371 F.3d 353
Randall K. MATAYA, Petitioner-Appellant,
Phillip A. KINGSTON, Warden, Respondent-Appellee.
United States Court of Appeals, Seventh Circuit.
Argued April 13, 2004.
Decided June 3, 2004.
Robert J. Palmer, May, Oberfell & Lorber, South Bend, IN, for Petitioner-Appellant.
James M. Freimuth, Office of the Attorney General, Madison, WI, for Respondent-Appellee.
Before FLAUM, Chief Judge, and POSNER and KANNE, Circuit Judges.
POSNER, Circuit Judge.
After exhausting his state remedies, Wisconsin lifer Randall Mataya sought federal habeas corpus, claiming that he had been convicted in violation of the rule of Brady v. Maryland, 373 U.S. 83, 83 S.Ct. 1194, 10 L.Ed.2d 215 (1963), which requires the prosecution to turn over to the defense evidence in its possession that would be helpful to the defendant, including evidence useful only for impeaching a prosecution witness. United States v. Bagley, 473 U.S. 667, 676-77, 105 S.Ct. 3375, 87 L.Ed.2d 481 (1985).
Pamela Claflin, age 35 and a frequenter of bars, happened one evening to be in a bar named Ma's Place in Manitowoc, Wisconsin, when she was accosted by Mataya, whom apparently she hadn't met before. They left together, amidst indications that she was drunk and he amorous, and drove off in Mataya's car, Claflin leaving hers in the bar's parking lot. Later that night, at some distance from the bar, a man named Cole heard "a very loud, strange, almost terrifying type of a scream," unlike any animal sound he had ever heard; minutes later he saw a car driving at a high speed from the small park-like area at the dead end of his street. As it happens, Cole had been involved in auto racing for many years and had rebuilt car engines hundreds of times, so the police conducted a test in which six automobiles, one of them Mataya's (of course Cole was not told which), were driven past Cole's home one at a time, and he picked Mataya's and one other car as most like the one he'd seen and heard that night.
After midnight on the fatal night Mataya's stepson saw Mataya applying bleach to bloodstains on his white pants and the next morning he saw him cleaning the interior of his car with spot remover. Mataya told him to say nothing of these things to the police.
A week later, Claflin having been reported missing, the police found her corpse in the park at the end of Cole's street, near a pond. She was naked except for her socks, and her body had been shoved under bushes and was almost entirely covered by weeds, grass, and sticks. Her clothes were scattered nearby. Claflin had worked for a cleaning service and according to her employer had carried her customers' keys — 50 or more of them — in a large purse that she had with her the day she vanished. The purse was never found.
Claflin's skull had been fractured by a heavy object, probably a rock. She had also been strangled. There were bite marks on one of her breasts, and a dentist who has testified frequently as an expert witness, after studying Mataya's teeth, testified that the bite marks had probably been made by those teeth, which the dentist described as "remarkable" (and therefore distinctive). One tooth was rotated 30 degrees; another tooth was missing and its absence had caused other teeth to shift in his mouth.
Having been the last person seen with Claflin before her disappearance, Mataya was immediately suspected of being the murderer and was questioned by the police even before the body was found. At first he denied that Claflin had been in his car when he left Ma's Place, but later he admitted she had been but said he'd dropped her off at another bar — but no one at the other bar saw her there. Later still he said he had blacked out after he left Ma's Place and didn't remember anything that had happened afterwards. He told the police that his wife would say he was wearing white pants the day of the murder and didn't come home until midnight, but that these things were untrue. When his step-daughter asked him whether he had murdered Claflin, he said he didn't know.
The state's principal witness was Donald Hertel, and it is in connection with his testimony that the Brady issue arises. Hertel, who admitted on the stand that he had been convicted ten times and that in exchange for his "cooperation and testimony" he was to receive $1,000 plus a favorable letter to the parole authorities, had known Mataya for a decade. Shortly after the murder, Hertel had absconded from a halfway house in which he was supposed to be living and had gone on a burglary spree with Mataya during which Mataya had admitted having killed a woman named Pamela. He had told Hertel, Hertel testified, that "they were making out on the hood of his car and he was twisting her nipple between his teeth and biting on her breast, and she shoved him back and told him to stop it and take her home or she would turn him in for attempted rape. He got mad and shoved her against a tree, she fell down and hit herself, hit her head on a rock, and then — ah — he crawled on top of her and started beating her in the head, in the temple area. He got up off of her, and she was gasping for air, and making weird noises, so her pants around her neck and drugged, and then he knew she was dead after the body went limp.... He dragged her and he told me that he covered her up with twigs and leaves and grass, and took most of her clothes off of her because he wanted to make it look like a rape.... He said he threw her pants away from the body." Hertel further testified that Mataya had told him that the murder had taken place in a "a little wooded area with a pond next to it," that the woman had had a large purse, that Mataya had thrown the purse into the pond and watched it sink weighed down by "a large set of keys," and that he had tried to get the bloodstains out of his white pants by soaking the pants in bleach.
But what Hertel did not tell the jury, and the prosecution did not tell Mataya's lawyer, was that Hertel had made a deal with the prosecution under which four burglary charges against him would be dropped in exchange for his testifying truthfully at Mataya's trial. Had Hertel been prosecuted and convicted of those charges, he might have been sentenced to 40 years in prison. The implications for his freedom are unclear, but may have been great. He was in prison when he negotiated the deal, because his parole on one of his previous convictions had been revoked; we do not know how long he was likely to remain there. He was also under threat of having his parole on another of his convictions revoked; the letter from the prosecutor to the parole authorities was intended to ward off that revocation. Apparently he was not facing any other new charges besides the four burglaries. The dropping of those charges may not have made him a free man immediately, but almost certainly spared him many years of imprisonment.
Mataya's trial lawyer could have used the deal, had he known about it, to further impeach Hertel's testimony. The deal provided a greater inducement to Hertel to play ball with the prosecution than the $1,000 reward plus the supportive letter to the parole authorities. United States v. Williams, 81 F.3d 1434, 1439-41 (7th Cir.1996). Both those inducements were disclosed to the jury; the implication was that they were the only inducements that he'd been offered. Hertel also disclosed his ten convictions, which doubtless reduced his credibility in the eyes of the jurors; yet disclosure of the deal he had struck with the prosecutors over the burglaries would have impeached him even more because the deal had given him a palpable and very substantial incentive to lie if need be, as well as freeing him from any real threat of prosecution for perjury if he did lie.
Brady requires the disclosure to the defendant of evidence in the prosecution's possession that is "material" to the defense; and in the law of evidence "material" just means bearing on an issue in the case. 1 McCormick on Evidence § 185, pp. 637-38 (John W. Strong ed., 5th ed.1999). But the Brady line of cases uses the word in a more demanding sense. Brady evidence is material only if there is a reasonable probability that disclosure to the defense would have resulted in the jury's acquitting the defendant. Banks v. Dretke, 540 U.S. 668, ___, 124 S.Ct. 1256, 1276, 157 L.Ed.2d 1166 (2004); Kyles v. Whitley, 514 U.S. 419, 434, 115 S.Ct. 1555, 131 L.Ed.2d 490 (1995); United States v. Boyd, 55 F.3d 239, 245 (7th Cir.1995).
Usually evidence that the prosecution's principal witness had been offered a strong inducement to testify against the defendant would satisfy the criterion of materiality. And there is no doubt that Hertel was the state's principal witness against Mataya. We are not impressed by the state's argument that the remaining evidence made his conviction a certainty regardless, though we'll note later in this opinion that the evidence of Mataya's guilt, even without his confession to Hertel, was considerable. Nor does the fact that Hertel began talking to the police before there was any agreement to drop the charges against him show, as the state also argues, that the agreement didn't operate as an inducement for him to lie on the stand. A prospective witness might lie to the police in the hope of obtaining concessions yet be unwilling to repeat his lies on the stand unless he obtained very generous concessions in bargaining with the prosecution over the "price" of his cooperation. Indeed from the get-go Hertel demanded as part of the inducement for his testifying for the government that it drop the burglary charges against him. He could not negotiate the deal without revealing to the government at least some of what he would be willing to testify to if the deal went through. This didn't make the revelation independent of the deal — quite the contrary.
But what is unusual about this case, and decisive against the Brady claim, is that Hertel's evidence was self-validating, which makes his motivation to fabricate irrelevant. The concept of self-authenticating evidence is familiar in the law of documentary evidence, Fed.R.Evid. 901, 902; Wis. Stat. §§ 909.01, .02; cf. Fed.R.Evid. 803(6), (8); United States v. Sutton, 337 F.3d 792, 797-98 (7th Cir.2003), though we cannot find any previous case that involved oral evidence.
Hertel may have been willing or even eager to lie in exchange for the dropping of the burglary charges. But we know that he didn't lie. We know this because he revealed details of the crime that, with just a few exceptions, he could have learned only from the murderer. All that he knew from other sources when he started telling the police details of Mataya's confession to him in an effort to strike a bargain were the name of the murder victim, that she had been killed in a park, and that her head had struck a tree (Mataya's wife had repeated to Hertel, what she had learned from the police, that Claflin had been found with bark embedded in her head). Everything else — the heavy purse, the strangulation (we take the garbled statement that "she was gasping for air, and making weird noises, so her pants around her neck and drugged, and then he knew she was dead after the body went limp" to mean that Mataya acknowledged to Hertel strangling Claflin with her pants), the fracturing of her skull, the proximity of a pond, the biting, the white pants, the bleaching, and her being naked beneath a covering of twigs, leaves, and grass — Hertel could have learned only from the murderer. There is no suggestion that he may have learned it from someone, not Mataya, who was the real murderer. Nor is there evidence or reason to believe that the police planted these facts in Hertel's head. And while more of the details may have come from Mataya's wife than she admitted on the stand, her source, like Hertel's, could only have been Mataya, either directly or through her son, Mataya's stepson, who had seen Mataya cleaning his pants and his car and might have told his mother. There is no reasonable possibility that Hertel simply made up the facts that he recounted to the police and that his fabrication just happened to correspond to the truth. The odds against such a coincidence between fiction and fact are astronomical.
In insisting that Hertel's motivations were material in the Brady sense, Mataya's lawyers ignore a distinction that is as old as Aristotle's treatise on rhetoric. We learn there that when the truth of an assertion cannot be verified, the character of the person making it becomes important (but not otherwise). Is he the kind of person who always tells the truth or the kind who yields to a strong inducement to lie? Given Hertel's criminal record, there was no reason to credit him with being a reliable witness or to doubt that he would lie to avoid a heavy prison sentence. This would matter greatly if his testimony had been unverifiable, as in such cases as Giglio v. United States, 405 U.S. 150, 154-55, 92 S.Ct. 763, 31 L.Ed.2d 104 (1972), and Crivens v. Roth, 172 F.3d 991, 998-99 (7th Cir.1999). But it was verifiable simply by comparing what he said with what only someone who had talked to the murderer would have known. Had a known liar found a written confession of Mataya, and a handwriting expert confirmed that the confession was indeed in Mataya's hand, the fact that the confession had been found by a liar would not undermine its veracity. It is the same here. The correspondence between the facts narrated by Hertel and the actual facts of the murder establish that, like the known liar in our example, Hertel was merely the transmission belt for Mataya's confession.
The confession by itself was compelling evidence of Mataya's guilt, and in addition it was corroborated, and not merely by Claflin's having indeed been murdered. Mataya's lawyer picks away at each piece of corroborating evidence (the dentist's, the stepkids', etc.) but in doing so overlooks another important point about proof: that a number of weak proofs can add up to a strong proof. Rowan v. Owens, 752 F.2d 1186, 1188-89 (7th Cir.1984); cf. United States v. Jakobetz, 955 F.2d 786, 793, 798-800 (2d Cir.1992). Suppose, by way of analogy, that someone claims that a coin is so weighted that when flipped it always comes up heads. So it is flipped, and, sure enough, it comes up heads. This is feeble evidence to prove that it always does this because even if the coin were perfectly balanced there would be a 50 percent chance that the first toss would come up heads. So the coin is flipped again and again and again and again and each time it comes up heads. Now the likelihood that the coin is perfectly balanced is much less. For example, in five tosses, the probability that an evenly balanced coin will not turn up tails even once is only 3.125 percent (1-.5).5 Suppose there were a 50 percent chance that the bite marks on Claflin's breast had not been made by Mataya, a 50 percent chance that the car that Cole saw and heard was not Mataya's (here the minimum probability was 50 percent, and the actual probability higher), a 50 percent chance that Mataya was not bleaching bloodstains out of his white pants the night of the murder, a 50 percent chance that he did not make the incriminating statements that his wife and stepdaughter reported, and a 50 percent chance that he did not make the incriminating statements to the police that they reported. Nevertheless the probability that none of these five highly incriminating events or sets of events had occurred would be only 3.125 percent — provided they were independent events in the sense that if one were false this wouldn't make it more likely that another was false as well. Branion v. Gramly, 855 F.2d 1256, 1265 (7th Cir.1988). The percentage figures are arbitrary, and we may assume without having to decide that without Hertel's evidence no reasonable jury could have found Mataya guilty beyond a reasonable doubt. Our point is only that because "it is wrong to view items of evidence in isolation when they point in the same direction," Rowan v. Owens, supra, 752 F.2d at 1188, Hertel's evidence was not only self-validating but also strongly corroborated by the other evidence that the state presented.
Mataya argues that a distinct violation of his rights occurred when the prosecution failed to correct Hertel's false testimony that he had been offered no inducements to testify beyond the reward and the favorable letter. But this violation of the prosecution's constitutional obligations was not distinct from the Brady violation in any realistic sense. Had the prosecution corrected Hertel's falsehood on the spot, as it should have done, that would have cured the Brady violation, United States v. Knight, 342 F.3d 697, 705-06 (7th Cir.2003); United States v. Higgins, 75 F.3d 332, 335 (7th Cir.1996); United States v. Scarborough, 128 F.3d 1373, 1376 (10th Cir.1997), but the failure to correct did not make the violation any more serious. All that Mataya is complaining about is that Hertel's deal with the prosecution was concealed from the jury, and the fact that the prosecution had two opportunities to drop the mask — before trial, when it could have told the defense about the plea bargain, and during trial, when Hertel lied — has no bearing on the prejudice to Mataya from the concealment. That prejudice, as we have been at pains to show, was nil.
Although the cases distinguish as a matter of nomenclature between knowing use of false testimony and the withholding of evidence favorable to the defense, the first being a Napue violation (Napue v. Illinois, 360 U.S. 264, 79 S.Ct. 1173, 3 L.Ed.2d 1217 (1959)), and the second a Brady violation, see, e.g., United States v. Boyd, supra, 55 F.3d at 243-45, the standard is the same. United States v. Williams, supra, 81 F.3d at 1438. And the violations are likely to merge in a case such as this in which the value of the withheld evidence to the defense was its potential utility for impeaching the government's witness as a liar. See, e.g., Giglio v. United States, supra, 405 U.S. at 153-55, 92 S.Ct. 763; cf. Banks v. Dretke, supra, 540 U.S. at ___ n. 11, 124 S.Ct. at 1271 n. 11. When the government's witness testifies, the prosecutor will probably know immediately that he's lying. (We hedge with "probably" because he may have forgotten or never known the impeaching fact that defense counsel is trying to elicit on cross-examination.)
One last point, and we are done. Recall that Hertel was given $1,000 as part of the consideration for his "cooperation and testimony." To pay a witness, other than an expert witness, for his testimony is irregular and in fact is unlawful in federal trials, 18 U.S.C. § 201(c)(2); United States v. Condon, 170 F.3d 687, 689 (7th Cir.1999), though it is claimed that federal prosecutors sometimes do pay witnesses for their testimony, J. Richard Johnston, "Why Is It OK for the Prosecution, But Not the Defense?," 11 Crim. Justice, Winter 1997, p. 21, and in United States v. Boyd, supra, 55 F.3d at 244, we noted the "scandalous" favors with which the government in that case had showered its prize witnesses. The practice of paying witnesses for their testimony apparently is forbidden in trials in the state courts of Wisconsin as well. Wisconsin Supreme Court Rule of Professional Conduct for Attorneys 3.4(b) comment. To pay in money, that is; immunity from prosecution, a lighter sentence, placement in a witness-protection program, and other breaks are lawful coin in this realm. United States v. Condon, supra, 170 F.3d at 688-91; United States v. Singleton, 165 F.3d 1297, 1302 (10th Cir.1999) (en banc).
But it is commonplace to offer money rewards for information leading to the apprehension of a criminal, and at argument the state's lawyer explained that the $1,000 paid Hertel had come from a fund for the payment of such rewards. Yet it was the state itself that in questioning Hertel at trial had denominated the $1,000 as compensation for cooperation and testimony, not merely for information leading to Mataya's arrest, though the information Hertel gave the police was decisive in their decision to arrest Mataya. (They had questioned him shortly after the crime, but had not arrested him.) Paying for testimony, as well as concealing evidence of inducements given to a government witness, are breaches of prosecutorial ethics, but do not, in this case at any rate, undermine the conviction. If anything, the fact that the jury was told that Hertel had been paid for his testimony made him a less credible witness — or rather would have made him such had it not been for the self-validating character of that testimony.