Opinion
No. 85 Civ. 4014(CBM).
April 29, 1988.
Henry Putzel, III, Obermaier, Morvillo Abromowitz, P.C. by Alan J. Brudner, New York City, for petitioner Walter Sanders.
Robert M. Morgenthau, Dist. Atty., New York County by Marc Frazier Scholl, Robert M. Raciti, New York City, for respondent James E. Sullivan.
OPINION
By opinion and accompanying order filed January 8, 1988, this court denied petitioner's motion for a writ of habeas corpus pursuant to 28 U.S.C. § 2254(a) (1982). Counsel for petitioner filed a notice of appeal in the Court of Appeals for the Second Circuit on February 8, 1988, and now request, by letter dated March 18, 1988, that this court permit the appeal to proceed by issuing a certificate of probable cause, as required by Fed.R.App.P. 22(b).
Our Court of Appeals has held that "the standard of probable cause to appeal requires the district court to find that the petition is not frivolous and that it presents some question deserving appellate review." Alexander v. Harris, 595 F.2d 87, 91 (2d Cir. 1979). Having examined the two grounds for appeal presented by petitioner's counsel, this court holds that the second ground does not present a question deserving appellate review. As to the first ground, although this court disagrees with counsel's proposed interpretation of the applicable law, the court also believes that interpretation is not frivolous. In consequence, the court grants petitioner's request, and, in conformity with Rule 22(b), now sets out the reasons for both its conclusions.
Four years after Alexander, the Supreme Court held in Barefoot v. Estelle, 463 U.S. 880, 893, 103 S.Ct. 3383, 3394, 77 L.Ed.2d 1090 (1983) that "a certificate of probable cause requires petitioner to make a `substantial showing of the denial of [a] federal right'" (quoting Stewart v. Beto, 454 F.2d 268, 270 n. 2 (5th Cir. 1971), cert. denied, 406 U.S. 925, 92 S.Ct. 1796, 32 L.Ed.2d 126 (1972)). It is not clear whether this standard is different from that set forth in Alexander. If anything, it is more stringent. The Second Circuit has not reconsidered Alexander since Barefoot, so that we cannot say whether it regards Barefoot as overruling the Alexander standard by implication. For the purposes of this opinion, this court assumes that Alexander governs.
Recantation of Testimony
Petitioner advanced four grounds for habeas relief in this court, and now renews the fourth. At trial, the State's principal witness, Carmelo Perez, testified that petitioner and one Omar Sabir accosted him in his hallway, and that petitioner and Sabir fired their guns in the general direction of Irma Semiday, Perez's common law wife, when she peeked out of their apartment to see what the disturbance in the hall was. The State's theory was that a shot from petitioner's gun killed Sabir. Later, while petitioner and Perez were coincidentally incarcerated at Sing, petitioner elicited a recantation from Perez that claimed that the fatal shot was fired by Irma Semiday, and that Perez lied at trial in order to protect her.
This court, noting a wide range of authority in its support, held that recantation of a material witness' trial testimony could establish a due process violation only if it was also established that the prosecution's use of the perjured testimony was knowing and intentional. Petitioner's counsel now renew the argument they presented in their Proposed Findings of Fact and Conclusions of Law that "under the circumstances of this case, the appropriate standard for a violation of due process as a result of the recantation of a material witness does not require that the prosecutor be aware of the perjured trial testimony." Letter of Alan J. Brudner, Esq., March 18, 1988.
Of course, the court must be at least reasonably satisfied that the testimony at trial was in fact perjured, rather than that the recantation is false. If the testimony is true and the recantation false, there can be no due process problem.
In support of this view, petitioner's counsel cite United States ex rel. Sostre v. Festa, 513 F.2d 1313, 1317 (2d Cir.), cert. denied, 423 U.S. 841, 96 S.Ct. 72, 46 L.Ed.2d 60 (1975), which relies on the three-pronged test for due process scrutiny of recanted trial testimony set out in Larrison v. United States, 24 F.2d 82, 87-88 (7th Cir. 1928):
[A] new trial should be granted when,
(a) The court is reasonably well satisfied that the testimony given by a material witness is false.
(b) That without it the jury might have reached a different conclusion.
(c) That the party seeking the new trial was taken by surprise when the false testimony was given and was unable to meet it or did not know of its falsity until after the trial.
Petitioner misapprehends the status of the Larrison test in the Second Circuit. Moreover, even assuming the Larrison test to apply, the court finds that Perez's recantation does not provide the basis for a new trial.
Petitioner originally contended that the prosecutor in his state court trial indeed knew that the testimony of Carmelo Perez was perjurious. Petitioner withdrew this contention after this court held an evidentiary hearing on this and other questions: "The Petitioner concedes following the evidentiary hearing that there is no substantial evidence to establish that the perjured testimony was used knowingly." Petitioner's Proposed Findings of Fact and Conclusions of Law at 28. Nonetheless, as counsel's letter states, it is still petitioner's view that in light of Larrison the use of Perez's testimony violated petitioner's right to due process.
Our Court of Appeals in Sostre used the Larrison test without commenting on whether the prosecutor must be aware of the allegedly perjurious character of the trial testimony for due process to be violated. Thus, this court construes petitioner's argument as one from silence: because the Larrison test does not mention prosecutorial knowledge, such knowledge is not, under that test, a requirement for a due process violation. Thus construed, petitioner's argument fails because it fails to appreciate the limitations the Second Circuit has placed on Larrison. In United States v. DeSapio, 435 F.2d 272 (2d Cir. 1970), cert. denied, 402 U.S. 999, 91 S.Ct. 2170, 29 L.Ed.2d 166 (1971), the Second Circuit noted the existence of an alternative. If one treats a recantation as newly discovered evidence, the test of whether a new trial is required is whether this new evidence "is so material that it would probably produce a different verdict, if the new trial were granted." 435 F.2d at 286 (quoting Berry v. State, 10 Ga. 511, 527 (1851)) (emphasis supplied). Larrison, the court noted, imposed the weaker condition that the jury might reach a different verdict. The court explained the difference by endorsing the Government's observation that " Larrison dealt with a situation where a principal prosecution witness had made an affidavit (later repudiated) that he had given false testimony at the instance of Post Office inspectors," and stated that "We are inclined to agree that the less severe Larrison test should be confined to cases of prosecutorial misconduct." Id. n. 14.
This holding moved up from the footnotes in United States ex rel. Rice v. Vincent, 491 F.2d 1326 (2d Cir.), cert. denied, 419 U.S. 880, 95 S.Ct. 144, 42 L.Ed.2d 120 (1974), in which our Court of Appeals observed that "[t]he standard adopted by the federal courts in determining whether a witness' recantation necessitates a new trial, at least where allegations of police or prosecutorial misconduct are made, is that stated in Larrison v. United States. . . ." Id. at 1331 (footnote omitted) (emphasis supplied). This limitation of Larrison to cases involving prosecutorial misconduct became more firmly entrenched in the Second Circuit's cases after Sostre with United States v. Stofsky, 527 F.2d 237, 245-46 (2d Cir. 1975), cert. denied, 429 U.S. 819, 97 S.Ct. 66, 50 L.Ed.2d 80 (1976). In this, its most extended discussion of the Larrison test, the Court of Appeals noted its own "increasing discomfort with the Larrison test in cases that do not involve prosecutorial misconduct," id. at 245 (citing cases), observed that strictly applied the Larrison test requires reversal in cases of perjury by a material witness as to minor matters, id. at 245-46, and pointed out that courts that purportedly embraced Larrison (including, the court might have mentioned, Larrison itself) scarcely ever reversed a conviction on Larrison grounds, id. at 246 n. 10.
In fact, it is arguable that in Stofsky, in which there was no allegation that the prosecution had knowingly used perjured testimony, the Court of Appeals rejects the Larrison test even for cases that do involve allegations of prosecutorial misconduct: "However, rather than adopt the Larrison test and violate it in application, we believe . . . that the time-honored `probability' standard is the more appropriate one for determining whether perjury calls for a new trial." 527 F.2d at 246.
Judge Leval, in a discussion that tends generally to undermine petitioner's reliance on Sostre, recently noted that not only is the Larrison test limited to cases of prosecutorial misconduct, but that United States ex rel. Rice v. Vincent, decided a year before Sostre, "indicates that the requirements of due process applicable to the states may be less stringent than the Larrison standard." United States ex rel. Scarincio v. Coughlin, 657 F. Supp. 433, 436 n. 2 (S.D.N.Y. 1987) (citing Rice, 491 F.2d at 1332). To the degree that this is true, petitioner's reliance on Larrison carries that much less weight.
If, however, the Second Circuit endorses Larrison only in cases involving prosecutorial misconduct, necessarily it has engrafted a knowledge requirement on the Larrison test. For in a Larrison case the misconduct, if any, attributed to the prosecutor is precisely the knowing use of perjured testimony. Here, however, in light of this court's evidentiary hearing, petitioner has disclaimed any allegation that the state prosecutor knowingly used perjured testimony. In any event, petitioner makes no other allegation of prosecutorial misconduct in this specific connection. Therefore, petitioner himself has foreclosed application of the Larrison test, under the governing law of this Circuit, to this case.
Even if Larrison were apposite, this court would be compelled to hold that petitioner has not satisfied its threepronged test. Petitioner admits as much in his Proposed Findings of Fact and Conclusions of Law: "Obviously, at trial Sanders believed that Perez and Semiday were lying and so testified; he was not taken by surprise." Proposed Findings and Conclusions at 30. Petitioner maintains, however, that this shows that "[t]he third part of the Larrison test must be viewed as inapplicable to the facts at bar." Id. The argument is that "it was only after Semiday had died that Perez was willing to tell the truth. Thus, Sanders had to wait until after the trial — indeed, until now — for the opportunity to exculpate himself." Id. This argument rests on a misreading of the third prong of the Larrison test. Petitioner clearly interprets the third prong to require a new trial whenever it is discovered after trial that a witness committed perjury, no matter what attitude the defendant took toward that witness' testimony at trial. This, however, would make the third prong surplusage. For the Larrison test is a test of recantations, and, assuming a recantation to be true, the defendant will always learn what he did not know at trial, namely that the witness' testimony was false. Thus, on petitioner's interpretation the third prong of the Larrison test is automatically met by the mere fact that a witness has recanted.
This court declines to impose such an interpretation on Larrison. Surely the intuition that underlies Larrison is that a recantation raises due process concerns when the defendant failed to meet the recanted testimony at trial — something that can happen for two reasons. First, the testimony may have occasioned such surprise at trial that defendant was unable to respond. Second, defendant may not have opposed the recanted testimony at trial because at the time of trial he believed it to be true. The potential prejudice to a defendant in either of these cases is clear. The present case is quite different. Here, although petitioner may justifiably say that he did not know at trial that Perez's testimony was false, he certainly believed it to be. Petitioner's counsel cross-examined Perez in some detail about the gunshots. Transcript at 149-53, People v. Sanders (Indictment No. 5530/80). Moreover, although petitioner's counsel did not try to impeach Perez's account of the shooting directly, she attacked his credibility by introducing his prior statements to the police, e.g., id. at 127, and before the grand jury, e.g., id. at 140-45. Thus, this court cannot hold that petitioner was unable to meet Perez's testimony so as to raise due process problems under Larrison, even if Larrison were apposite under the law of this Circuit.
This disposition of petitioner's contention makes it unnecessary for this court to consider the first and second prongs of the Larrison test. This court is persuaded by Second Circuit precedent, in particular DeSapio, Vincent, and most of all Stofsky, that our Court of Appeals would reject petitioner's proposed attempt to apply Larrison to his case. Nonetheless, because the Second Circuit has not ruled squarely on the question whether Larrison, when it applies, requires knowing prosecutorial misconduct, this court holds that question is deserving of appellate review. Petitioner's first proposed ground is nonfrivolous, and a certificate of probable cause to appeal shall issue.
Cumulative Prejudice
Petitioner's counsel inform the court that petitioner's second ground for appeal will be that "even if no single flaw at the trial rose to the level of a Constitutional deprivation, nevertheless the cumulative effect of all the problems at trial was to deprive Mr. Sanders of due process." Letter of Alan J. Brudner, Esq., March 18, 1988. Petitioner so contended in his Proposed Findings and Conclusions. See id. at 35. Strangely, in that submission petitioner cited only a New York Court of Appeals case for the proposition that individually harmless errors could cumulatively cause a due process deprivation. Petitioner does not, either there or in counsel's letter, call this court's attention to federal habeas cases that assert this proposition.
Petitioner also argued that a specific alleged error, the trial court's discharge of a juror who had expressed doubts about her ability to deliberate objectively, pushed the trial over the bounds permitted by due process. Proposed Findings and Conclusions at 33-34. This court fully discussed petitioner's claim that the discharge of the juror was error, and rejected it. Sanders v. Sullivan, 701 F. Supp. 1000, 1002-1003 (S.D.N Y 1988).
This court has discovered only one Second Circuit case that reversed a conviction because of the cumulative effects of individual trial errors. United States v. Alfonso-Perez, 535 F.2d 1362 (2d Cir. 1976). Of course, one case is enough: Alfonso-Perez establishes that the Second Circuit observes a "cumulative error" rule. Our problem is that it does not tell us what that rule is: it did not set forth a test for determining when the cumulative effects of individual trial errors call for reversal of a conviction or habeas relief. The cases in other Courts of Appeals that do purport to set forth such a test are surprisingly few, and it is not clear to this court that they are consistent. Thus, in United States v. Jones, 482 F.2d 747 (D.C. Cir. 1973), the Court of Appeals for the District of Columbia Circuit stated that "although certain errors standing alone might be insufficient to overcome a verdict, these errors may exert a cumulative effect such as to warrant reversal. The critical injury is an analysis of the `probable impact, appraised realistically, of the particular [errors] upon the jury's fact-finding function.'" 482 F.2d at 749-50 n. 1 (quoting United States v. Wharton, 433 F.2d 451, 457 (D.C. Cir. 1970)). Judge Nathaniel Jones of the Court of Appeals for the Sixth Circuit, citing Jones among other cases, has steadfastly argued for a cumulative error rule that, though consistently stated in his decisions, is not obviously equivalent to that of Jones. See Matlock v. Rose, 731 F.2d 1236, 1242 (6th Cir. 1984) ("Errors that might not be so prejudicial as to amount to a deprivation of due process when considered alone, may cumulatively produce a trial setting that is fundamentally unfair."), cert. denied, 470 U.S. 1050, 105 S.Ct. 1747, 84 L.Ed.2d 812 (1985); Payne v. Janasz, 711 F.2d 1305, 1316 (6th Cir.) (Jones, J., dissenting) ("Errors which standing alone may be deemed harmless or insufficiently prejudicial to amount to a denial of due process may cumulatively produce a trial setting which is fundamentally unfair."), cert. denied, 464 U.S. 1019, 104 S.Ct. 552, 78 L.Ed.2d 726 (1983); Walker v. Engle, 703 F.2d 959, 963 (6th Cir.) ("Errors that might not be so prejudicial as to amount to a deprivation of due process when considered alone, may cumulatively produce a trial setting that is fundamentally unfair."), cert. denied sub nom. Walker v. Marshall, 464 U.S. 962, 104 S.Ct. 396, 78 L.Ed.2d 338 (1983).
Alfonso-Perez involved a direct appeal from a federal district court conviction, not a collateral attack on a state proceeding by way of habeas review. This court assumes that this distinction is immaterial for present purposes.
See also, e.g., United States v. Araujo, 539 F.2d 287, 292 (2d Cir.), cert. denied sub nom. Rivera v. United States, 429 U.S. 983, 97 S.Ct. 498, 50 L.Ed.2d 593 (1976) (rejecting cumulative-error argument in particular case but noting that "[u]nder other circumstances, there might be force to this argument.").
The lack of a generally accepted test for cumulative error is dramatically illustrated by Payne v. Janasz, in which the majority opinion was written by Senior District Judge Robert Van Pelt of the District of Nebraska, who ten years before had written the majority opinion in Jones, on which Judge Jones has relied in articulating his rule of fundamental fairness — nowhere more explicitly than in his dissent from Judge Van Pelt's opinion in Payne.
In view of the above, if this court had held that multiple errors had been committed in petitioner's trial, but that those errors were harmless, this case might well raise a question deserving of appellate review, justifying the issuance of a certificate of probable cause. That was not the holding of this court's January 8, 1988 opinion, however. That opinion differs from all the cases discussed herein in one crucial respect: this court has already found that petitioner's allegations of error are without substance. The January 8, 1988 opinion held that the trial judge simply did not commit error in those instances in which petitioner alleged he did. The cumulative-error rule, whatever form it takes, can only come into play after errors have been discovered; if no one error requires reversal, the whole body of error is to be assessed for prejudicial effect. If, in the January 8, 1988 opinion, this court had agreed with petitioner that there had been errors in the state court trial, and if it had found those errors individually harmless, it would then have gone on to determine whether their cumulative effect required reversal, as petitioner contended at the time. This court did not agree with petitioner, however. Finding no error in the conduct of the trial, this court had no occasion to apply cumulative-error considerations. That is why the question was not addressed in the January 8, 1988 opinion: it did not arise. Petitioner should not now be allowed an opportunity to reargue the question under the guise of a request for a certificate of probable cause.
For the reasons stated above, the court concludes that petitioner's Larrison claim raises a question worthy of appellate review, but that his cumulative-error claim does not. The court therefore grants petitioner's application for a certificate of probable cause.