slippage." Dr. Brady explained that skin (on a corpse) that has been soaking in water for a long time becomes very soft, and that it is very easy for the outer layer of such skin to be rubbed off accidentally when the body is moved or handled. This is called "skin slip," and it exposes the lower layer of the skin, which is bright red.
In analyzing Paradis' Brady claim in light of the evidence introduced at the hearing, the district court concluded that the notes were material under Brady for the following reasons: Dr. Brady's testimony that Currier died before Palmer and that Palmer aspirated water were "instrumental in establishing jurisdiction." Dr. Brady testified unequivocally at Paradis' trial that Currier died before Palmer, but Notes I ("no time of death either") could have been used to impeach him on this point. Notes I ("spoonful water in lungs" and "dead when went in water") could also have been used to impeach him regarding Palmer's alleged aspiration of water. The uncertainty expressed in Notes III ("may have been alive when went [in water]") could have been used to the same effect.
The district court further found that Paradis' lack of access to Notes I ("not sexually assaulted") and Notes III ("no gross evid. sex. molest."), coupled with the erroneous police report, prevented Brown from cross-examining Dr. Brady regarding the labia wound, an issue that "might have refuted jurisdiction." But because there was conflicting evidence regarding whether Brown was given color photographs that showed the blood on Palmer's ear and nose, the court found that Paradis had not carried his burden of proving that this information was not disclosed to him prior to trial.
The district court concluded that neither set of notes would have been independently admissible evidence, and that Notes I might not even have been admissible to impeach. But it also concluded that they could have led to the discovery of admissible evidence, and that the nondisclosure of the notes "appears to have put the defense team at a substantial disadvantage in preparing their case." For all of these reasons, the district court concluded that the notes undermined confidence in the verdict, that they were consequently material under Brady, and that the writ should therefore be granted.
We review de novo the district court's decision to grant a habeas petition under 28 U.S.C. S 2254. Bribiesca v. Galaza, 215 F.3d 1015, 1018 (9th Cir. 2000). The district court's findings of fact are reviewed for clear error. Lopez v. Thompson, 202 F.3d 1110, 1116 (9th Cir. 2000). We may affirm on any ground supported by the record, even if it differs from the rationale of the district court. Bonin v. Calderon, 59 F.3d 815, 823 (9th Cir. 1995).
Under Brady, the prosecution has a constitutional obligation to disclose exculpatory evidence to a criminal defendant if it is "material" either to guilt or to punishment. Brady, 373 U.S. at 87. This obligation extends to impeachment evidence, United States v. Bagley, 473 U.S. 667, 676 (1985), and to evidence that was not requested by the defense, id. at 682. See also United States v. Agurs, 427 U.S. 97, 107-10 (1976). Evidence is material if "there is a reasonable probability that, had the evidence been disclosed to the defense, the result of the proceeding would have been different. A `reasonable probability' is a probability sufficient to undermine confidence in the outcome." Bagley, 473 U.S. at 682; see also Kyles v. Whitley, 514 U.S. 419, 434 (1995); United States v. Kennedy, 890 F.2d 1056, 1058-59 (9th Cir. 1989). The final determination of materiality is based on the "suppressed evidence considered collectively, not item by item. " Kyles, 514 U.S. at 436-37.
The State argues that on remand the district court improperly shifted the burden to the State to disprove Paradis' Brady claim. In support of this argument, the State relies on the following passage from the district court's opinion:
The Ninth Circuit Court of Appeals previously concluded that the notes conflicted with Dr. Brady's testimony at trial and that counsel was not given the right to investigate or cross-examine on it. The court of appeals further concluded that the defense was put to an actual disadvantage as previously outlined in this decision. Nothing in the evidentiary hearing occurred or was presented that would make these notes less material.
The State claims that this passage shows that the district court required the State to disprove the "speculative conclusions" contained in our prior opinion, rather than requiring Paradis to prove his Brady claim.
The argument lacks merit. The quoted passage follows approximately five pages of analysis in which the district court explains the ways in which the evidence introduced at the evidentiary hearing supports Paradis' claim. In addition, the district court found against Paradis in one important respect precisely because he had "not supported his claim that this information [i.e., the color photographs] was not disclosed to him prior to trial," thus indicating that the court did not relieve Paradis of the burden of proving his claims. Finally, the passage on which the State relies can be interpreted in a way that implies no illicit burden shifting--in referring to our prior conclusions and saying that nothing at the hearing "would make these notes less material," the court apparently meant only that our conclusions were borne out, and not undermined, by what transpired at the hearing.
In sum, the record does not support the argument that the district court shifted the burden of proof to the State.
B. "Sufficiently undermined"
The State argues that because the district court stated that it could not determine that Haws' notes would have "sufficiently undermined the credibility of Dr. Brady or jurisdiction in this case," the court must have erred when it went on to find the notes to be material under Brady.
We do not find the State's argument to be persuasive. An earlier portion of the district court's opinion contains a comprehensive and accurate statement of the standard of review for Brady claims, including determinations of materiality. In addition, the court explicitly found that "there is a reasonable probability that had Notes I and III been disclosed to the defense, the result of the proceeding would have been different, and the suppression of the notes undermines confidence in the outcome of this trial," thereby applying the correct standard for materiality.
While the court did not explain what it meant by "sufficiently undermined," it immediately distinguished that issue from the issue of materiality; the court therefore apparently did not intend that the word "sufficiently" be understood to mean "sufficiently to be material." Thus, the court presumably meant that Paradis did not conclusively prove, or prove by a preponderance of the evidence, that he would have been acquitted if he had access to the notes prior to trial. Thus, the court could not "definitively determine" that the notes would have undermined jurisdiction or Dr. Brady's testimony sufficiently to guarantee acquittal. The State's interpretation of the passage, in contrast, renders the passage patently logically inconsistent, and the State has provided no reason why the passage should be so interpreted.
In sum, the State has not shown that the district court's statement that it could not "definitively determine" that jurisdiction and Dr. Brady's testimony would have been "sufficiently undermined" conflicts with the court's finding that the notes were material under Brady.
The State argues that the district court erred by failing to consider all of the evidence regarding jurisdiction that was introduced at trial. The State claims that the location and condition of Palmer's body differed importantly from the location and condition of Currier's body. The State argues that these facts together with the "incline, the terrain, the fence and the foliage in the area," and the fact that the jury was permitted to view the scene where the bodies were found, adequately supported a finding of jurisdiction.
This argument fails on several levels. First, the district court was clearly aware of the evidence regarding the location and condition of Palmer's body when it was found. All of that evidence is described in the district court's opinion.
Second, the location of Palmer's body was not in dispute--the crucial jurisdictional question was where Palmer was murdered, not where her body was found. Given that her body was found near Currier's (within 20 to 60 feet) and that no one suggested or argued that Currier was murdered in Idaho, the location of her body does not strongly suggest that she was murdered in Idaho. The State has not explained how the distance of 20 to 60 feet between the two bodies, the differences in their condition, the foliage and so forth, or the jury's view of the scene conclusively, or even strongly, supports jurisdiction in Idaho.
Third, materiality under Brady is in any event not evaluated under a sufficiency of the evidence test. Kyles, 514 U.S. at 434-35. That is, the defendant "need not demonstrate that after discounting the inculpatory evidence in light of the undisclosed evidence, there would not have been enough left to convict." Id. Rather, the defendant only needs to show a "reasonable probability" that the trial would have resulted in a different verdict had the evidence been disclosed. Id. (stating that one proves a Brady violation by "showing that the favorable evidence could reasonably be taken to put the whole case in such a different light as to undermine confidence in the verdict"). It is true that if the verdict is supported by overwhelming evidence of guilt, then undisclosed exculpatory evidence that undermines only a small part of the prosecution's case may not be sufficient to undermine confidence in the outcome. But, for the reasons explained above, the evidence that the State pointsto regarding jurisdiction is far from overwhelming.
The district court found that Dr. Brady's testimony was "instrumental in establishing jurisdiction," and the State has not challenged that conclusion. Because the district court also found that the notes, or evidence that they could have led to, would have contradicted and undermined Dr. Brady's testimony at trial, it concluded that confidence in the outcome of the trial had been undermined. The citation of other evidence regarding jurisdiction does not contradict these findings; the State's argument therefore fails.
D. "Speculation" regarding the discovery of admissible evidence
The district court found, and Paradis concedes, that Notes II were not properly Brady material. The district court further found that Notes I and III would not have been independently admissible evidence at Paradis' trial (presumably because of hearsay problems), and that Notes I might not even have been admissible as impeachment evidence (presumably because of even more hearsay problems).4 Nonetheless, the court concluded that Notes I and III could have led Paradis' defense team to discover admissible evidence and that they "appear[ ] to have put the defense team at a substantial disadvantage in preparing their case." The court ultimately found that Notes I and III were material and that the State's failure to disclose them constituted a Brady violation.
The State argues that because the district court based its decision on the bald assertion, unsupported by further analysis, that the notes could have led to the discovery of admissible evidence, the court based its decision on "speculation" in contravention of Wood v. Bartholomew, 516 U.S. 1 (1995).
There is no uniform approach in the federal courts to the treatment of inadmissible evidence as the basis for Brady claims. See generally Felder v. Johnson, 180 F.3d 206, 212 & n.7 (5th Cir. 1999) (discussing various approaches taken by different circuits and collecting cases). Some circuits have held that if evidence is itself inadmissible, then it cannot be material under Brady. See, e.g., Hoke v. Netherland, 92 F.3d 1350, 1356 n.3 (4th Cir. 1996). Others allow that inadmissible evidence can be material under Brady, if it could have led to the discovery of admissible evidence. See, e.g. , Wright v. Hopper, 169 F.3d 695, 703 & n.1 (11th Cir. 1999); Madsen v. Dormire, 137 F.3d 602, 604 (8th Cir. 1998).
The Supreme Court has likewise provided little guidance. In Bartholomew, the Court did not categorically reject the suggestion that inadmissible evidence can be material under Brady, if it could have led to the discovery of admissible evidence. The purported Brady material at issue was two previously undisclosed polygraph tests that were inadmissible under applicable state law, even to impeach. Bartholomew, 516 U.S. at 6. This court concluded that the polygraph tests could have led the petitioner to depose the tested witnesses and thereby obtain exculpatory evidence, but the Supreme Court found this to be "mere speculation," because the test results were consistent with the witnesses' testimony at trial and with the petitioner's guilt. Id. The Court also rejected the suggestion that the polygraph results might have affected the petitioner's trial counsel's preparation, because counsel had himself testified to the contrary at the evidentiary hearing. Id.at 7. But other than its disapproval of such "mere speculation," the Court did not rule out the possibility of basing a Brady claim on inadmissible evidence that could have led to the discovery of admissible evidence.
It appears that our Circuit's law on this issue is not entirely consistent. Compare Coleman v. Calderon, 150 F.3d 1105, 1116-17 (9th Cir. 1998), rev'd on other grounds , 525 U.S. 141 (1998), and Kennedy, 890 F.2d at 1059 ("To be material under Brady, undisclosed information or evidence acquired through that information must be admissible. " (emphasis added)), and id. at 1060-61 (discussing the admissible evidence that the suppressed information at issue could have led to), with United States v. Sarno, 73 F.3d 1470, 1505 (9th Cir. 1995) (stating that nondisclosure of inadmissible evidence does not give rise to a Brady violation) (citing Kennedy).
The instant case does not require resolution of that possible conflict, however, because under Ninth Circuit law "[e]vidence is material if it might have been used to impeach a government witness, because if disclosed and used effectively, it may make the difference between conviction and acquittal." Carriger v. Stewart, 132 F.3d 463, 481 (9th Cir. 1997) (en banc) (citation and internal quotation marks omitted). Evidence can be "used to impeach" a witness even if the evidence is not itself admissible, even to impeach. For example, if Haws' notes record Elliott's hearsay reports of Dr. Brady's hearsay statements, then the notes themselves would not be admissible, even to impeach Dr. Brady. But if Dr. Brady's hearsay statements, reflected in the notes, contradict his in-court testimony, then the notes could be used to impeach Dr. Brady by leading the defense team to call Elliott to testify regarding Dr. Brady's prior inconsistent statements, which, as such, would not be hearsay.5
Because evidence is material if it could have been used to impeach a key prosecution witness sufficiently to undermine confidence in the verdict, the district court's decision must be affirmed. First, although the district court found that Notes III would not have been independently admissible, it did not find that they could not have been used for cross examination. Notes III would have been very useful to Paradis in cross-examining Dr. Brady regarding the labia wound, because they indicated (1) that Dr. Brady saw no evidence of bleeding ("evid no bleeding"), which is not the same as the lack of evidence of "vital reaction" that was described in the autopsy report, and (2) that Dr. Brady thought that there was blood, not "skin slip," at Palmer's nose and ear, and that the blood could have immediately followed her strangulation.6
Second, disclosure of the fact that there was no evidence of a sexual assault ("not sexually assaulted" in Notes I, and "no gross evid. sex. molest." in Notes III) could have been used to impeach Dr. Brady because it would have freed Brown to cross-examine Dr. Brady regarding the labia wound, an issue that the district court found "might have refuted jurisdiction." Dr. Faith testified that Brown did not question Dr. Brady regarding the labia wound because he feared implications of sexual assault. The district court found that the police report given to Brown had actually stated that Palmer was sexually assaulted. The court thus found that the erroneous police report and the nondisclosure of the notes together had a "chilling effect" on Brown's ability to cross-examine Dr. Brady regarding the labia wound. If Brown had had access to the note "not sexually assaulted" in Notes I, he could have used this information to enable him to impeach Dr. Brady by cross examining him regarding the labia wound, a cross examination that would have been guided by the impeachment evidence in Notes III.
Third, the remainder of Notes I could have been used to impeach Dr. Brady, and the evidentiary hearing itself indicated what admissible evidence could have been discovered if the notes had been disclosed. At the evidentiary hearing, Elliott testified that he saw only a small amount of fluid come out of Palmer's lungs and that at the autopsy Dr. Brady said that he could not tell who had died first. Elliott's testimony, if Paradis had known to ask for it (on the basis of "no time of death either" and "spoonful water in lungs" in Notes I), could have directly impeached and undermined Dr. Brady's testimony regarding Palmer's alleged aspiration of water and his opinion that Currier died first. Elliott also testified that it was after the autopsy that he first suggested the idea to Dr. Brady that Palmer had aspirated water and that, at that time, Dr. Brady did not adopt the theory, but said that he would have to "check it out." This testimony, had Paradis known to ask for it (on the basis of "siphon prob" in Notes I), would have undermined the purported scientific basis of the theory by showing that it originated with a police officer rather than a pathologist, was not accepted by Dr. Brady when it was first suggested to him, and was later adopted by Dr. Brady, even though he had received no new medical information. Cf. Kyles, 514 U.S. at 443 (finding an item of undisclosed evidence to be material because it would have "destroy[ed] confidence" in a witness' story by "raising a substantial implication that the prosecutor had coached him to give it").
For all of these reasons, the district court was not "merely speculating" when it concluded that there is a reasonable probability that the result of the trial would have been different if Notes I and III had been disclosed. On the contrary, its conclusion has solid support in the record. Notes III would have been valuable impeachment evidence, part of Notes I would have freed Brown to cross-examine Dr. Brady regarding the labia wound, and the remainder of Notes I would have been useful to impeach Dr. Brady either by cross examination or via Elliott's testimony, as the evidentiary hearing showed.
The State argues that the district court clearly erred when it interpreted the phrase "no time of death either" in Notes I to refer to comparative times of death, rather than to specific times of death of each victim. Haws testified at the evidentiary hearing that he remembered that the discussion at that point in the June 24 meeting was not about which victim died first, but about the specific date on which the victims died. The State argues that if that is what the note means, then it could not have been used to impeach Dr. Brady, because Dr. Brady testified only regarding comparative times of death.
The argument lacks merit, for several reasons. First, Elliott testified that at the autopsy Dr. Brady said that he could not determine which victim had died first. Because Elliott was the only person who attended both the autopsy and the June 24 meeting, his testimony provides evidentiary support for the court's apparent finding that that phrase in Notes I referred to comparative times of death. The court was under no obligation to believe Haws' testimony. In light of Elliott's testimony, the record does not support the State's contention that the court clearly erred.
Second, even if the court did clearly err, the error is of no consequence. If, as the State argues, "no time of death either"actually reflects the fact that Dr. Brady could not determine the date on which Palmer died, then it is still exculpatory evidence--it indicates that Dr. Brady could not determine whether Palmer died on June 21, which is the only date on which there was any established connection between Paradis and the Mellick Road site. Moreover, had the notes been disclosed, they would have led Paradis to depose, or at least to interview, Elliott, because he was the only source of information from the autopsy at the June 24 meeting. And the evidentiary hearing shows what Elliott's resulting testimony would have been--according to Elliott, Dr. Brady was unable to determine even comparative time of death, which impeaches Dr. Brady's testimony at trial.
For all of these reasons, the State's argument fails. The district court did not clearly err in interpreting the notes as it did, but even if it did clearly err, the notes would still be exculpatory and would still have been useful for impeachment.
Because the district court's decision is fully supported by the record, and because the State has presented no meritorious argument that the decision was in error, the district court's judgment conditionally granting the writ of habeas corpus, unless the State grants Paradis a new trial within 120 days, is
Because Notes II are no longer at issue, we omit our previous analysis dealing with those notes.
Brown's fear was based on a police report, furnished to him by the prosecution, which stated that Palmer had been sexually assaulted. As we point out in Part III.D, below, this report later proved to be false.
All of the evidence recited in this section was introduced at the trial by the prosecution.
Notes III were Haws' notes from a conversation he had with Dr. Brady immediately before Gibson's trial. Notes I were Haws' notes from the June 24 meeting. The relevant portion of Notes I appears under the heading "Pathologist report." Because Elliott was the only person at the June 24 meeting who had attended the autopsies or had any information about them, the notes evidently reflect Elliott's report of what he learned by observing, and talking to Dr. Brady, during the autopsies.
The notes could also be used to refresh Dr. Brady's recollection of what he stated at the autopsy.
The district court found that Paradis failed to prove that he had not received color photographs of Palmer's body that displayed the blood on her nose and ear, and concluded that he had thus failed to prove that he did not have access to "this information" prior to trial. The court did not specify what "this information" was, but its analysis does not contradict the observations we have made. In particular, even if Paradis did have the color photographs, they would not have disclosed to him what Notes III would have, namely, that Dr. Brady had said that there was blood, not skin slip, at Palmer's nose and ear, and that the blood could have immediately followed her strangulation.