Charles Hemstreet v. Charles Greiner, Superintendent

367 F.3d 135 | 2d Cir. | 2004

Lead Opinion

Judge B.D. PARKER dissents in a separate opinion.

OAKES, Senior Circuit Judge.

Petitioner Charles Hemstreet seeks ha-beas relief on the grounds that he received ineffective assistance of counsel from both his trial and appellate counsel, leading to his conviction and incarceration for second-degree murder. Hemstreet argues that *137counsels’ failure to pursue the issue of the state’s intimidation of a crucial defense witness prejudiced the outcome of his trial and appeal. The United States District Court for the Southern District of New York, Charles L. Brieant, Judge, granted Hemstreet’s habeas petition, finding that Hemstreet’s Sixth Amendment right to effective counsel was violated when appellate counsel failed to raise on appeal the issue of trial counsel’s performance with respect to the intimidated witness. We agree with the district court that Hemstreet was denied a fair trial and that his appellate counsel’s performance was deficient and prejudicial. Accordingly, we affirm the grant of a writ of habeas corpus to Hem-street.

BACKGROUND

In January 1998, petitioner Hemstreet was convicted of the second-degree murder of his business partner, Kenneth Hiep. Hiep’s body, which had numerous stab wounds, was found in a state park in Rock-land County in February 1992. Three days earlier, on the last day he was seen, Hiep was with Hemstreet and Hemstreet’s friend, Patrick Bentz, at the Old Fashioned Bar in Nyack, New York. At trial, it was the prosecution’s theory that when the three left the bar that evening, Hemstreet and Bentz drove Hiep to a secluded place, stabbed him repeatedly, then threw him over an embankment in the state park. The defense theory was that Hemstreet and Bentz dropped Hiep off at the end of his driveway, then drove to Hemstreet’s house and dropped him off, at which point Bentz, driving Hemstreet’s car, returned to pick up Hiep. Bentz and Hiep then went out partying, after which Bentz killed Hiep.

Bentz was tried and convicted for Hiep’s murder. Hemstreet was also convicted, but that conviction was reversed in 1996 due to grand jury error. Prior to Hem-street’s retrial in New York Supreme Court in 1997, his trial counsel learned that a witness named Jeanette Bucci, who at the time of Hiep’s murder was a barmaid at a nightclub called Lace, had given exculpatory evidence in 1992 to detectives investigating the case. Bucci swore in a 1997 affidavit that she had told the detectives that she had seen Hiep and Bentz together at Lace, without Hemstreet and after the time at which Hemstreet said he had been dropped off by Bentz.

Hemstreet’s trial counsel named Bucci as a potential witness in the retrial and asked the prosecution to turn over any statements Bucci had made to the police in 1992. A few days later, trial counsel informed the court that Bucci had refused to meet with him because detectives from the prosecutor’s office had visited her mother and sister and warned them that, if Bucci testified, “they were in for a lot of trouble.” Although trial counsel asked the court to stop such threats, counsel did nothing further in response to the fact that Bucci had been intimidated into not testifying. Following the retrial, Hemstreet was again convicted and sentenced by the New York Supreme Court to a term of 25 years to life.

On direct appeal, Hemstreet’s appellate counsel did not raise the claim that, in light of trial counsel’s failure to seek a remedy for the state’s intimidation of a witness, Hemstreet had been denied a fair trial. Hemstreet’s conviction was affirmed by the Appellate Division on March 27, 2000. See People v. Hemstreet, 270 A.D.2d 499, 706 N.Y.S.2d 337 (2d Dept.2000).

In August 2001, Hemstreet applied for a writ of error coram nobis in the Appellate Division, arguing that he was deprived of effective assistance of counsel because his appellate counsel had failed to raise the issue of trial counsel’s performance with *138respect to the intimidation of Bucci.1 Hemstreet’s petition was denied in January 2002 in a brief order that did not include any factual findings. See People v. Hemstreet, 290 A.D.2d 458, 788 N.Y.S.2d 583 (2d Dept.2002).

Hemstreet then filed his habeas petition in federal court, relying on the same ineffective assistance of counsel argument that he made in state court. Pursuant to Sparman v. Edwards, 154 F.3d 51 (2d Cir.1998), the district court ordered Hem-street to request a response to his claims from both his trial and appellate counsel. Trial counsel responded, stating that he could not “with any degree of certainty assert why Ms. Bucci was not called as a witness during the course of the trial.” Appellate counsel did not respond.

On November 14, 2002, the district court granted Hemstreet’s habeas petition, holding that appellate counsel violated Hem-street’s Sixth Amendment rights by failing to raise valid claims that Hemstreet had been denied a fair trial by virtue of trial counsel’s ineffectiveness. The district court found that trial counsel’s failure to seek relief for the intimidation of Bucci, a crucial defense witness, was so deficient and prejudicial as to deprive Hemstreet of a fair trial under Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). As the district court stated:

No reasonable basis exists for police to get Bucci’s name off the defense trial witness list and then visit her at her home with force and arms, with the intent and foreseeable result that she would become suddenly unavailable to the defense. This is a classic situation of actions tending to prevent a fair trial. No plausible basis in trial tactics could justify failure to pursue the issue by demanding a hearing in the trial court, and no tactical decision can justify failure to raise the issue on direct appeal.

The district court therefore concluded that the Appellate Division’s rejection of Hem-street’s ineffective assistance of counsel claim was an unreasonable application of clearly established federal law under 28 U.S.C. § 2254(d)(1) and that Hemstreet was entitled to habeas corpus relief. The state respondent subsequently appealed.

DISCUSSION

When considering on appeal a district court’s grant of a habeas petition, we review de novo the court’s legal conclusions and review its factual findings for clear error. Anderson v. Miller, 346 F.3d 315, 324 (2d Cir.2003). Here, because the state court coram nobis decision did not include any factual analysis, the district court was not required to give deference to state court factual findings as is typically required. See Channer v. Brooks, 320 F.3d 188, 195 (2d Cir.2003) (per curiam) (“[W]here a state court does not resolve a question of fact, no presumption of correctness can possibly attach with respect to that issue.”). We therefore must determine whether the district court’s determinations of fact were clearly erroneous, such that we are left with “the definite and firm conviction that a mistake has been committed.” Ortega v. Duncan, 333 F.3d *139102, 106-07 (2d Cir.2003) (internal quotation omitted).

Additionally, as with all habeas petitions brought after the effective date of the Anti-Terrorism and Effective Death Penalty Act of 1996, a federal court cannot grant habeas in a case in which there was an adjudication on the merits in a state court proceeding unless the adjudication “resulted in a decision that was contrary to, or involved an unreasonable application of, clearly established Federal law as determined by the Supreme Court of the United States.” 28 U.S.C. § 2254(d)(1) (2003); see also Williams v. Taylor, 529 U.S. 362, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000). In this case, the district court correctly found that the state court’s denial of Hemstreet’s coram nobis application was an adjudication on the merits. The parties do not dispute this finding. Our review is thus focused on whether the state court’s denial rested upon an unreasonable application of Strickland, which is the relevant clearly established federal law. See Aparicio v. Artuz, 269 F.3d 78, 94 (2d Cir.2001).

The test for ineffective assistance of counsel established in Strickland requires a defendant to show that counsel’s performance “fell below an objective standard of reasonableness” and that “there is a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different. A reasonable probability is a probability sufficient to undermine confidence in the outcome.” 466 U.S. at 688, 694, 104 S.Ct. 2052. This two-prong test applies to the performance of appellate counsel as well as trial counsel. Aparicio, 269 F.3d at 95 (citing Evitts v. Lucey, 469 U.S. 387, 396-97, 105 S.Ct. 830, 83 L.Ed.2d 821 (1985)). To determine that the state court applied Strickland in an unreasonable manner, we cannot simply conclude that the court’s application was erroneous; rather, we must find that there was “ ‘some increment beyond error.’ ” See Eze v. Senkowski, 321 F.3d 110, 124-25 (2d Cir.2003) (quoting Francis S. v. Stone, 221 F.3d 100, 111 (2d Cir.2000)).

Appellant argues that the state court’s decision was consonant with Strickland because Hemstreet’s appellate counsel made a reasonable strategic choice not to present an ineffective assistance of counsel claim and because Hemstreet cannot demonstrate that the outcome of his appeal would have been different if the claim had been raised. Appellant also argues that trial counsel’s performance was effective because it involved a tactical decision not to call Bucci as a witness. Because the crux of Hemstreet’s claim with respect to his appellate counsel’s performance concerns the performance of his trial counsel, we address that issue first.

The record shows that, prior to trial, Hemstreet’s counsel knew that Bucci had material and exculpatory information regarding where, when, and with whom the murder victim was last seen. Bucci was not called as a witness at trial. The district court found as a matter of fact that Bucci became unavailable as a witness to Hemstreet because she had been threatened by detectives from the prosecutor’s office. The district court also made the complementary factual finding that trial counsel was not employing trial strategy when he did not call Bucci as a witness.

The record does not definitively establish whether the prosecution intimidated Bucci. At best, the record reflects that Bucci refused to meet with Hemstreet’s trial counsel, and refused to testify on Hemstreet’s behalf, after detectives visited her mother and sister. The detectives apparently informed these family members that “they were in for a lot of trouble” if Bucci testified. Although Bucci character*140ized this conduct as a threat, the state trial court did not expend any significant effort on investigating whether Bucci had been intimidated. Moreover, neither the prosecution nor Hemstreet’s trial counsel pressed for an investigation into the detectives’ conduct, and trial counsel made no effort to seek a remedy from the court for any threats to which Bucci and her family had allegedly been subjected.

Despite the absence of any extensive information about the detectives’ conduct, we are not persuaded that the district court’s findings of fact with respect to the reasons for Bucci’s unavailability were clearly erroneous. There is no evidence in the record that suggests that she became unavailable for any reason other than the alleged threats made by the detectives. Indeed, the prosecution did not directly rebut the accusations of intimidation leveled by Hemstreet’s trial counsel, but instead made statements suggesting that little was known about the actual conversation between the detectives and Bucci’s family. The prosecutor indicated that he “did not believe” the accusations and that he would “never condone” that type of conduct by the detectives, but offered no specific information whatsoever to demonstrate that the detectives did not in fact intimidate Bucci by threatening her family. Under these circumstances, we conclude that the district court’s findings as to Bucci’s unavailability were not clearly erroneous.

In addition, we conclude that the district court did not make a clearly erroneous factual determination when it found that Hemstreet’s trial counsel failed to call Buc-ci as a witness for reasons that did not implicate trial strategy. The record does not support an inference that Bucci was willing to testify on Hemstreet’s behalf but was not called to the stand for strategic reasons.- To the contrary, the meager information available in the record suggests that Hemstreet’s trial counsel was unable to secure Bucci’s participation in the trial after the detectives allegedly intimidated her. We note that the Sparman affidavit submitted by trial counsel does not provide an alternative explanation for the absence of Bucci’s testimony. We therefore find no basis to overturn the district court’s factual findings.

Based on these findings of fact, we agree with the district court that the professional performance of Hemstreet’s trial counsel was sufficiently deficient so as to fall below an objective standard of reasonableness. Although trial counsel was well aware that Bucci could offer significant exculpatory information that could lead the jury to acquit Hemstreet, and that she had been intimidated into not testifying, he made no effort to pursue an investigation into the matter or to seek a remedy on his client’s behalf. Additionally, though he could have sought to compel her attendance at trial, he altogether failed to secure her testimony through such means. Given that “prosecutorial intimidation that dissuades a potential defense witness from testifying for the defense can ... violate the defendant’s right to present a defense,” United States v. Williams, 205 F.3d 23, 29 (2d Cir.2000), trial counsel’s conduct falls squarely within the range of deficient representation.

We also agree with the district court that trial counsel’s deficient conduct affected the outcome of Hemstreet’s trial. As the state acknowledges, the prosecution’s case against Hemstreet was, by and large, a circumstantial one. Much of the evidence introduced by the prosecution did not directly tie Hemstreet to Hiep’s murder, and, to the extent that some evidence may be characterized as supposed confessions by Hemstreet, the defense cast serious doubt on the testimony of the wit*141nesses who reported those confessions. In short, the evidence offered by the prosecution at trial was inconclusive rather than overwhelming.

Ultimately, the prosecution’s strongest evidence against Hemstreet was (a) that Hiep was last seen in the company of both Bentz and Hemstreet prior to his murder and (b) that no one had ever seen Bentz alone with Hiep. This evidence allowed the prosecution to effectively argue that both Hemstreet and Bentz, not Bentz alone, had murdered Hiep. However, as mentioned above, Bucci swore in an affidavit that she had seen Hiep and Bentz together at Lace, without Hemstreet, after the time when Hemstreet said he had been dropped off at home by Bentz.

“Evidence that the victim was last seen with someone other than the defendant” is “likely to raise a reasonable doubt in a juror’s mind” about the defendant’s guilt. O’Guinn v. Dutton, 88 F.3d 1409, 1419-20 (6th Cir.1996) (Merrit, C.J., concurring). Here, if Hemstreet’s trial counsel had tried to secure Bucci as a witness, she could have offered testimony that would have directly contradicted the prosecution’s best evidence against Hemstreet. Because the prosecution built a case on the basis of inconclusive evidence, Bueci’s testimony “would likely have created a reasonable doubt in the minds of the jurors had they been made aware of’ it. Id. at 1419. Accordingly, trial counsel’s failure to pursue this critical testimony or to seek a remedy for its loss through intimidation raised a reasonable probability that the result of Hemstreet’s trial would have been different but for his unprofessional errors. See Williams, 205 F.3d at 29 (noting the effect of prosecutorial intimidation where it dissuades a defense witness from testifying and citing cases).

In sum, we conclude that the record available to appellate counsel raised an obvious question regarding the effectiveness of Hemstreet’s trial counsel.

Our inquiry now proceeds to the question upon which this habeas petition turns: Did Hemstreet’s appellate counsel provide ineffective assistance to him by failing to raise on direct appeal the issue of trial counsel’s deficient performance? In order for us to conclude that he did, we must determine both that appellate counsel’s conduct was objectively unreasonable and that the outcome of Hemstreet’s appeal would likely have been different had the issue been raised of whether Hemstreet received a fair trial. See Pavel v. Hollins, 261 F.3d 210, 216 (2d Cir.2001); Lindstadt v. Keane, 239 F.3d 191, 198-99 (2d Cir.2001).

We have held that a habeas petitioner can show ineffective assistance of appellate counsel if he demonstrates that, on appeal, “counsel omitted significant and obvious issues while pursuing issues that were clearly and significantly weaker.” Jackson v. Leonardo, 162 F.3d 81, 85 (2d Cir.1998) (internal quotation omitted). The district court here found that appellate counsel’s conduct was unreasonable because the issue of trial counsel’s ineffectiveness was quite clear from the record, but appellate counsel failed to argue it and instead pursued a claim that the verdict was against the weight of the evidence.

Although we “indulge a strong presumption that counsel’s conduct falls within the wide range of reasonable professional assistance,” Strickland, 466 U.S. at 689, 104 S.Ct. 2052, we cannot characterize the performance of Hemstreet’s appellate counsel as “reasonable.” Here, trial counsel had not pursued any remedy for the intimidation of a crucial witness for the defense which arguably violated Hemstreet’s right to a fair trial. This deficiency stood out clearly in the trial record. The failure to *142recognize it as the most significant argument on appeal was an error that compromised Hemstreet’s right to meaningful representation. To characterize appellate counsel’s failure to raise the issue as a strategic choice, as we are urged to do by the appellant here, is to ignore the more likely explanation that appellate counsel made a severe mistake in judgment.

Moreover, despite being in the best position to defend his performance on appeal, appellate counsel declined to respond to Hemstreet’s Sparman request and did not provide any evidence to contradict the conclusion that he simply failed to recognize the ineffective assistance claim. Without an alternative explanation, and faced with the evidence in the record before us, we agree with the district court that “no tactical decision can justify failure to raise the issue on direct appeal.”

In order to meet the second prong of the Strickland test, Hemstreet must demonstrate that appellate counsel’s unreasonable conduct prejudiced the outcome of his appeal, such that our confidence in the outcome is undermined. See Strickland, 466 U.S. at 694, 104 S.Ct. 2052. To establish prejudice in the appellate context, a petitioner must demonstrate that there was a reasonable probability that his claim would have been successful before the [state’s highest court].” Mayo v. Henderson, 13 F.3d 528, 534 (2d Cir.1994) (internal citations and quotation marks omitted). We find that the standard has been met here. In the context of a murder trial in which the evidence against Hemstreet was inconclusive, a defense witness with highly material and exculpatory testimony was intimidated, and trial counsel failed to act in the face of such intimidation, Hemstreet’s right to a fair trial was compromised. Given that his trial attorney provided Hemstreet with ineffective assistance of counsel that affected the outcome of his trial, there is a “reasonable probability” that the New York Court of Appeals would have responded to the circumstances of Hemstreet’s representation in the same manner that we do: It would have been concerned that an error of constitutional dimensions had been made by Hemstreet’s trial counsel. Raising this issue on direct appeal would have in all likelihood caused the New York Court of Appeals to do something other than affirm Hemstreet’s conviction in a one-sentence order. We therefore conclude that appellate counsel’s failure to raise the issue undermines confidence in the outcome of Hemstreet’s appeal.

Although we have found that Hemstreet meets both prongs of the Strickland test, his habeas petition can only be granted if the state court’s decision that he did not receive ineffective assistance of counsel represents an “unreasonable,” not simply incorrect or erroneous, application of federal law. See Williams, 529 U.S. at 412, 120 S.Ct. 1495. In evaluating whether this criterion is met, we must determine that the decision “reflect[s] some additional increment of incorrectness such that it may be said to be unreasonable.” Aparicio, 269 F.3d at 94. In light of the magnitude of appellate counsel’s error, the particularly inadequate performance of trial counsel, and the inconclusive nature of the circumstantial case made against Hemstreet, we hold that the Appellate Division’s rejection of Hemstreet’s ineffective assistance of counsel claim was not only incorrect, but objectively unreasonable as well. See Lindstadt, 239 F.3d at 205 (holding that state court decision to deny ineffective assistance claim was unreasonable when Strickland prongs had been met).

In sum, we hold that Hemstreet has shown that the performance of his appellate counsel was objectively unreasonable and prejudiced his appeal in satisfaction of *143the Strickland test. The state appellate court’s denial of Hemstreet’s coram nobis petition “involved an unreasonable application of clearly established Federal law, as determined by the Supreme Court of the United States,” 28 U.S.C. § 2254(d)(1), and Hemstreet’s habeas petition accordingly must be granted.

CONCLUSION

In light of the foregoing, we affirm the decision of the district court.

. Hemstreet also argued that his appellate counsel was ineffective because he failed to raise two other issues on appeal: (1) that trial counsel had failed to make a Brady claim with respect to notes taken by the detective who interviewed Bucci; and (2) that trial counsel failed to raise a double jeopardy issue based on pervasive prosecutorial misconduct. Hemstreet repeated these two arguments in his habeas petition before the district court below. The district court considered the arguments but found them unpersuasive, and its decision on these claims has not been appealed to this court.






Dissenting Opinion

B.D. PARKER, JR.,

dissenting.

According to the majority, Hemstreet is entitled to habeas relief for essentially three reasons: (1) his trial counsel was ineffective, (2) his appellate counsel was ineffective for failing to raise his trial counsel’s ineffectiveness, and (3) the state court’s determination that he was not deprived of effective assistance of appellate counsel was, under AEDPA, “objectively unreasonable” in light of clearly established federal law, 28 U.S.C. § 2254(d)(1); see Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). Since I disagree with these conclusions, I respectfully dissent. I cannot find that Hemstreet has shown his appellate counsel’s performance to have been inadequate or that he suffered prejudice from a purportedly deficient performance. I conclude that he, therefore, could not meet the more stringent burden of showing that his appellate counsel’s performance was “objectively unreasonable” under AEDPA.

Since I do not believe the record establishes ineffectiveness of trial counsel, I do not believe appellate counsel can be faulted for not pursuing the claim on direct appeal. The majority concludes that trial counsel’s failure to pursue any remedy for the intimidation of a crucial witness was a violation of Hemstreet’s rights that clearly stood out in the trial record and that Appellate counsel’s failure to recognize and to pursue it as the most significant argument on appeal amounted to constitutionally inadequate representation. Since I believe the record was anything but clear, I do not believe the argument stood out or that appellate counsel was required by Strickland to raise it.

Under New York law, it is extremely difficult successfully to pursue a claim of ineffective assistance of counsel on direct appeal and such claims are typically pursued on collateral attack under § 440.10, N.Y.Crim. Proc. Law. § 440.10.1 People v. Gil, 285 A.D.2d 7, 12, 729 N.Y.S.2d 121 (N.Y.App. Div., 1st Dep’t 2001) (noting that it is a “rare case” that such a claim is appropriately pursued on direct appeal). This rule is driven by a need to examine facts and circumstances about the performance of counsel that lie outside of the trial record, which provides the only set of facts available on direct appeal. An ineffective assistance claim brought on collateral attack affords the court the ability to develop facts specifically concerning trial counsel’s performance and the circumstances of counsel’s decisions. In cases such as this one, where none of these types of information are available and the record concerning trial counsel’s motivations and reasoning is sparse, New York courts have routinely denied claims of ineffective assistance of counsel on direct appeal. See, e.g., People v. Love, 57 N.Y.2d 998, 1000, 457 N.Y.S.2d 238, 443 N.E.2d 486 (1982) (affirming conviction on direct appeal because “we cannot conclude that defen*144dant’s counsel was ineffective simply by reviewing the trial record without the benefit of additional background facts that might have been developed had an appropriate after-judgment motion been made pursuant to CPL 440.10” (internal quotation marks omitted)); People v. Jones, 55 N.Y.2d 771, 773, 447 N.Y.S.2d 242, 431 N.E.2d 967 (1981) (“Reviewing the record without the benefit of knowing, as might have been developed had an appropriate after-judgment motion been made, what was in defense counsel’s mind, we cannot conclude that defendant’s counsel was ineffective.”); People v. Zeh, 289 A.D.2d 692, 734 N.Y.S.2d 306, 308 (N.Y.App. Div., 3d Dep’t 2001) (noting that the “Court is being asked to decide the issue of effective assistance of counsel on direct appeal when many of the claims would require us to resort to supposition and conjecture” and stating that “prudence dictates that the issue of ineffective assistance of counsel be raised in a post-trial application” where a complete record can be made) (internal quotation marks omitted); People v. Figueroa, 254 A.D.2d 226, 679 N.Y.S.2d 304, 304 (N.Y.App. Div., 1st Dep’t 1998) (finding that an ineffective assistance of counsel claim was inappropriate on direct appeal since it was based on facts not in the record and trial counsel had not had a chance to explain her trial tactics).

Here, I do not believe the trial record was sufficiently developed such that appellate counsel was compelled to bring an ineffective assistance of trial counsel claim. The record with respect to what actually occurred between Jeanette Bucci and the investigating officers and what trial counsel did and why is sketchy and inconclusive. The majority concludes that the record shows that, prior to trial, Hemstreet’s counsel knew that Bucci had material and exculpatory information regarding where, when, and with whom the murder victim was last seen. The majority goes on to conclude that Bucci was unavailable as a witness because she had been threatened by law enforcement officers and that trial counsel’s decision not to call her was not a strategic choice on his part.

However, the only material in the record as to what occurred is a statement by Hemstreet’s trial counsel raising the issue to the trial judge, saying of Bucci that “she told me that her family was visited, her mother and sister were visited by two police officers,” and “[one of the officers] had indicated to the mother and sister than if [Bucci] testifies they were in for a lot of trouble.” Tr. of Jury Selection, at 301-02. There is nothing in the record to indicate the context or background of this exchange, or why trial counsel or the court did not pursue the issue. In fact, during the District Court proceeding, trial counsel submitted an affidavit explaining that he had no recollection as to why the witness was not called. A number of explanations as to why this issue was not pursued at trial are possible: perhaps there were concerns about Bucei’s credibility; perhaps trial counsel was negligent; perhaps Bue-ci’s testimony would have been inconsistent with other defense evidence; or perhaps trial counsel decided not to pursue the issue for currently unknown tactical reasons. Each of these possibilities has one thing in common: it is speculative. Consequently, the record does not support any conclusions about why Bucci was not called.

Had the issue of trial counsel’s effectiveness been raised by way of a § 440.10 motion in the state trial court, as it should have been,2 a record concerning what actu*145ally occurred could have been developed. Then, the state courts would, in the first instance, have had an opportunity to determine — on a complete record — whether relief was appropriate. We would then, if necessary, have been in a position to determine whether constitutional violations occurred and what remedies were appropriate. Given such an inconclusive record, I am not prepared to conclude that Hem-street has overcome the “strong presumption that counsel’s conduct falls within the wide range of reasonable professional assistance.” Aparicio v. Artuz, 269 F.3d 78, 95 (2d Cir.2001) (quoting Strickland, 466 U.S. at 689, 104 S.Ct. 2052) (internal quotation marks omitted). We should be especially wary about eroding that presumption on a record such as the one before us where no other aspect of trial counsel’s performance in a long, emotionally charged murder trial has been challenged.

Thus, appellate counsel’s failure to pursue trial counsel’s supposedly ineffective assistance on direct appeal is not evidence of constitutionally deficient performance. To the contrary, it reflects a decision not only consistent with, but arguably compelled by, New York law, since the lesson to be drawn from the cases is that, given the state of the record, pursuit of an ineffective assistance claim on direct appeal would almost surely have been futile.

Furthermore, even if there were a color-able argument that Hemstreet’s appellate counsel was deficient for failing to raise trial counsel’s purported ineffectiveness, Hemstreet has failed to show prejudice. First, as shown above, Hemstreet should have pursued his ineffective assistance claim on a § 440.10 motion to the trial court as required by New York law. Caballero, 42 F.3d at 740; People v. Ramos, 63 N.Y.2d 640, 643, 479 N.Y.S.2d 510, 468 N.E.2d 692 (1984). As New York law clearly establishes that it would have been premature and probably futile to raise such a claim on direct appeal, petitioner could not have suffered prejudice from appellate counsel’s failure to do so. The fact that Hemstreet can still pursue a § 440.10 remedy is a pellucid reason why he has suffered no prejudice.

Additionally, the case against Hemstreet was independently strong. In April 1992, and again in September 1993, Hemstreet confessed to killing Kenneth Hiep. He told Nancy Wright, an old friend, that Hiep’s death was ordered by two individuals — Sal and Carmine Franco — and that “it had to be done; it was a contract.” Later, in September 1993, he told Robert Horne-man, whose father was an old business partner of Hemstreet’s, that “[Hiep] didn’t do the right thing ... I made my bones with [Hiep].” Both witnesses testified against Hemstreet at trial. In addition to these confessions, the state marshaled considerable circumstantial evidence establishing that Hemstreet and Hiep, once close friends, became estranged as a result of business dealings. Hemstreet had apparently complained to his friends that Hiep had stolen a considerable sum of money from a business they jointly owned. On the night of Hiep’s death, Hemstreet and Hiep were seen having a heated argument at a local bar. When Hiep’s son had to leave the bar early, Hemstreet offered to give Hiep a ride home. The state also presented evidence that Hemstreet’s *146hands displayed fresh cuts and injuries the day after Hiep’s death that were not present the night before. When asked about the cause of these cuts on two separate occasions, Hemstreet gave inconsistent explanations. Further, the state showed that Hemstreet’s Cadillac had substantial blood stains on the front passenger seat— stains that were then cleaned by Hem-street’s associate the day after Hiep’s disappearance. Hemstreet also inexplicably tried to dissuade Hiep’s family from calling the police to report Hiep’s disappearance, even though he himself had suggested that Hiep had been kidnapped. Finally, a search of Hemstreet’s Cadillac confirmed the presence of Hiep’s blood throughout the passenger seat area. Slash cuts in the passenger seat headrests and blood stains on the lining of the roof indicated that Hiep had been fatally assaulted inside petitioner’s car.

In light of the plethora of evidence against Hemstreet, I am not persuaded that, but for trial counsel’s failure to call Bucci and pursue the intimidation claim, a reasonable probability existed that Hemstreet would have been acquitted. I suggest that petitioner’s appellate counsel could easily have reached the same conclusion. As previously observed, this conclusion could have been reached in the context of New York law clearly dictating that ineffectiveness claims are properly raised on direct appeal only in “rare cases” and that a § 440.10 motion to vacate judgment on the grounds of trial counsel’s ineffectiveness remained open.3 In light of these factors, I do not believe Hemstreet has established a reasonable probability that, but for his appellate counsel’s performance, his appeal to the Appellate Division would have been successful. See Smith v. Robbins, 528 U.S. 259, 285, 120 S.Ct. 746, 145 L.Ed.2d 756 (2000). Therefore, he has not established prejudice.

The majority states that the District Court concluded that the Appellate Division’s rejection of Hemstreet’s ineffective assistance of counsel claim was an unreasonable application of clearly established federal law under 28 U.S.C. § 2254(d)(1). I disagree. In fact, the District Court failed even to mention, much less apply, the AEDPA standards anywhere in its opinion. In my view, AEDPA requires the denial of Hemstreet’s habeas petition. Hemstreet sought a writ of coram nobis in the Appellate Division, alleging that his appellate counsel was ineffective because he did not raise trial counsel’s supposed ineffectiveness in failing to demand a hearing, sanctions, or a mistrial as a result of prosecutorial interference with a potential defense witness. The Appellate Division denied the petition, concluding that “[t]he appellant has failed to establish that he was denied the effective assistance of appellate counsel.” Hemstreet, 290 A.D.2d at 458, 738 N.Y.S.2d 583. This decision constitutes an adjudication on the merits *147entitled to fall AEDPA deference. Sellan v. Kuhlman, 261 F.3d 303, 311-12 (2d Cir.2001). Under AEDPA, Hemstreet is not entitled to habeas relief unless he establishes that the Appellate Division’s conclusion amounted to an “objectively unreasonable” application of clearly established federal law, Williams v. Taylor, 529 U.S. 362, 409, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000), a test that falls somewhere in between “merely erroneous” and “unreasonable to all reasonable jurists,” Francis S. v. Stone, 221 F.3d 100, 109 (2d Cir.2000). Thus, it is not enough to find, as the majority does, that appellate counsel, in their opinion, was ineffective. AEDPA requires a showing that the Appellate Division’s conclusion that petitioner had not received ineffective assistance of counsel “objectively unreasonable.” Because I conclude that Hemstreet failed to make even the minimal showing that appellate counsel’s performance was inadequate under Strickland, I also find that he failed to meet the more stringent burden of showing the state court’s decision to have been “objectively unreasonable.”

For these reasons, I respectfully dissent.

. The statute states, in relevant part: "At any time after the entiy of a judgment, the court in which it was entered may, upon motion of the defendant, vacate such judgment upon the ground that ... the judgment was obtained in violation of a right of the defendant under the constitution of this state or of the United States.” N.Y.Crim. Proc. Law § 440.10(l)(h).

. A motion in the trial court to vacate the judgment would not be barred on the basis of § 440.10(2)(b), which compels a court to deny such a motion based upon the defendant’s *145"unjustifiable failure to raise such ground or issue upon an appeal." N.Y.Crim. Proc. Law § 440.10(2)(c). This is because, as discussed below, I do not believe that "sufficient facts appear on the record of the proceedings underlying the judgment to have permitted, upon appeal ..., adequate review of the ground or issue raised upon the motion.” Id. Here, review of an ineffective assistance of trial counsel claim clearly required further factual development and, therefore, remains open.

. Additionally, Hemstreet's appellate counsel cannot be found ineffective for failing to file a § 440.10 motion in state trial court. A right to effective assistance of counsel only exists where a right to counsel exists in the first place. See Wainwright v. Torna, 455 U.S. 586, 587-88, 102 S.Ct. 1300, 71 L.Ed.2d 475 (1982). Further, the right to counsel only extends to the first appeal as of right. Evitts v. Lucey, 469 U.S. 387, 394, 105 S.Ct. 830, 83 L.Ed.2d 821 (1985). Thus, Hemstreet had no right to counsel on collateral attack pursuant to § 440.10 and, therefore, could not have been deprived of effective assistance of counsel when his appellate counsel failed to pursue such a motion. See Chalk v. Kuhlmann, 311 F.3d 525, 529 (2d Cir.2002) (holding that a New York defendant was not deprived of effective assistance of counsel when his lawyer failed to file for discretionary appeal to the Court of Appeals because the defendant's right to counsel did not extend to a discretionary appeal).

midpage