Opinion
No. 04 Civ. 5455 (DLC) (GWG).
January 17, 2006
REPORT AND RECOMMENDATION
James Truesdale brings this petition for writ of habeas corpus pursuant to 28 U.S.C. § 2254. Following a jury trial in New York State Supreme Court, Bronx County, Truesdale was convicted of various criminal charges and was sentenced to a term of imprisonment. He is currently released on parole. For the reasons stated below, the petition should be denied.
I. BACKGROUND
A. The Crime
Because the only issue raised by Truesdale pertains to jury selection, the evidence presented at trial is not relevant to this petition. In brief, the testimony at trial showed that Truesdale was one of three youths who pushed their way into a lighted security booth occupied by two residential security guards in October 1995. The three youths assaulted one of the guards, stole the other's money and wallet, and threatened both of their lives. Both guards recognized the three youths as previously having been in the lobby and courtyard of the building. In addition, the guards saw the three youths drinking and smoking for two to three hours in the lobby on the night of the attack. Using photographs, each guard identified Truesdale as one of the perpetrators. See Memorandum of Law, undated ("Resp. Mem.") (annexed to Affidavit in Opposition, filed Feb. 23, 2005 (Docket #8) ("Aff. in Opp.")), at 2. One of the guards also identified Truesdale at a lineup. See Petitioner's Amended Memorandum of Law and Appendix in Support of Petition for a Writ of Habeas Corpus, filed Aug. 19, 2005 (Docket #17) ("Pet. Am. Mem."), at 3-4.
B. Voir Dire
Jury selection for the trial of Truesdale and his two co-defendants began on May 28, 1998. The defense and the prosecution could each exercise up to 15 peremptory challenges for the selection of 12 jurors, and an additional 2 peremptory challenges for the selection of two alternates. See N.Y. Crim. Proc. Law § 270.25(2)(b). Under the trial court's jury selection process, a subset of jurors from the venire was seated in the jury box, and challenges for cause and peremptory challenges were exercised as to those jurors. During the first round, 14 members of the venire were seated in the jury box. (V. 128). Of these, 3 were excused by the court for cause with the parties' consent, leaving 11 prospective jurors. (V. 22, 134, 182, 191). After the prospective jurors in the box were questioned, the prosecutor exercised 5 peremptory challenges. (V. 191). The defense exercised 2 peremptory challenges, leaving 4 jurors from the first round of jury selection to be seated on the petit jury. (V. 191, 194). 3 of those 4 jurors were African-American. (See V2. 33).
"V." refers to a document that includes transcripts from proceedings held on various dates (Docket #20). "V2" refers to a portion of the transcript from the proceedings on June 2, 1998, which is bound separately (Docket #21).
There were 20 members of the venire left for the second round of jury selection, and 14 of them filled the jury box while the remaining 6 were seated in the front row of the courtroom. (V. 192, 196). After challenges for cause (V2. 28-31), the trial judge asked the parties for their challenges only with respect to 8 identified prospective jurors. (V2. 31-32). The prosecutor exercised 3 peremptory challenges among these 8 prospective jurors (V. 32), whereupon counsel for one of Truesdale's co-defendants stated:
Your Honor, before we have a peremptory challenge we have a Batson [v. Kentucky, 476 U.S. 79 (1986)] [sic]. Every single one of jurors [sic] challenged by the People so far have been black. The People have specifically skipped over in their challenge in this specific rounds [sic] . . . white people. Everyone we've picked up has been black. The first round and in this round, Judge. And Mr. Freeman in this round is the only black that he didn't challenge and he is a security guard, which, of course, Your Honor knows that their chief witness is a security guard [sic] I think they're banking on the fact there's some affinity there.
(V2. 32-33). Truesdale's counsel joined in the motion. (V2. 33). The trial judge asked the prosecutor if a prima facie case existed and the prosecutor asserted that it did not, noting that the prosecution had so far not opposed the seating of 3 African-Americans on the petit jury. (V2. 33-34). After asserting that the Batson motion was groundless, the prosecutor provided non-discriminatory reasons for some but not all of his peremptory challenges, and then asserted that his challenges had "not been based upon racial lines at all." (V2. 33-35).
Prior to the first exercise of challenges, the court stated, "To the extent possible, I'd ask one defense attorney to speak for all three, but I'm not going to foreclose anybody from supplementing." (V. 186). Counsel for one of Truesdale's co-defendants offered to speak on behalf of the three defendants. (V. 186-87).
In response, defense counsel noted that "all three defendants are male African-Americans," that "the challenges that were exercised [were] 100 percent African-American challenges," and that it was irrelevant that the prosecutor did not challenge other African-American prospective jurors; rather, it was relevant only that 100% of the jurors that the prosecutor did challenge were African-American. (V2. 35). Defense counsel argued that there was no "discernable . . . reason" for the prosecutor's challenges other than "racial composition." (V2. 35-36). The prosecutor then asserted that all the prospective jurors in the first round available to him were "non-white," whereupon counsel for a co-defendant stated that this was false because one of those jurors was "not an African-American." (V. 36). The court agreed that the juror was "not an African-American." (See V. 36). Then, the court ruled that petitioner and his co-defendants had not established a prima facie case of intentional discrimination:
[U]pon consideration of defendants' Batson challenge, I would note that in [the] first panel, there were ten African-American prospective jurors out of a total of 11 available for challenges — available for peremptory challenges, I should say. Of that group, the People exercised three out of seven. In the second panel, of the first eight, there are four African-American prospective jurors totaling 14 African-American jurors out of a total of 19. Of these, People have challenged a total of six. When measured with all the information provided me, I conclude first that neither the percentage of peremptories in themselves or any other facts relied upon by the defendants satisfy me that a prima [facie] case of discrimination on the People's part has been made out. I would note somewhat parenthetically as to [one juror], the record at a minimum, verged on a challenge for cause as to her and I see no pattern of discrimination and defendants['] Batson objection is overruled.
(V2. 37-38). It is undisputed that the trial court incorrectly stated that the prosecution had exercised 3 peremptory challenges in the first round. See Resp. Mem. at 9. Instead, the transcript shows that the prosecution exercised 5 challenges in that round (V. 191), making a total of 8 challenges at the time of the Batson claim (not 6, as the trial judge stated).
After the Batson objection, the racial characteristics of the remaining venire members and subsequent challenges were not set forth in the record. The Batson issue was not raised again at trial.
C. Verdict and Sentence
At the conclusion of the trial, the jury convicted Truesdale of Robbery in the Second Degree (New York Penal Law ("NYPL") § 160.10(1)), two counts of Attempted Robbery in the Second Degree (NYPL §§ 110/160.10(1)), Assault in the Second Degree (NYPL § 120.05(2)), and Criminal Possession of Stolen Property in the Fifth Degree (NYPL § 165.40). Aff. in Opp. ¶ 5; Pet. Am. Mem. at 1. On August 5, 1998, he was sentenced to concurrent terms of 10 years on the robbery count, 7 years each on the attempted robbery and assault counts, and one year on the possession of stolen property count. Aff. in Opp. ¶ 5; Pet. Am. Mem. at 1-2.
D. Direct Appeal
Truesdale appealed his conviction to the Appellate Division, First Department. In his brief, Truesdale raised two issues: (1) the investigation procedures were unduly suggestive; and (2) he "was denied his federal and state rights to due process and equal protection where, even though the prosecutor had exercised all eight of his challenges against African-Americans, the court denied [his] Batson claim without requiring the prosecutor to supply race-neutral reasons for all of the challenges." Brief for Defendant-Appellant, dated June 2002 ("Pet. App. Br.") (reproduced as Ex. 1 to Aff. in Opp.), at 19, 27.
On October 1, 2002, the Appellate Division affirmed Truesdale's conviction and the lower court's Batson ruling. People v. Truesdale, 299 A.D.2d 289, 290 (1st Dep't 2002). With respect to the Batson application, the Appellate Division's ruling was as follows:
Defendant's application pursuant to Batson v. Kentucky 476 U.S. 79 (1986), was properly denied since he failed to establish a prima facie case of discrimination. See People v. Brown, 97 N.Y.2d 500, 507-508 (2002). Given the racial composition of the panel, defendant's numerical argument was not so compelling as to be conclusive by itself, and it was not corroborated by any other evidence.See People v. Miller, 298 A.D.2d 194 (N.Y.App.Div. 1st Dep't 2002). Truesdale, 299 A.D.2d at 290 (citation information added).
Truesdale sought leave to appeal to the New York Court of Appeals. See Letter from Robert Budner to Your Honor, dated Jan. 8, 2003 (reproduced as Ex. 4 to Aff. in Opp.), at 1-2. In his application letter, Truesdale made the same arguments as he did on appeal to the Appellate Division. Id. On April 23, 2003, leave was denied. People v. Truesdale, 99 N.Y.2d 659 (2003).
E. The Instant Habeas Corpus Petition
Truesdale timely filed this petition for a writ of habeas corpus on July 14, 2004. See Petition Under 28 U.S.C. § 2254 for Writ of Habeas Corpus by a Person in State Custody, filed July 14, 2004 (Docket #3) ("Petition"). Truesdale's petition alleges one ground for habeas relief: that he "was denied [his] right to equal protection under the law when, during the jury selection, the Prosecutor exercised a statistically significant pattern of peremptory challenges to exclude African-Americans." Petition at 5. Following briefing, see Petitioner's Memorandum of Law and Appendix in Support of Petition for a Writ of Habeas Corpus, filed July 14, 2004 (Docket #4) ("Pet. Mem."); Aff. in Opp.; Resp. Mem.; Petitioner's Reply Memorandum of Law in Support of Petition for a Writ of Habeas Corpus, filed April 19, 2005 (Docket #10) ("Pet. Reply"), Truesdale was granted permission to file an amended memorandum of law in light of a jury selection transcript he had recently received and intervening federal court decisions. See Memorandum Endorsement, filed July 25, 2005 (Docket #16). On August 19, 2005, Truesdale filed the amended memorandum of law. See Pet. Am. Mem. Respondent filed an affidavit and memorandum in response. See Supplemental Affidavit, filed Sept. 8, 2005 (Docket #18) ("Supp. Aff. in Opp."); Supplemental Memorandum of Law (annexed to Supp. Aff. in Opp.). Truesdale then filed another reply memorandum of law. See Petitioner's Supplemental Reply Memorandum of Law, filed Sept. 23, 2005 (Docket #19) ("Pet. Supp. Reply Mem.").
II. APPLICABLE LEGAL PRINCIPLES
A. Exhaustion
Before a federal court may consider the merits of a habeas claim, a petitioner is first required to exhaust his available state court remedies. See 28 U.S.C. § 2254(b)(1) ("An application for a writ of habeas corpus on behalf of a person in custody pursuant to the judgment of a State court shall not be granted unless it appears that . . . the applicant has exhausted the remedies available in the courts of the State. . . .");accord Daye v. Attorney Gen. of New York, 696 F.2d 186, 190-91 (2d Cir. 1982) (en banc), cert. denied, 464 U.S. 1048 (1984). To exhaust a habeas claim, a petitioner is required to have presented the federal nature of each claim to all levels of the state appellate courts. See, e.g., Duncan v. Henry, 513 U.S. 364, 365-66 (1995) (per curiam); Picard v. Connor, 404 U.S. 270, 275-76 (1971); Grey v. Hoke, 933 F.2d 117, 119 (2d Cir. 1991); Daye, 696 F.2d at 191.
B. Standard of Review of Petitions for Habeas Corpus Under 28 U.S.C. § 2254
Pursuant to the Antiterrorism and Effective Death Penalty Act of 1996, a petition for writ of habeas corpus may not be granted with respect to any claim that has been "adjudicated on the merits" in the state courts unless the state court's adjudication: "(1) 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; or (2) resulted in a decision that was based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding." 28 U.S.C. § 2254(d). For a claim to be adjudicated "on the merits" within the meaning of 28 U.S.C. § 2254(d), the state court must base its decision on "the substance of the claim advanced, rather than on a procedural, or other, ground." Sellan v. Kuhlman, 261 F.3d 303, 311 (2d Cir. 2001). It is not necessary for the state court to refer to the federal aspect of a claim or to any federal law for the deferential standard to apply. Id. at 312.
In Williams v. Taylor, the Supreme Court held that a state court decision is "contrary to" clearly established federal law only "if the state court applies a rule that contradicts the governing law set forth" in Supreme Court precedent or "if the state court confronts a set of facts that are materially indistinguishable from a decision [of the Supreme Court] and nevertheless arrives" at a different result. 529 U.S. 362, 405-06 (2000). Williams also held that habeas relief is available under the "unreasonable application" clause only "if the state court identifies the correct governing legal principle from [the Supreme Court's] decisions but unreasonably applies that principle to the facts of the prisoner's case." Id. at 413. A federal court may not grant relief "simply because that court concludes in its independent judgment that the relevant state-court decision applied clearly established federal law erroneously or incorrectly." Id. at 411. Rather, the state court's application must have been "objectively unreasonable."Id. at 409.
C. Law under Batson v. Kentucky Batson established a three-step process for determining whether peremptory challenges have been exercised discriminatorily:
First, the defendant must make a prima facie showing that the prosecutor has exercised peremptory challenges on the basis of race. Second, if the requisite showing has been made, the burden shifts to the prosecutor to articulate a race-neutral explanation for striking the jurors in question. Finally, the trial court must determine whether the defendant has carried his burden of proving purposeful discrimination.Hernandez v. New York, 500 U.S. 352, 358-59 (1991) (citingBatson, 476 U.S. at 96-98). Here, the second and third stages of the Batson process were not reached. (V2. 38). The only issue before this Court is whether the trial court's conclusion that Truesdale had not made a prima facie showing underBatson was an unreasonable application of clearly established federal law.
A defendant "may make out a prima facie case of purposeful discrimination by showing that the totality of the relevant facts gives rise to an inference of discriminatory purpose." Batson, 476 U.S. at 93-94. A trial court must consider "all relevant circumstances" in deciding whether there is a prima facie case of discrimination. Id. at 96. The Court in Batson gave two examples of such circumstances: (1) a "pattern of strikes" against persons of a particular race or gender "might give rise to an inference of discrimination," and (2) statements or questions during voir dire "may support or refute an inference of discriminatory purpose." Id. at 97 (internal quotation marks omitted); see Tankleff v. Senkowski, 135 F.3d 235, 249 (2d Cir. 1998) ("In considering whether a defendant has made out aprima facie case . . ., we believe that courts should consider how many members of the cognizable racial group are in the venire panel from which the petit jury is chosen, the pattern of strikes against racial group jurors in the particular venire, the prosecutor's statements and questions during selection, as well as any other relevant circumstances."); accord Overton v. Newton, 295 F.3d 270, 277-278 (2d Cir. 2002); see, e.g., Anderson v. Superintendent, Elmira Correctional Facility, 360 F. Supp. 2d 477, 493-94 (E.D.N.Y. 2005) (considering the pattern of strikes and the prosecutor's statements in determining whether petitioner made out a prima facie case).
With respect to whether a pattern of strikes constitutes aprima facie case, the Second Circuit has held that "a rate of minority challenges significantly higher than the minority percentage of the venire would support a statistical inference of discrimination." United States. v. Alvarado, 923 F.2d 253, 255 (2d Cir. 1991) ("Alvarado II"). Alvarado II was a case before the Second Circuit on direct appeal rather than an application of the deferential standard of review under 28 U.S.C. § 2254(d). Nonetheless, the Second Circuit has stated in a habeas corpus case, albeit in dictum, that it had "no doubt that statistics, alone and without more, can, in appropriate circumstances, be sufficient to establish the requisite prima facie showing under Batson." Overton, 295 F.3d at 278 (citing cases). In addition, in Green v. Travis, 414 F.3d 288, 299 (2d Cir. 2005), which was a case on habeas review, the court found that there had been a prima facie case of discrimination based only on a statistical disparity in the prosecutor's challenges and stated, in support of its finding, that "[t]he Batson court itself noted that a pattern of strikes can constitute a prima facie case."
The burden of establishing a prima facie case is "minimal" and "similar to that placed on plaintiffs in Title VII and equal protection jurisprudence." Overton, 295 F.3d at 279 n. 10. In fact, the Supreme Court has recently confirmed that, at the first step of a Batson challenge, the possibility of discrimination does not have to be "more likely than not"; rather, a prima facie case requires only "evidence sufficient to permit the trial judge to draw an inference that discrimination has occurred." Johnson v. California, 125 S. Ct. 2410, 2417 (2005).
Respondent argues that Green and Johnson are not relevant to this case because they were decided after the Appellate Division's affirmance of Truesdale's conviction and thus do not constitute "clearly established" Supreme Court precedent under 28 U.S.C. § 2254(d)(1). See Resp. Supp. Mem. at 3. These cases, however, merely confirm the holdings in Batson itself: that a "pattern of strikes" can constitute a prima facie case and that only an "inference of discrimination" is necessary. Batson, 476 U.S. at 93-94, 97.
III. DISCUSSION
A. Exhaustion of Truesdale's Batson Claim
Respondent argues that some of Truesdale's Batson arguments before this Court were never presented to the state trial court or the state appellate court, and thus those arguments cannot be considered here because they are unexhausted. Resp. Mem. at 6.
At trial, petitioner argued during his Batson objection that a prima facie case was established by the fact that 100% of the prosecutor's peremptory challenges were against African-Americans, that non-African-Americans were in the venire, and that 2 of the 6 African-Americans not challenged by the prosecutor possessed qualities perceived to be prosecution friendly. (V2. 33-37). In his appeal to the Appellate Division, Truesdale argued that a prima facie case had been established at trial because the 100% African-American challenge rate was disproportionate to the 74% African-American population in the venire. Pet. App. Br. at 29-30. He attempted to buttress hisprima facie assertion by arguing that, in addition to the statistical showing of discrimination, the prosecutor evinced his discriminatory intent by incorrectly stating that the first round of prospective jurors consisted of only "non-white[s]," even though one of the prospective jurors was "white." Pet. App. Br. at 30.
In the instant habeas proceeding, Truesdale presents some additional statistical arguments. For example, he compares the rate at which the prosecutor challenged African-Americans with the rate at which he challenged non-African-Americans. Pet. Am. Mem. at 27-28. He also offers the African-American exclusion rate as evidence of a prima facie case by itself. Id. at 25. Truesdale did not make these arguments in state court. See Pet. App. Br. at 27-31; (V2. 33-37). Consequently, respondent asserts that Truesdale has "failed to exhaust them." Resp. Mem. at 6-7; Resp. Supp. Mem. at 2.
The cases cited by respondent, however, see Resp. Mem. at 6-7, require only that the claim made in the federal habeas proceeding must also have been presented to the state courts. Here, Truesdale's presented his claim as a "Batson" claim. See Pet. App. Br. at 27-30. The Second Circuit has made clear that, while the constitutional claim presented before the federal court must be "substantially the same" as that presented in state court, that does "not mean that there can be no substantial difference in the legal theory advanced." Daye v. Attorney Gen. of State of N.Y., 696 F.2d 186, 192 n. 4 (2d Cir. 1982). Recently, the Supreme Court made a distinction between "evidence that must be presented to the state courts to be considered by federal courts in habeas proceedings and theories about that evidence."Miller-El v. Dretke, 125 S. Ct. 2317, 2326 n. 2 (2005). InMiller-El, the Court noted in the context of a Batson claim that the petitioner's "comparisons of black and nonblack venire panelists, along with [his] arguments about the prosecution's disparate questioning of black and nonblack panelists and its use of jury shuffles" were arguments based on the "the transcript ofvoir dire," which "was before the state courts." Id. Accordingly, Miller-El concluded that, even though petitioner's arguments had not been made in state court, they were properly before the federal courts. Id.
Here, Truesdale has not submitted new evidence. He has merely offered additional support for his Batson claim through statistical arguments based on the same numbers that were considered by the state courts. These new arguments do not alter the substance of his Batson claim. Rather, he has used the facts already in the record — in particular, the number of jurors present and struck — to enhance his analysis of whether there was an improper "pattern of strikes." Because Truesdale's claim remains essentially the same one that was presented to the state courts — that the prosecutor's conduct supported an inference of discrimination under Batson — it is properly presented here.
B. The Merits of the Batson Claim
1. Statistical Evidence: Pattern of Strikes
Taking both rounds together, there were 19 people available in the venire to be peremptorily challenged. Of these 19 people, 14, or 74%, were African-Americans. When it came to striking jurors, the prosecutor used 8 out of his 8 strikes to strike African-American jurors, or 100% of his strikes.
The question raised by a prima facie analysis underBatson is whether the peremptory strikes used by an attorney give rise to "an inference of discriminatory purpose." Batson, 476 U.S. at 93-94. A "pattern of strikes" indicating discrimination is one way to establish an inference of discrimination. Id. at 97. To decide whether such an inference is raised, the attorney's actual strikes must be compared with the number of strikes that would have been expected had there been no discrimination at all. With all other variables being equal, we would expect that the rate of strikes actually used against members of a protected group would mirror their proportion in the available venire. But if the rate of strikes against the protected group is "significantly" higher than the expected rate, an inference of discrimination may be drawn.Alvarado II, 923 F.2d at 255.
Because African-Americans made up 74% of the venire in Truesdale's case, we would expect that 74% of the prosecutor's 8 strikes, or 6 strikes (rounding up from 5.89), would have been used to eliminate African Americans had the strikes been made entirely by chance. Instead, 100%, or 8 strikes, were used against African-Americans.
One could imagine a rule under which any deviation at all from the expected rate of strikes would be sufficient to raise an inference of discrimination. But the laws of statistics teach us that even events we know to be random do not always occur in particular applications at an expected rate. For example, if we have a black bag containing 100 red candies and 100 blue candies, and if we have someone reach into that bag without looking to remove 10 candies, we understand that in each instance the person will not remove 5 red candies and 5 blue candies. For each group of 10, it will frequently come out 6 reds and 4 blues (or vice-versa), and with somewhat less frequency 7 blues and 3 reds, and so forth. But merely because the number of candies in a given set of 10 is not 5 reds and 5 blues, we would not believe that this fact alone was enough to demonstrate that the person removing the candies was not doing so on a blind basis.
Recognizing this principle, courts reviewing Batson claims have typically compared the degree to which the rate of peremptory challenges against a particular group differs from the expected rate. As noted, this means we must decide if the rate of challenges to members of the group was "significantly higher" than the percentage of that group in the venire. Alvarado II, 923 F.2d at 255. Here, the difference between these rates is approximately 36%. That is, African-Americans were struck at a rate of 100% (8/8), which was 36% higher than the rate we would have expected based on their presence in the venire, that is, 74% (14/19). Thus, if we look only at the percentages involved, the question before us is whether a challenge rate that is 36% higher than the expected rate is "significantly higher" under Alvarado II. More precisely, the question presented is whether the Appellate Division unreasonably applied Supreme Court precedent in concluding that this difference did not raise an inference of discrimination.
In Alvarado II, the Second Circuit performed an analysis of strikes for Batson purposes in a case that arose on direct review. There, the prosecutor had challenged 50% of the minorities (3 of 6) in the selection of the jury of 12 and 57% (4 of 7) in the selection of the jury of 12 plus alternates. 923 F.2d at 255. The Batson challenge in Alvarado II was made after jury selection was completed. Id. The court did not know the minority percentage of the venire because it was absent from the record. Instead, the court used the minority percentage of the population of the district from which the venire was drawn (29%). Id. at 256. Given this discrepancy (50-57% versus 29%), the Court held that "a challenge rate nearly twice the likely percentage of the venire strongly supports a prima facie case under Batson." Id.
This comparison of raw percentages has become common. In holding that "statistics, alone and without more, can, in appropriate circumstances, be sufficient to establish the requisite prima facie showing under Batson," Overton cited to two cases that performed a similar analysis. 295 F.3d at 278-79 n. 9. In Turner v. Marshall, 63 F.3d 807, 813 (9th Cir. 1995), a prima facie case was established when the government used 56% of its peremptory challenges against African-Americans where the venire as a whole was 30% African-American. Id. at 813. As in Alvarado, the rate of challenges was thus almost twice the proportion in the venire. InJones v. Ryan, 987 F.2d 960 (3d Cir. 1993), a prima facie case was established where the prosecutor used 75% of his peremptory challenges on minorities and the venire was 20% minority; thus his challenge rate was "nearly four times the minority percentage of the venire." Id. at 971.
Cases in this Circuit have followed Alvarado II's lead by conducting comparisons of these two percentages. Thus, in United States v. Moore, 4 F. Supp. 2d 319 (S.D.N.Y. 1998), the Court found that a prima facie case had been made where "of the 19 jurors available for challenge by defense counsel through the first four rounds, only 11 (or 58%) were white males, yet defendant exercised all eight of his challenges [i.e., 100%] toward this group." Id. at 321. The rate of strikes in Moore was 72% greater than what would have been expected. The Court concluded that "[t]he chance of this happening randomly was minute." Id.
Following the methodology in Alvarado II and the other cases, and applying the deferential standard of review mandated by 28 U.S.C. § 2254(d), we would be hard-pressed to conclude that New York State had "unreasonably applied" Supreme Court precedent in Truesdale's case. The statistical disparity (100% vs. 74%) is far less than twice the likely percentage cited in Alvarado II — or the disparities cited in Turner and Jones. Instead, the percentage of strikes actually used by the prosecutor (100%) was merely 36% greater than the likely or expected percentage (74%) based on the venire. We cannot say that the 36% elevated rate of strikes shows that the state court acted contrary to "clearly established Federal law, as determined by the Supreme Court," 28 U.S.C. § 2254(d).
This conclusion is consistent with other habeas cases that have found no error in a state court's declining to draw an inference of discrimination. For example, in Collado v. Miller, 157 F. Supp. 2d 227 (E.D.N.Y. 2001), the prosecution used (at most) 39% (7/18) of its challenges to strike Hispanic jurors, while the Hispanic population of the county from which the jury was drawn — used as a proxy for the racial composition of the venire — was 25%. The increase of actual challenges over expected challenges was 56%, an increase far greater than the 36% increase in this case. Denying the petition for writ of habeas corpus, the court held that
[a]bsent any other indicia of racial bias, the disparity here between the rate of the prosecution's challenges of Hispanics and the racial composition of the venire is not sufficiently significant for this Court to find that the state court's decision was per se incorrect. Although another court might find a prima facie Batson violation based on the statistical disparity in petitioner's case, the state court did not necessarily err in concluding otherwise, and certainly did not advance an "objectively unreasonable" application of Batson.Id. at 234. Other habeas cases in the Second Circuit have similarly concluded that the state court acted properly where the increase in the actual challenge rate over the expected challenge rate was larger than the 36% increase in this case. See, e.g., Collins v. Beaver, 2005 WL 1801603, at *7 (E.D.N.Y. July 27, 2005) (50% increase); Rodriguez v. Lord, 2001 WL 1223864, at *20 (S.D.N.Y. Oct. 15, 2001) (51% increase).
A sampling of cases reveals that in instances where courts found that a prima facie case had been made, the court each time saw percentage increases over the expected number of strikes far greater than the 36% seen here — even if the courts did not themselves always make a direct comparison of percentages. See, e.g., Johnson, 125 S. Ct. at 2417 (258%); Alvarado II, 932 F.2d at 255 (97%); Moore, 4 F. Supp. 2d at 320-21 (72%);Anderson, 360 F. Supp. 2d at 480-82, 493, 494, 496 (61-79%);Rose v. Senkowski, 2003 WL 21698240, at *3, *6, *13 (E.D.N.Y. July 8, 2003) (110-144%); Turner, 63 F.3d at 813 (87%);Jones, 987 F.2d at 971 (275%); see also Miller-El v. Cockrell, 537 U.S. 322, 331, 342 (2003) (173%) (analyzing the third step of a Batson claim).
Nonetheless, this Court is loathe to rest its analysis solely on the comparison of raw percentages that is typically undertaken. This is because a comparison of these percentages does not provide a true assessment of the degree to which an attorney's strikes of jurors of a protected group are due to chance. As noted, the core question in determining a prima facie Batson violation is whether the actual strikes made by a prosecutor are so disproportionate to the expected strikes that we have reason to believe that the difference is not due merely to chance. See, e.g., Miller-El, 537 U.S. at 342 (noting that "[h]appenstance [was] unlikely to produce [the] disparity" between the number of African-Americans on the venire and the number actually stricken). A more precise methodology for determining whether a particular result is due to "happenstance" is to use measures of statistical significance.
The method for determining statistical significance that has been explicitly approved by the Supreme Court in the context of evaluating alleged patterns of discrimination in the grand jury selection process is that of standard deviations. In Castaneda v. Partida, which dealt with the use of statistics to determine if a grand jury fairly reflected the population from which it had been convened, the Supreme Court explained the standard deviation calculation as follows:
[I]n any given drawing some fluctuation from the expected number is predicted. The important point, however, is that the statistical model shows that the results of a random drawing are likely to fall in the vicinity of the expected value. The measure of the predicted fluctuations from the expected value is the standard deviation, defined for the binomial distribution as the square root of the product of the total number in the sample . . . times the probability of selecting [a juror from a particular group] . . . times the probability of [not] selecting a [juror from that group] . . . As a general rule for . . . large samples, if the difference between the expected value and the observed number is greater than two or three standard deviations, then the hypothesis that the jury drawing was random would be suspect to a social scientist.430 U.S. 482, 496 n. 17 (1977) (citation omitted). The test has been used in other instances as well. As the Second Circuit has described the test in a housing discrimination case:
A standard tool for assessing statistical significance is the two-standard deviation test. The test is used to determine whether a deviation from the expected norm is small enough to be attributable to chance, or so large that random chance could not reasonably account for the outcome. The greater the number of standard deviations, the less likely it is that chance is the cause of any difference between the expected and observed results.Davis v. New York City Housing Authority, 278 F.3d 64, 75 (2d Cir. 2002) (citations and internal quotation marks omitted). The two-standard-deviation threshold has frequently been used for purposes of determining whether an "inference" of discrimination has been raised in the employment discrimination context. See, e.g., Smith v. Xerox Corp., 196 F.3d 358, 366 (2d Cir. 1999) (two standard deviations is "generally consider[ed] th[e] level of significance sufficient to warrant an inference of discrimination" in employment discrimination cases) (citingOttaviani v. State Univ. of New York, 875 F.2d 365, 371-72 (2d Cir. 1989)).
A standard deviation analysis provides a more accurate methodology for gauging whether strikes by an attorney are "significantly higher" than expected, Alvarado II, 923 F.2d at 255, than a mere comparison of the percentage increase in the actual strikes over the expected strikes. The standard deviation analysis is superior to a comparison of raw percentages because it gives a more precise description of the significance that should be attached to the difference between an actual result and an expected result.
One court reviewing a Batson claim has agreed in principle, though it did not actually apply the analysis. See Wilson v. Cockrell, 2002 WL 32590134, at *7 (N.D. Tex. Sept. 24, 2002). A commentator has argued that a more refined statistical test would be a better formula for judging statistical significance in Batson challenges than a standard deviation analysis. See Stephen R. DiPrima, Selecting a Jury in Federal Criminal Trials After Batson and McCollum, 95 Colum. L. Rev. 888, 922-24 n. 208 (1995). Application of this formula, however, is complex and thus unsuited to judicial review — at least in the absence of testimony or other evidence from an expert witness. Accordingly, we feel compelled to rely on the standard deviation formula since it represents a far better mechanism for measuring statistical significance than the comparison of raw percentages.
Consider, for example, a jury pool that is 6.5% minority and a prosecutor who uses 2 of her 16 strikes, that is, 12.5%, to eliminate minority jurors. One might assume that an inference of discrimination was appropriate because the rate of strikes was almost twice (92% greater than) the actual percentage of minorities in the jury pool. This result might be said to be consistent with Alvarado II, which found that a 97% difference was enough to raise an inference of discrimination. 932 F.2d at 255. On the other hand, we also intuitively recognize that the difference between the expected rate of strikes (1 in 16, or 6.25%) and the actual rate (2 in 16) reflects a difference of only one strike, and may not be significant. Thus, we would probably conclude that, despite the 97% increase reflected in the actual strikes, an inference of discrimination should not necessarily be drawn.
As it turns out, this was precisely the result reached inSoria v. Johnson, 207 F.3d 232, 237-38 (5th Cir.), cert. denied, 530 U.S. 1286 (2000), from which these facts are taken. The Fifth Circuit in Soria held that "[a] deviation from the norm of but a single strike simply does not so clearly raise an inference of racial discrimination that a factfinding to the contrary must be disturbed on appeal." 207 F.3d at 238; accord Jamison v. Duncan, 2001 WL 1352918, at *9 (S.D.N.Y. Nov. 2, 2001). Notably, the actual number of strikes in the Soria case was only .97 standard deviation away from the expected number, far less than the "two or three" standard deviations test articulated in Castaneda.
Castaneda recommended the "two or three" standard deviation test for "large samples." 430 U.S. at 496 n. 17. While it is unclear if Castaneda was implying that smaller samples should require a greater number of standard deviations, there is certainly no reason to use any lesser number merely because the sample size is small. See generally Vogelsang v. Patterson Dental Co., 1989 WL 89951, at *6 (D. Minn. Feb. 17, 1989) ("[W]hen working with a small sample size, a court may require greater statistical disparity in order to infer discriminatory bias . . . [G]iven the small sample size, it would be imprudent for the Court to rely on a statistical disparity of less than two standard deviations as proof sufficient to establish a prima facie case of age discrimination.") (citing EEOC v. American National Bank, 680 F.2d 965, 969 n. 6 (4th Cir. 1982)).
In the employment discrimination context, the Second Circuit also has recognized that the value of the standard deviation analysis "drops off" when small sample sizes are used. Waisome v. Port Authority of New York and New Jersey, 948 F.2d 1370, 1379 (2d Cir. 1991). Waisome concluded that this means only that "other indicia raising an inference of discrimination must be examined." Id. Thus, nothing in Waisome suggests that a small sample size should result in the use of a lesser number of standard deviations as a yardstick — let alone that it should cause a court to resort to the inexact process of comparing raw percentages.
Conversely, sometimes the standard deviation analysis will raise an inference of discrimination that would not have been readily apparent from a mere comparison of raw percentages. This is particularly likely when larger numbers are considered. For example, imagine a venire that was 60% minority and a prosecutor who used 13 of 15 (that is, 87%) of her strikes against minorities. On a raw percentage basis, the increase in the actual strikes (87%) over the expected strikes (60%) would be only 44%. But a standard deviation analysis would show that the actual number of strikes, 13, was 2.1 standard deviations away from the expected number of strikes (9, or 60% of 15). In such a situation, a court might conclude that the strikes raised an inference of discrimination.
In Truesdale's case, the actual number of strikes (8) differs from expected number of strikes (5.89) by 1.69 standard deviations. Under the Castaneda standard and the "two-standarddeviation" test used in the employment context, see Smith, 196 F.3d at 366, we would not conclude that the difference between the observed and expected rate was statistically significant forBatson purposes on habeas review.
We calculate this number as follows: The standard deviation equals the square root of the product of 8 (the number of challenges) times 14/19 times 5/19, which is 1.2455. The expected number of challenges against African-American jurors was 5.89 (that is, the product of the prosecutor's challenges (8) times the likelihood of picking an African-American juror (14/19)). To determine the number of standard deviations by which the actual chosen number deviated from the expected number, we subtract the expected number of struck jurors (5.89) from the actual number (8) and divide this number by the standard deviation. Thus, the observed result (8 challenges against African-American jurors) was 1.69 standard deviations away from the expected result, 5.89.
This Court could see good reason to consider the possibility that an inference of discrimination might be raised even for standard deviations of less than "two or three." See, e.g., Kilgo v. Bowman Transp., Inc., 789 F.2d 859, 872 (11th Cir. 1986) (cautioning against drawing a bright-line rule as to the number of standard deviations to be used as a threshold for determining disparate impact in an employment discrimination case) (citing cases). But for purposes of habeas review, where we must decide only if the state court engaged in an "unreasonable application" of federal law, it is proper to defer to a state court's decision refusing to attach an inference of discrimination to a pattern of strikes that turns out to be below the level of significance that was explicitly approved by the Supreme Court in a related context, see Castaneda, 430 U.S. at 496 n. 17.
For example, in Alvarado II, the court compared the percentage of minorities in the venire to the percentage of strikes (29% vs. 57%). 923 F.2d at 255. Because the strikes were made at "nearly twice" the rate that was expected, id., Alvarado II concluded that an inference of discrimination should be drawn. But if the standard deviation analysis had been performed on the strikes in Alvarado II, the number of actual strikes (4/7) was only 1.5 standard deviations away from the expected strikes (2.0), a number less than the "two or three standard deviations" bar set in Castaneda, see 430 U.S. at 496 n. 17. Alvarado II is a case arising on direct review, however, and thus, unlike the present case, it was not required to give special deference to the determination of the trial court.
In addition to his arguments comparing the percentage of African-Americans struck to the percentage of African-Americans on the venire, Truesdale makes another statistical argument using the identical figures. He argues that the rate at which African-Americans were struck — 57% (8 out of 14) — by itself supports an inference of discrimination in his case. Pet. Am. Mem. at 25, 27-28. He points in particular to Turner, which, in finding a prima facie case, considered "the statistical fact that the prosecutor had used peremptory challenges to exclude five African-Americans out of a possible nine African-American venirepersons." 63 F.3d at 812. Truesdale argues that the percentage at issue in Turner (56%) is not so different from the percentage challenged here, 57% (8 out of 14) and thus provides an inference of discrimination. Pet. Am. Mem. at 27-28.
The flaw in this reasoning is that the ratio of struck members of a group to the total number of members of that group is meaningless in the absence of information as to the context in which that exclusion took place. For example, even a 100% exclusion rate might be insignificant. If an available venire consisted of 12 persons, only 1 of whom was of the protected group, and the prosecutor struck 11 venirepersons, including the sole member of the protected group, the resulting 100% exclusion rate would not cause concern. This is because the percentage of strikes used against the protected group must be compared with the percentage of the available venire as a whole that was stricken. In this example, the 100% rate of strikes used against the protected group (1/1) must be compared to the 92% (11/12) of the strikes used against the available venire. Our instinct that the 100% exclusion rate is not significant in this context is confirmed by the fact that the difference between the actual number of persons in the protected group stricken (1) and the expected number (.92) is not statistically significant (.3 standard deviations).
Of course, this principle works in the other direction. A low exclusion rate would be suspicious if it represented a disproportionate share of the attorney's overall exclusions. Thus, if a prosecutor struck 4 African-Americans out of 12 African-Americans on a 40-person panel, and used 1 strike to remove a non-African-American, the rate at which African-Americans were stricken would be only 33% (4 out of 12). But the 33% figure must be compared with the percentage of the available venire as a whole that was stricken, which is 12.5% (5 out of 40). In these circumstances, the 33% strike rate might raise an inference of discrimination because, based on the number of African-Americans in the available venire (12) and the percentage of the venire that was stricken (12.5%), it should have been expected that 1.5 African-Americans would be stricken (12 times 12.5%). The difference between 1.5 strikes and 4 strikes turns out to be statistically significant based on the standard deviation test (2.18 standard deviations).
Thus, in Truesdale's case, the 57% rate of strikes against African-Americans (8 out of 14) must be compared with the 42% rate of strikes used against the total population (8 out of 19). Not only do these percentages seem close on a raw basis, the difference in these numbers is not statistically significant (1.14 standard deviations).
In a related vein, Truesdale points to the fact that the prosecutor challenged 0% of the non-African-American prospective jurors. See Pet. Am. Mem. at 27; Pet. Reply Mem. at 9. Again, this percentage is meaningful only in the context of the pattern of overall strikes. Thus, the 0% number must be compared with the 42% rate of strikes used against the total population (8 out of 19). While these two percentages (0% vs. 42%) reflect a significant disparity in absolute terms, the difference between the number of strikes of non-African-Americans we would have expected based on chance, 2 (that is 42% of 5), and the number of strikes actually made (0) is not great in absolute terms — rather, it is a difference of only two strikes. Accordingly, it is not surprising to find that the 0 strikes represents 1.9 standard deviations from the expected number of 2 — again, less than the "two or three" standard deviations contemplated byCastaneda.
We recognize that it would be absurd to expect a trial judge in the middle of voir dire to undertake the sort of mathematical analysis conducted herein to determine what inference should be drawn from a particular pattern of strikes. At the same time, we are convinced that a reviewing court cannot simply compare percentages of actual strikes to percentages of expected strikes without an understanding of whether the difference between those percentages has any statistical significance. The standard deviation analysis, while not perfect and perhaps not even dispositive, is certainly a better tool for conducting a review of whether a trial court properly drew or failed to draw an inference of discrimination than the mere comparison of percentages. If nothing else, it serves a useful check on a court's intuition regarding whether inferences of discrimination should be drawn from a pattern of strikes.
2. Whether There is Additional Evidence of a Pattern of Discrimination
While statistics alone may be sufficient to raise an inference of discrimination, they are not necessary to make such a showing. That is, other factors may be used, either alone or in combination with a statistical showing, to meet step one of theBatson process. See Batson, 476 U.S. at 97. The only non-statistical piece of evidence offered by Truesdale consists of one of the prosecutor's comments made during his Batson objection. Pet. Am. Mem. at 30. The prosecutor noted that all the prospective jurors in the first round available to him were "non-white," following which counsel for a co-defendant alleged that this was false because one of those jurors was "not an African-American." (V. 36). The trial judge agreed with defense counsel. (V. 36). Truesdale argues that this "supports the inference that the prosecutor directed 100% of his challenges against African-Americans by choice, and not be [sic] necessity." Pet. Am. Mem. at 31. It is unclear why Truesdale reaches this conclusion as there is no apparent connection between the prosecutor's purported error regarding the race of one juror and the reasons that underlay his challenges. To the extent Truesdale is arguing that the prosecutor's mentioning race at all reflects discriminatory intent, this argument is meritless as the prosecutor discussed the juror's race only because defense counsel had made the Batson challenge. Thus, he appropriately tried to identify the races of members of the first round of venirepersons. To the extent Truesdale's argument rests on the assumption that the prosecutor was mistaken as to the race of the juror at issue, Truesdale is incorrect. It is possible that this juror, Helen Sabella-Hernandez, could have been accurately characterized as both "non-white" and "not an African-American." (V2. 36). Finally, even if the prosecutor had actually misidentified a non-African-American juror as African-American, there is no logical connection between this mistake and the inference that the prosecutor had deliberately stricken the other jurors on account of their race.
Truesdale points out that five of the six African-American jurors who were not challenged by the prosecution had law enforcement connections and makes an additional argument, not presented before the Appellate Division, that "the prosecutor appears to have excepted from his pattern of discrimination jurors who were overtly pro-prosecution." Pet. Am. Mem. at 30. While these facts might constitute evidence relevant to a claim of discrimination by the prosecutor, they are insufficient to make the state courts' finding that there was no inference of discrimination an unreasonable application of Supreme Court precedent.
3. Summary
The Appellate Division did not unreasonably apply Supreme Court law in concluding that the prosecutor's conduct did not raise an inference of discrimination. Conclusion
In light of this recommended disposition, it is unnecessary to address respondent's argument that the Batson claim should be rejected because trial counsel failed to create a record sufficient to establish a prima facie case. See Resp. Mem. at 15 ("defense counsel never calculated, or even hinted at, the percentage of black prospective jurors in the venire or excluded by the prosecutor's peremptory challenges").
For the foregoing reasons, Truesdale's petition should be denied.
PROCEDURE FOR FILING OBJECTIONS TO THIS REPORT AND RECOMMENDATION
Pursuant to 28 U.S.C. § 636(b)(1) and Rule 72(b) of the Federal Rules of Civil Procedure, the parties have ten (10) days from service of this Report and Recommendation to file any objections. See also Fed.R.Civ.P. 6(a), (e). Such objections (and any responses to objections) shall be filed with the Clerk of the Court, with copies sent to the Hon. Denise L. Cote, 500 Pearl Street, New York, New York 10007, and to the undersigned at 40 Centre Street, New York, New York 10007. Any request for an extension of time to file objections must be directed to Judge Cote. If a party fails to file timely objections, that party will not be permitted to raise any objections to this Report and Recommendation on appeal. See Thomas v. Arn, 474 U.S. 140 (1985).