Guzman v. State

941 So.2d 1045 (2006)

James GUZMAN, Appellant,
v.
STATE of Florida, Appellee.

No. SC04-2016.

Supreme Court of Florida.

June 29, 2006. Rehearing Denied October 27, 2006.

William Jennings, Capital Collateral Regional Counsel, Eric C. Pinkard, Assistant CCRC—Middle Region, Tampa, FL, for Appellant.

Charles J. Crist, Jr., Attorney General, Tallahassee, FL, and Kenneth S. Nunnelley, Senior Assistant Attorney General, Daytona Beach, FL, for Appellee.

PER CURIAM.

James Guzman, a prisoner under sentence of death, appeals a circuit court order denying his motion for postconviction relief under Florida Rule of Criminal Procedure 3.851. We have jurisdiction. See art. V, § 3(b)(1), Fla. Const. We affirm the denial of relief.

I. BACKGROUND

In September 1992, James Guzman was convicted of the armed robbery and first-degree murder of David Colvin. He was subsequently sentenced to death. On direct appeal, this Court reversed and remanded for a new trial. Guzman v. State, 644 So.2d 996 (Fla.1994). Guzman was retried in a bench trial in December 1996. He again was convicted of armed robbery and first-degree murder and sentenced to death. This Court upheld the conviction and sentence on direct appeal. Guzman v. State, 721 So.2d 1155 (Fla.1998). In his postconviction proceedings, we affirmed the trial court's denial of all but one of the claims raised.^[1]Guzman v. State, 868 So.2d 498 (Fla.2003). The one denial we did not affirm was Guzman's assertion that the State violated Giglio v. United States, 405 U.S. 150, 92 S.Ct. 763, 31 L.Ed.2d 104 (1972). Instead of addressing the merits of this claim, we clarified the Giglio standard and remanded the claim to the circuit court for further consideration. On remand, the circuit court again denied the claim. Guzman appeals this denial and requests that this Court vacate his conviction and sentence and order a new trial.

The following facts, taken from our earlier opinion, are pertinent to the resolution of Guzman's Giglio claim:^[2]

On August 12, 1991, David Colvin's body was found lying face down on the bed in the motel room where he lived. Colvin had been stabbed nineteen times. A samurai sword that belonged to Colvin was propped up in a light fixture above his bed; however, no blood or fingerprints were found on the samurai sword. The medical examiner determined that Colvin died between 3 p.m. and midnight on August 10.

After Colvin's body was found, police officers interviewed other residents of the motel where Colvin had lived. About a week before the murder, Guzman and Martha Cronin, a prostitute and a crack cocaine addict, had begun living together at the motel. The police interviewed both Guzman and Cronin. Each denied having any information about Colvin's murder. On August 16, 1991, the State published in two local newspapers a reward offer of $500 for information about the case.

The police investigation failed to lead to an arrest until November 23, 1991, when Cronin was arrested on prostitution charges. Cronin volunteered to testify about Colvin's murder in exchange for a deal in her own case. Cronin then told the police that Guzman had confessed to her that he killed Colvin. The police took Cronin to a motel and paid for her room. Cronin used the room for prostitution and used crack cocaine; then she left the motel. The police later rearrested Cronin. On January 3, 1992, the police paid Cronin $500 by money order delivered to the Volusia County jail. The police detective who arranged the payment could not recall when she first discussed the reward money with Cronin.

. . . .

At trial, the medical examiner testified that the weapon used to kill Colvin was a single-edged knife or knife-like object with a slightly curved, heavy blade. The medical examiner could not identify the murder weapon used, but he said that Colvin's samurai sword could have inflicted some of Colvin's wounds and that a survival knife like one owned by Guzman could have inflicted other wounds.

Guzman's fingerprints were on the telephone in Colvin's room. There were blood stains on other parts of the phone, but Guzman's fingerprints on the phone were not bloody. Blood and saliva samples were taken from Guzman, but nothing was matched to anything found in Colvin's room. No other physical evidence connected Guzman to the murder.

Guzman testified at trial that on the day before the murder, Guzman helped Colvin move from one room to another in the motel. Guzman said that he used the phone in Colvin's room at that time and again on the morning of August 10. Cronin confirmed that Guzman telephoned her from Colvin's room.

On the morning of August 10, Guzman and Colvin left the motel in Colvin's car. . . . Guzman testified that he and Colvin returned to the motel at about noon. Guzman said that he gave Colvin's car and room keys back to Colvin and returned to his own room, where Cronin was getting ready to go to work as a prostitute. Cronin left the room at around noon.

Guzman testified that at about 3 p.m., Cronin returned to the room accompanied by Curtis Wallace. Guzman said that Wallace gave him a diamond ring, asking Guzman to trade the ring for crack cocaine. It is undisputed that on August 10, at around 4 p.m. or 5 p.m., Guzman took the ring, which had belonged to Colvin, to a drug dealer named Leroy Gadson. Guzman sold the ring to Gadson for drugs and cash. Guzman testified that he then returned to the room and gave Wallace some of the drugs.

Cronin's testimony at trial contradicted Guzman's. Cronin said that on the morning of August 10 Guzman told her that he was going to drive Colvin to the bank. Cronin stated that Guzman returned to their room at about 11 a.m. and showed her Colvin's car keys and room keys, saying he was going to help Colvin move to another room in the motel. Cronin said she left the room at about 11 a.m. to work as a prostitute, and returned at about 2:30 p.m. She said that at about 3 p.m. Guzman came back to their room, looking upset and carrying a garbage bag that contained white rags. Cronin said that Guzman told her he killed Colvin. She said Guzman told her that Colvin woke up while Guzman was in the process of robbing him, so Guzman hit Colvin in the head and then stabbed him with the samurai sword. Cronin said that Guzman showed her a ring and some cash he had taken from Colvin. Cronin identified the ring at trial. Cronin said that Guzman told her before the murder that Colvin would be easy to rob because he was always drunk and usually had money. Cronin testified that Guzman had said in a separate conversation that if he ever robbed anyone he would kill them, and that Guzman was holding his survival knife when he said this.

Cronin said that when she was arrested for prostitution in November 1991, she offered to tell the arresting officers who killed Colvin. However, Cronin denied that she received any deal for her testimony against Guzman. She said she was taken to a motel room for protection, but that she used the room for prostitution and continued to use crack cocaine, so she got no deal from the State. The detective who paid the $500 to Cronin also testified at trial, stating that Cronin received no deal for her testimony against Guzman.

Guzman's counsel attempted to impeach Cronin by bringing out that she was a prostitute and a drug addict, that she testified against Guzman while she faced charges of prostitution, and that she was angry at Guzman because he was involved with other women. Guzman's counsel also presented the testimony of Carmelo Garcia, who said Cronin told him in February of 1992 that Guzman had not killed anyone and that Cronin admitted she had lied to the police because she had been arrested.

Paul Rogers, a jailhouse informant, corroborated Cronin's testimony against Guzman. Rogers and Guzman shared a jail cell during the spring of 1992. At trial, Rogers testified that Guzman said that he robbed and killed Colvin. Rogers testified that Guzman told him that he used Colvin's key to enter Colvin's room, and that Colvin woke up while Guzman was robbing him. Rogers said that Guzman told him that he hit Colvin in the head with a samurai sword and stabbed him ten or eleven times. Rogers said Guzman confessed that he took Colvin's ring and some cash, cleaned up the sword, and put everything in the dumpster.

Guzman's counsel attempted to impeach Rogers by asking if Rogers had read Guzman's trial papers, which Guzman kept in the cell they shared, but Rogers denied reading Guzman's papers. Rogers also denied learning of the case by reading the newspaper. Rogers admitted that after he initially told police that Guzman confessed to him, Rogers had signed an affidavit saying he knew nothing about Colvin's murder and indicating that he would not testify against Guzman.

Guzman, 868 So.2d at 501-03 (footnote omitted).

In our last opinion, we made the following determination regarding Guzman's request for relief under Giglio:

The first two prongs of the Giglio test are satisfied in this case. [The false testimony prong was satisfied because] [b]oth Cronin and the lead detective on the case testified falsely at trial that Cronin received no benefit for her testimony against Guzman other than being taken to a motel rather than jail when she was arrested. . . . The knowledge prong is satisfied because the knowledge of the detective who paid the reward money to Cronin is imputed to the prosecutor who tried the case.

The only disputed issue with respect to Guzman's Giglio claim is the third prong, which requires a finding that the false testimony presented at trial was material.

Id. at 505 (citation omitted). As previously stated, we clarified the standard to be applied in the third prong of Giglio, stating that a trial court must determine "whether there is any reasonable likelihood that the false testimony could have affected the court's judgment as the factfinder." Id. at 507. Having clarified the legal standard, we remanded this issue to the circuit court for reconsideration. Specifically, we remanded the issue because the circuit court's order denying postconviction relief did not state that there was no reasonable likelihood that the false evidence regarding a $500 payment to Cronin could have affected the judgment of the factfinder. Moreover, we wanted to ensure that the burden of proof was placed upon the right party, in particular, that the State was required to prove that the false testimony was harmless beyond a reasonable doubt. Id.

On remand, the circuit judge who had presided over the bench trial and the subsequent motion for postconviction relief reconsidered the claim. He again denied it. He determined that the testimony on the $500 payment to Cronin was not material under the third prong of Giglio. In reaching this ultimate conclusion, the circuit judge found that the State had met its burden of demonstrating that the false testimony was harmless beyond a reasonable doubt. Specifically, he found that "[t]he State has demonstrated that no Giglio violation occurred due to the ample impeachment and corroboration of Cronin's testimony, and the independent evidence of the Defendant's guilt."

II. ANALYSIS

We agree with the circuit judge that the false statements by Cronin and Detective Sylvester do not meet the materiality prong of Giglio. To explain this result, we begin by stating the standard of review and the applicable law. We then apply this law to the facts of this case.

A. Standard of Review and Applicable Law

"This Court applies a mixed standard of review to Giglio claims, `defer[ring] to the factual findings made by the trial court to the extent they are supported by competent, substantial evidence, but review[ing] de novo'" the application of the law to the facts. Suggs v. State, 923 So.2d 419, 426 (Fla.2005) (quoting Sochor v. State, 883 So.2d 766, 785 (Fla.2004)). Because the question before us addresses only the application of law, we review it de novo.

Again, "[t]o establish a Giglio violation, it must be shown that: (1) the testimony given was false; (2) the prosecutor knew the testimony was false; and (3) the statement was material." Guzman, 868 So.2d at 505. In our previous opinion, we found that Guzman had established the first two prongs and that only the third prong remained at issue. Id. We also made it clear that the State bears the burden of proof on this prong. We stated that it must prove that the presentation of the false testimony was harmless beyond a reasonable doubt. Id.; see also Davis v. State, 915 So.2d 95, 122 n. 7 (Fla.2005), cert. dismissed, ___ U.S. ___, 126 S.Ct. 1649, 164 L.Ed.2d 357 (2006). To meet the harmless error standard, the State must establish "that there is no reasonable possibility that the error contributed to the conviction." State v. DiGuilio, 491 So.2d 1129, 1138 (Fla.1986).

Our previous opinion in this case used the term "reasonable likelihood" in setting out the materiality standard of a Giglio claim. Guzman, 868 So.2d at 506-08. This term appears in the United States Supreme Court opinion in Giglio and elsewhere. However, assessing materiality in terms of a "reasonable likelihood" of a different result may foster confusion because of the similarity of that term to the "reasonable probability" standard applied to claims of suppression of evidence favorable to the defense under Brady v. Maryland, 373 U.S. 83, 83 S.Ct. 1194, 10 L.Ed.2d 215 (1963). In our previous opinion in this case, we acknowledged a lack of clarity in our precedent that led to improper merging of the two standards. Guzman, 868 So.2d at 506. As we stated then, the two standards are not the same. The test of materiality under Brady is whether disclosure of the evidence to the defense would have created a reasonable probability, sufficient to undermine confidence in the outcome, of a different result. Cardona v. State, 826 So.2d 968, 973 (Fla.2002). The same test applies under the prejudice prong of a claim of ineffective assistance of counsel pursuant to Strickland v. Washington, 466 U.S. 668, 694, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). See Rutherford v. State, 727 So.2d 216, 219-20 (Fla.1998); see also Trepal v. State, 846 So.2d 405, 438 (Fla.2003) (Pariente, J., specially concurring).

As we stated in our previous opinion in this case, the test of materiality under Giglio is more "defense friendly" than the Brady materiality test. Guzman, 868 So.2d at 507. In fact, the test under Giglio is the same as the harmless error test of Chapman v. California, 386 U.S. 18, 87 S.Ct. 824, 17 L.Ed.2d 705 (1967), and DiGuilio. See United States v. Bagley, 473 U.S. 667, 680, 105 S.Ct. 3375, 87 L.Ed.2d 481 ("[T]he fact that testimony is perjured is considered material unless failure to disclose it would be harmless beyond a reasonable doubt."); Guzman, 868 So.2d at 508 ("The State bears the burden of proving that the presentation of the false testimony was harmless beyond a reasonable doubt."). The DiGuilio harmless error test requires the State to prove "that there is no reasonable possibility that the error contributed to the conviction." 491 So.2d at 1138. In the interest of greater clarity, the term "reasonable possibility" is preferable to "reasonable likelihood," the term used by the trial judge in this case, in assessing the materiality prong of a Giglio claim. Whatever terminology is used, the dispositive question is whether the State has established beyond a reasonable doubt that the knowing use of perjured testimony, or failure to disclose the perjury once it was discovered, did not affect the verdict.

B. Application of the Law

The circuit judge denied Guzman's claim on the grounds that the false testimony was not material. To affirm this determination, we have to conclude that there is no reasonable possibility that the false testimony of Cronin and Detective Sylvester regarding the $500 reward affected the verdict. That standard has been met here. For the reasons relied upon by the trial court, we conclude that the State has established beyond a reasonable doubt that the false testimony of the witnesses had no effect on the verdict.

The trial court found Cronin's credibility as a witness was amply impeached and that Cronin's testimony incriminating Guzman was independently corroborated and supported at trial. Our own de novo review of the record supports these conclusions. Indeed, impeachment of Cronin was substantial. She testified about her crack cocaine addiction, numerous arrests for prostitution, and agreement to testify against Guzman in exchange for a lesser charge on prostitution. She also acknowledged that she told witness Garcia that Guzman had not killed anyone but that she had lied to the police in accusing Guzman because she had been arrested and that she was angry with Guzman over his relations with other women. The judge also received evidence that the State had paid for a motel room and meals for Cronin.^[3] In light of this ample impeachment, the circuit judge was justified in concluding that Cronin and Detective Sylvester's false testimony regarding the $500 reward was of "limited significance" and "merely cumulative and immaterial." We agree with this conclusion. The addition of the truthful testimony about the $500 reward would not have made a material difference in Cronin's credibility to the finder of fact.

Second, the record fully supports the circuit judge's finding that Cronin's testimony regarding Guzman's guilt was independently corroborated and supported by other record evidence. In particular, the testimony of both Dr. Steiner and Rogers supports the circuit judge's finding. Dr. Steiner, the medical examiner, supported Cronin's testimony at trial by testifying that the samurai sword and the survival knife could have caused the victim's injuries. Rogers, the jailhouse witness who shared a cell with Guzman, testified that Guzman admitted committing the crime. It is also undisputed that, shortly after the murder, Guzman sold the victim's ring to Leroy Gadson, a known drug dealer, for drugs and cash. This evidence of Guzman's guilt, wholly independent of Cronin, supports her testimony. In light of this independent and corroborating evidence, we conclude that there is no reasonable possibility that the false testimony regarding the $500 reward could have affected the judgment of the factfinder.^[4]

Inherent in the above analysis is our conclusion that the State met its burden of establishing that the false testimony was harmless beyond a reasonable doubt.

III. CONCLUSION

The circuit judge faithfully answered the two questions posed in our remand to him. Ensuring that the State bore the appropriate burden of proof, he determined that Guzman failed to meet the materiality prong of Giglio and denied relief. Having reviewed the question de novo, we affirm that decision.

It is so ordered.

PARIENTE, C.J., and WELLS, LEWIS, QUINCE, CANTERO, and BELL, JJ., concur.

ANSTEAD, J., dissents with an opinion.

ANSTEAD, J., dissenting.

Because I conclude that the majority's analysis and holding on the Giglio v. United States, 405 U.S. 150, 92 S.Ct. 763, 31 L.Ed.2d 104 (1972), claim is contrary to controlling precedent of this Court and the United States Supreme Court, I am compelled to dissent. Contrary to this controlling precedent, the majority opinion sends out an alarming signal as to this Court's concern with the presentation of perjured testimony in criminal trials generally, and in death penalty cases in particular.

Giglio Standard

It is apparent that in denying relief, the trial judge below erred both in applying an erroneous standard for Giglio claims and in subjectively deciding that the Giglio error would have made no difference to him personally, as the trial judge who previously tried the case. In essence, the trial court made it the burden of the defendant to prove that there was a "reasonable likelihood" that the uncontested Giglio error did affect the trial court's judgment. Further, the trial court erred in basing its determination solely on the fact that there was other evidence of guilt and impeachment, and in ignoring our warning in DiGuilio that the harmless error test "is not a sufficiency-of-the-evidence, a correct result, a not clearly wrong, a substantial evidence, a more probable than not, a clear and convincing, or even an overwhelming evidence test." State v. DiGuilio, 491 So.2d 1129, 1139 (Fla.1986). As the majority opinion correctly demonstrates, there is no "reasonable likelihood" component of our DiGuilio test. Rather, the test requires the State to prove that there is no reasonable possibility that the error could have affected the verdict. However, having acknowledged that the trial court was in error in applying an erroneous standard, the majority essentially approves this flawed analysis in affirming the order denying relief.

Subjective Standard

Initially, it is important to note that both the State and the trial judge are in error in advocating and applying a subjective standard in evaluating the effect of the Giglio error. The trial judge's order denying the Giglio claim sets out the court's reasoning:

That the Court states in answer to the proper question under Giglio, as presented by the Supreme Court of Florida, as to whether or not there is any reasonable likelihood that the false testimony could have affected the Court's judgment as to the fact finder in this case, that this Court's answer to that question is no, there is not any reasonable likelihood that the false testimony could have affected this Court's judgment as the fact finder in this case.

(Emphasis supplied.) This order makes clear that the trial court not only applied an erroneous "reasonable likelihood" test to the Giglio claim, but also used a subjective standard in applying the test. It is apparent on the face of the order that the "this Court" referred to is the actual trial judge in the case who heard both the original proceeding and the postconviction Giglio claim. The trial court's use of the words "this Court's judgment" leaves no doubt that the judge was referring to himself "as the fact-finder in the case."

Indeed, the State, in its brief, expressly advocates that we follow this erroneous subjective path by asserting that because the postconviction trial judge was the same judge who had heard the original case, his subjective determination that the error would have made no difference to him should control:

In the factual situation presented by Guzman's case, the "subjective-objective" dichotomy is a non sequitor [sic] because in considering the Giglio violation Judge Johnson was not called on to evaluate the possible effect of the error on a jury—he was directed to determine what the effect would have been on his verdict.

Answer Brief of Appellant at 24 (emphasis in original). The State makes clear that it supports the view that the trial judge was indeed called upon to subjectively determine the effect of the error "on his verdict." The State even emphasizes the possessive "his." But that is not the law. Rather, the State and the trial judge have expressly advocated and adopted an erroneous view of the law which the majority now fails to acknowledge and confront.

Bagley

The opinion in the federal case of Bagley v. Lumpkin, 798 F.2d 1297 (9th Cir.1986), demonstrates the errors made here by the trial judge and overlooked by the majority. In Bagley, the defendant requested that the government disclose, pursuant to Brady v. Maryland, 373 U.S. 83, 83 S.Ct. 1194, 10 L.Ed.2d 215 (1963), any impeachment evidence in its possession concerning "any deals, promises, or inducements made to [State] witnesses." Bagley, 798 F.2d at 1298. The government responded with affidavits from two witnesses stating they had not been promised any rewards for their statements. Id. at 1298-99. After conviction, defendant Bagley learned that the two State witnesses had lied when they stated they had not been promised any reward for their cooperation, as they had been paid expense money for their cooperation. Id. at 1299. A government agent recommended that the two witnesses receive $500 each, and they were actually paid $300 each. Id. When this claim was presented in postconviction proceedings, the lower court judge who had presided over the nonjury trial denied Bagley's Brady claim on the issue of materiality, much as the trial court denied Guzman's Giglio claim:

After an evidentiary hearing, a United States magistrate recommended to the district court that it deny Bagley's motion. The district court judge was the same judge who conducted the bench trial and imposed sentence. In its order denying relief, the district court stated that it was "in a unique position of being able to know what effect the disclosure . . . would have had upon the decisions made by this Court in the criminal prosecution." He concluded that "disclosure would have had no effect at all upon its finding that the government had proved beyond a reasonable doubt that defendant was guilty."

Id. On appeal, the Ninth Circuit rejected the trial judge's decision and subjective evaluation of the failure to disclose:

The proper inquiry is an objective one: whether "the Government's failure to assist the defense by disclosing information that might have been helpful in conducting cross-examination" undermines confidence in the outcome of the trial. Therefore, the district judge erred when, in ruling on the section 2255 motion, he stated that the disclosure of the contracts would not have affected his decision. The inquiry is not how this or any other judge, as the trier of fact, would subjectively evaluate the evidence. It is, rather, how the absence of the evidence objectively might have affected the outcome of the trial.

The district court further erred by failing to recognize that the ATF contracts revealed that the witnesses lied under oath. The district court's findings of fact in this action take into account only the extent to which the contracts demonstrate possible bias or prejudice. It is inconceivable that evidence of perjury would not, as an objective matter, affect a factfinder's assessment of a witness' credibility. When the evidence shows that the government's only witnesses lied under oath, it is contrary to reason that confidence in the outcome of the case would not objectively be undermined. This is particularly true here because the lies came from the only witnesses who testified against Bagley and the lies related to the reasons why they testified. Evidence of bias and prejudice is certainly material for impeachment, but lies under oath to conceal bias and prejudice raise the impeachment evidence to such a level that it is difficult to imagine anything of greater magnitude that would undermine confidence in the outcome of any trial.

Id. at 1301 (emphasis added) (citations omitted). The majority simply ignores the Bagley court's holdings on the subjective evaluation by the trial judge; and the patent prejudicial effect of the concealment of "bias and prejudice" and even greater prejudicial effect of "lies under oath to conceal bias and prejudice."

It is apparent in Guzman's case that the trial judge erred in the same two respects discussed in Bagley: first, in applying a subjective test, and second, in concluding that an objective fact-finder would not consider the perjury of witnesses in assessing their credibility. Unfortunately, and contrary to the analysis in Bagley, the majority essentially finds no fault with these errors in approving the trial court's flawed order. Contrast this conclusion with the Bagley court's conclusion that "it is difficult to imagine anything of greater magnitude that would undermine confidence in the outcome of any trial." Id. And, of course, the harmless error standard applied to Giglio claims is an even more defense-friendly standard than the "undermine confidence" standard applied in Bagley.

Harmless Error Standard

Of course, as the above authorities make clear, there is no way that a proper Giglio harmless error analysis can be conducted using a subjective test. But that is not the only problem here. As the majority acknowledges, the trial court's analysis is also flawed because it utilizes a "reasonable likelihood" standard to evaluate the Giglio error rather than the harmless error standard set out in DiGuilio:

The harmless error test, as set forth in Chapman and progeny, places the burden on the state, as the beneficiary of the error, to prove beyond a reasonable doubt that the error complained of did not contribute to the verdict or, alternatively stated, that there is no reasonable possibility that the error contributed to the conviction.

491 So.2d at 1135. And, contrary to the trial court's "other evidence" analysis, we expressly noted in DiGuilio that it mattered not in the assessment of the harmfulness of legal error that there may be other evidence of guilt. Id. at 1136-37. Once again, however, the majority overlooks the trial court's mistaken analysis.

Napue

As with Bagley, the federal courts also provide us with proper guidance on the correct application of the harmless error test in a similar situation. In Napue v. Illinois, 360 U.S. 264, 79 S.Ct. 1173, 3 L.Ed.2d 1217 (1959), a case with almost identical circumstances to those here, the United States Supreme Court mandated a new trial where a prosecutor had given a prosecution witness a vague and indefinite oral promise of possible future help in seeking a sentence reduction, but the prosecution witness later denied any promises of assistance when cross-examined at trial. Id. at 266-67, 79 S.Ct. 1173. Despite the fact of the uncertainty and indefiniteness of the assurance, the Court in Napue declared:

The principle that a State may not knowingly use false evidence, including false testimony, to obtain a tainted conviction, implicit in any concept of ordered liberty, does not cease to apply merely because the false testimony goes only to the credibility of the witness. The jury's estimate of the truthfulness and reliability of a given witness may well be determinative of guilt or innocence, and it is upon such subtle factors as the possible interest of the witness in testifying falsely that a defendant's life or liberty may depend. As stated by the New York Court of Appeals in a case very similar to this one, People v. Savvides, [1 N.Y.2d 554, 154 N.Y.S.2d 885,] 136 N.E.2d 853, 854-855:

"It is of no consequence that the falsehood bore upon the witness' credibility rather than directly upon defendant's guilt. A lie is a lie, no matter what its subject, and, if it is in any way relevant to the case, the district attorney has the responsibility and duty to correct what he knows to be false and elicit the truth. . . . That the district attorney's silence was not the result of guile or a desire to prejudice matters little, for its impact was the same, preventing, as it did, a trial that could in any real sense be termed fair."

Second, we do not believe that the fact that the jury was apprised of other grounds for believing that the witness Hamer may have had an interest in testifying against petitioner turned what was otherwise a tainted trial into a fair one.

Id. at 269-70, 79 S.Ct. 1173 (parallel citations omitted). Unfortunately, and directly contrary to the United States Supreme Court's analysis in Napue, the majority misapplies our DiGuilio harmless error test by simply dismissing the significance of the perjured testimony and relying on other evidence of guilt and impeachment. As the Supreme Court said in Napue, "we do not believe that the fact that the jury was apprised of other grounds for believing that the witness Hamer may have had an interest in testifying against petitioner turned what was otherwise a tainted trial into a fair one." Id. at 270, 79 S.Ct. 1173. Why that reasoning should not apply here, where the facts are even more compelling, is left unexplained by the majority.

This Case

Because the Giglio violation here demonstrates that the State knowingly condoned perjured testimony by both the State's key prosecution witness and the lead police detective investigating the case, there can be little doubt that the State has not carried its burden to objectively demonstrate beyond a reasonable doubt that the error could not possibly have affected a rational fact-finder. See Guzman v. State, 868 So.2d 498, 506 (Fla.2003) (recognizing the State's burden on a Giglio claim to demonstrate harmless error beyond a reasonable doubt).^[5] Obviously, a rational and objective fact-finder would not only have considered the fact that Cronin, the most important witness for the State, was paid for her testimony, but would also have considered the fact that both this crucial witness and the lead detective in the case perjured themselves when they denied under oath that any compensation was paid to Cronin. And, critically, it would have been of especial concern to the fact-finder that this crucial State witness had previously and repeatedly denied any knowledge of the case and only implicated the defendant after the State offered compensation to her.

One need only read this Court's assessment of the evidence and Cronin's critical role in the case to see that an objective fact-finder would have necessarily considered both the undisclosed evidence and the perjury in evaluating the testimony of both important witnesses for the State. See Guzman, 868 So.2d at 501-09. Indeed, this Court itself observed that "the State paid Cronin $500, a significant sum to an admitted crack cocaine addict and prostitute." Id. at 505. Our opinion also details other consideration provided to this critical witness and the numerous twists and turns in her stories, including her initial denials to the police of any knowledge of the murder and her statements to others that Guzman had not killed anyone and that she had lied to the police about his involvement. Id. at 501-04. The bottom line is that Cronin was the key witness in the case and the credibility of her testimony was critical to the State's case against Guzman. After making this plain in our prior opinions, we ignore it today.

In addition, the majority mischaracterizes the nature and weight of the other evidence in the case. For example, the majority treats the medical examiner's testimony as particularly important when the examiner actually testified that he could not identify the defendant's knives as the murder weapons, and that any knife or knife-like object three to four inches in length could have inflicted the fatal wounds. (The weapon that inflicted the wounds could have been "any knife three to four inches at least in length or knife-like object.") Of course, we know there are thousands, if not millions, of knives of such length in this country. As to the jailhouse informant, we know this witness was a seven-time convicted felon who had access to the defendant's court records about the case in his cell and who also swore under oath that "Guzman never confessed to me." As to the ring, Guzman testified at trial that he got the ring from Curtis Wallace, another suspect in the case, and sold it for $250 and a quantity of cocaine which he shared with Wallace and Cronin. In fact, only Cronin testified that Guzman got the ring from the victim. It is apparent that this "other evidence" of guilt is hardly overwhelming and is insufficient to render the perjury of Cronin and Detective Sylvester immaterial.

Conclusion

Ironically, the majority's analysis makes it sound like Cronin's testimony was so severely impeached that her credibility could not have been further undermined. In other words, it could not have gotten any worse. Yet, at the same time, it is apparent that it is this critical witness's testimony that has been relied upon at all levels of judicial scrutiny to sustain both a conviction of premeditated murder and a sentence of death. There is something wrong with this picture. Today's majority has simply chosen to ignore our own harmless error law and the controlling pronouncements from our nation's highest Court in Napue concerning the effect of perjury of important prosecution witnesses on the fairness of a trial. Because our law and these pronouncements should especially be honored in death penalty proceedings, I respectfully dissent.

NOTES

[1] Guzman's initial postconviction motion was filed pursuant to Florida Rule of Criminal Procedure 3.850. On remand, his motion was reclassified under rule 3.851, Collateral Relief After Sentence of Death Has Been Imposed and Affirmed on Direct Appeal, pursuant to a change in the rules of criminal procedure.

[2] A full statement of the underlying facts can be found in our recent opinion in Guzman, 868 So.2d 498.

[3] Notably, during the evidentiary hearing on the initial postconviction motion, Guzman's defense counsel revealed that, prior to trial, Guzman made him aware of allegations that Cronin had received the $500. Defense counsel chose not to pursue discovery or questioning at trial regarding the allegations.

[4] In part, the dissent asserts that the trial court judge erred by subjectively deciding that the Giglio error made no difference to him as the trier of fact. Dissenting op. at 1053. Nothing in the record supports this assertion. The findings of fact and conclusions of law in this trial judge's September 2004 order indicate that his determination was the objective one required by Giglio as articulated in our opinion remanding this cause to him. The order reads in pertinent part:

That the Court states in answer to the proper question under Giglio, as presented by the Supreme Court of Florida, as to whether or not there is any reasonable likelihood that the false testimony could have affected the Court's judgment as to the fact finder in this case, that this Court's answer to that question is no, there is not any reasonable likelihood that the false testimony could have affected this Court's judgment as the fact finder in this case.

Unlike Bagley v. Lumpkin, 798 F.2d 1297 (9th Cir.1986), where the trial judge used his "unique position" as the trier of fact to reach his conclusion, the trial judge here made no such error in his analysis of the impact of the false testimony in this case.

[5] See also State v. DiGuilio, 491 So.2d 1129, 1135 (Fla.1986) (stating that the burden is on the State to demonstrate beyond a reasonable doubt that there is no reasonable possibility that the error could have contributed to the verdict).