Summary
holding that an attorney's failure to object to plain prosecutorial misconduct cannot constitute legitimate trial strategy
Summary of this case from Hanna v. PriceOpinion
No. 98-2250.
Argued: March 14, 2000.
Decided and Filed: October 6, 2000.
Appeal from the United States District Court for the Eastern District of Michigan at Detroit. No. 97-71849 — Gerald E. Rosen, District Judge.
ARGUED: Debra A. Gutierrez, STATE APPELLATE DEFENDER OFFICE, Detroit, Michigan, for Appellant.
Janet A. Van Cleve, OFFICE OF THE ATTORNEY GENERAL, HABEAS CORPUS DIVISION, Lansing, Michigan, for Appellee.
ON BRIEF: Debra A. Gutierrez, STATE APPELLATE DEFENDER OFFICE, Detroit, Michigan, for Appellant.
Janet A. Van Cleve, OFFICE OF THE ATTORNEY GENERAL, HABEAS CORPUS DIVISION, Lansing, Michigan, for Appellee.
Before: JONES, BATCHELDER, and CLAY, Circuit Judges.
The petitioner, Rufus Washington, a Michigan prisoner convicted of second-degree criminal sexual conduct, appeals the district court's dismissal of his § 2254 habeas petition. His petition alleged prosecutorial misconduct and ineffective assistance of counsel. We find that the prosecutor's misconduct was sufficiently egregious to violate Washington's due process rights, that Washington's trial counsel was ineffective in not objecting to that conduct, and that the state court did not reasonably apply the relevant law in finding otherwise. We are thus compelled to REVERSE and issue the writ.
I.
In 1991, Washington was charged with first-degree criminal sexual conduct against complainant Tamara Beard in violation of Mich. Comp. Laws § 750.520b(1)(a), and with being a habitual offender-fourth offense under Mich. Comp. Laws § 769.12(a). On January 8, 1991, a Michigan jury found Washington guilty of second-degree criminal sexual conduct, in violation of Mich. Comp. Laws § 750.520c. Washington then pleaded guilty to being a habitual offender-fourth offense. On March 7, 1991, the trial court sentenced Washington to twenty to thirty years' imprisonment, which was later reduced to seventeen to thirty years' imprisonment.
A.
At the time of the crime, Washington was the boyfriend of Ms. Cora Beard, the mother of the complainant. Washington and Beard lived together, along with Beard's five sons and one daughter. Tamara Beard, the daughter, testified at trial that Washington had sexually abused her on two occasions in 1989. The first incident occurred when Tamara was nine years old. Tamara testified that on Memorial Day, 1989, she, Washington, her mother, and a brother traveled to Lansing, Michigan. On arriving home, she and her brother went to sleep on a couch in the den, while her mother went to work. Later that night, Tamara awoke in her mother's bedroom, with her brother asleep next to her; she did not remember how she got there. Washington was also in the bed. According to Tamara, Washington then ordered her to take off her gown and panties or he would kill her. He then inserted a finger in her vagina. When Tamara asked him to stop, he refused. J.A. at 154. He did not stop for "a long time." J.A. at 155. Washington then ordered Tamara to hold his penis, which she did. Tamara testified that she did not tell her mother of the incident because Washington had threatened to kill her if she did so. In June 1989, a similar episode occurred. Again, after falling asleep on the couch, Tamara awoke to Washington's demand that she remove her panties. He once again inserted his finger into her vagina. While the family continued to live with Washington, Tamara told no one of these incidents.
Tamara was 11 years old at the time of the trial.
In February 1990, Cora Beard moved out of the house she shared with Washington and began to stay with her mother, bringing her three youngest children with her — including Tamara. At trial, she explained that she moved out because Washington had been both mentally and physically abusive, and because he had begun drinking heavily. After a short time, and after Washington pleaded for her to return, Beard decided to move back into the house with Washington. At that time, Tamara told her mother about Washington's actions the prior year. Cora Beard immediately took Tamara to the hospital, where a social worker examined her. Beard also took Tamara to Dr. Joyce Woodson of the Michigan Department of Social Services, who interviewed Tamara on four occasions. After these interviews, Dr. Woodson contacted the Flint police.
At trial, although Washington acknowledged that he and Cora Beard often fought, and that he drank alcohol, he denied committing the alleged assaults on Tamara: "I never touched Tamara sexually in any form." J.A. at 200. He explained further:
[T]he only conclusion I can come to is that she wanted to maybe protect her mother or there's a possibility that they wasn't visiting their grandmother that regular and that she might want to stay over there, that's the only reason I can think of, you know. Because I certainly didn't ever mistreat [the children].
J.A. at 200. On cross-examination, Washington denied the prosecution's questions about whether he drank heavily. When the prosecutor asked whether he "d[id] something to Tamara," Washington responded, "[n]o, I didn't." J.A. at 209. Although he acknowledged that he had at one time told Cora Beard he would kill her, he denied ever threatening Tamara. J.A. at 211.
A month passed between the time of Washington's testimony and closing arguments. This is because at the close of the State's case, Washington's counsel requested that Dr. Wayne Ross, the doctor who examined Tamara at the hospital, testify; Dr. Ross's testimony was important to Washington's case because after examining Tamara, the doctor had noted that Tamara "denies penetration," J.A. at 191. Because Dr. Ross was bedridden after emergency back surgery and unavailable for trial, the court agreed to take Washington's testimony out of order. The trial was thereafter adjourned until Dr. Ross could testify. On January 8, 1991, with Dr. Ross still bed-ridden, the parties agreed to stipulate to the content of Dr. Ross's notes. Specifically, the court reporter read into evidence the following notes: "Ten month's ago mom's live in touched her down there, referring to the crotch area. Patient denies penetration. Child playful." J.A. at 235.
B.
Washington raises two issues in this appeal. First, he challenges several instances of alleged prosecutorial misconduct during the State's examination of Washington and closing argument. Second, he challenges as constitutionally ineffective his counsel's failure to object to this misconduct, as well as his flawed articulation of the burden of proof. These issues require a detailed look at portions of the trial record.
The record shows that the prosecutor extensively berated Washington's character before the jury, emphasizing that Washington did not work, beat Cora Beard regularly, consumed alcohol excessively, and did not make payments on Cora's home. These statements emerged initially in the State's cross-examination of Washington. First, the prosecutor emphasized that Washington moved into the Beard's home — that "it was their home," not his. J.A. at 201. At a later point, the prosecutor tried to elicit that the car Washington and Cora Beard used was not Washington's, but hers. J.A. at 206 (noting that Washington "didn't have a car" himself). The prosecutor then questioned Washington's work history, doubting the reasons he provided to explain periods when he was out of work: "You're telling the jury the virus kept you from working? . . . Then all of a sudden you got this virus? . . . So you didn't keep this other job too long either then?" J.A. at 201-02. Next, the prosecutor emphasized that Washington "drank a lot" when he was living at the Beard home. J.A. at 202. Finally, the prosecutor emphasized several times that Washington "smacked" Cora Beard regularly. J.A. at 203-08. Only for a brief portion of the cross-examination did the prosecutor focus on the alleged acts against Tamara Beard, which Washington denied. During this cross-examination, defense counsel, Sanford Keston, only objected one time — stating, after the prosecutor asked Washington if it was his "life style to smack the lady," that he was "repeating the same question." J.A. at 210.
The prosecutor again focused on Washington's character during his closing argument. Summarizing the sequence of events, he reminded the jury that "Rufus here stays in the house that he doesn't support, that he doesn't own." J.A. at 246. The prosecutor speculated as to why Tamara did not initially tell her mother of the incidents: "You heard him testify that when things didn't go his way, he just kind of smacked the girlfriend whose house he was living in. . . . I think he talked about hitting and smacking and it wasn't hard but it was pretty hard. . . ." J.A. at 247-48. This showed that Tamara "ha[d] reasons to be fearful of some guy who says he's going to kill her. . . ." J.A. at 249. Finally, the prosecutor asked the jury to consider whether Tamara's allegations "fit" the general description they had heard of Washington's character. He then answered the question himself: "I'll tell you this, it fits what we know about him in terms of the evidence of this case, what you've been told about him." J.A. at 254.
He certainly would fit that kind of pattern that was his kind of lifestyle that may allow it to happen. The alcoholism, the drugs as testified to by the mother. . . . The irrational behavior. . . . That he's living in her home, driving her car, kind of smacks her around when something doesn't quite suit him. Does that sound like a logical compassionate family type of person or does that sound like someone who lives a little bit out of control, someone who has no control in fact over there [sic] own life, someone who doesn't act according to logic. . . . Whether the motivation is a pervert, a child molester or just a selfish alcoholic inhumane, for whatever reason the results are the same on the victim. . . .
J.A. at 254-55. Keston did not object to these statements.
Finally, in his rebuttal, the prosecutor directly instructed the jury to consider Washington's character in deciding whether he committed the alleged crime.
Does what Tamara [told] you happened fit him in the testimony that was told you about him, coming from other people and from his own mouth. . . . [D]oes it fit the self-serving, illogical selfish non-compassionate, no emotional interest in a family type of person? . . . I mean it would be different if he was portrayed in a different light during this case and it was something contradictory with what Tamara testified happened. . . .
J.A. at 270-71. He asked whether Washington was "a person that just acted irrational caused by drugs and alcoholism and a general not caring about other people, just caring about him and his resentment," and concluded once again that the alleged crime "sure fits him." J.A. at 271. In his penultimate statement, he urged the jury to consider Washington's "lifestyle" in assessing the case: "All I ask you to do is look at the case carefully," including "Rufus Washington's character as he portrayed it to you on the stand." J.A. at 275.
Washington also argues that the prosecutor committed misconduct when, in closing, he characterized facts never introduced into evidence. Specifically, the prosecutor stated:
This child talked to her mother[;] this child talked to the doctor. This child talked to the social service worker in detail. She testified. This child talked to Sergeant Elford in detail. This child went through a preliminary examination and cross examination where there was cross examination and this child testified before you and nowhere for the most part based upon what happened, has it changed.
J.A. at 255 (emphasis added). Washington argues that none of the people the prosecutor referred to had testified about what Tamara had told them — nor could they have, since such testimony would have been inadmissible hearsay. Therefore, the prosecutor's statement that the story had never changed was not supported by any evidence. Once again, Keston did not object to this statement.
Finally, Washington argues that Keston was constitutionally ineffective for misstating the prosecution's burden of proof. Indeed, the record evinces that on several occasions during his closing argument, Keston stated that the State had to prove its case by clear and convincing evidence. Simultaneously, however, Keston several times mentioned that the prosecutor had to prove every element of the alleged crime beyond a reasonable doubt.
C.
Ultimately, the jury acquitted Washington on the charged offense of first-degree criminal sexual conduct, which requires a finding of penetration. See Mich. Comp. Laws § 750.520b(1)(a). It returned a guilty verdict on second-degree criminal sexual conduct, which does not require penetration. See Mich. Comp. Laws § 750.520c.
After sentencing, Washington filed an appeal for his conviction and sentence with the Michigan Court of Appeals, and filed a motion to remand to the trial court for a Ginther hearing on his claim of ineffective assistance of counsel. The appeals court granted the motion, and the trial court held a hearing on September 9, 1993, where it concluded that Keston had not been sufficiently ineffective to violate Strickland. Regarding Keston's failure to object to the prosecutor's character attacks, the court held that the evidence was not objectionable because it went to the victim's fear of Washington. Next, the court concluded that Keston had not been ineffective for failing to object to the prosecution's statement that Tamara's story had never changed, finding Keston's explanation to reflect sound trial strategy. Finally, the court concluded that although Keston used the words "clear and convincing" in his opening and closing arguments, the jury had been properly instructed on the reasonable doubt standard. In sum, the court concluded, trial counsel's actions were not "so serious that counsel was not functioning as the `counsel' guaranteed the defendant by the Sixth Amendment." J.A. at 51 (quoting Strickland v. Washington, 466 U.S. 668, 687 (1984)). "Defendant was charged with CSC, first degree and was convicted of the lesser offense of CSC, second degree. On balance defense counsel appears to have done a good job with a very difficult case." J.A. at 51.
In Michigan, a hearing on ineffective assistance claims is held pursuant to People v. Ginther, 212 N.W.2d 922 (Mich. 1973).
On April 10, 1995, the Michigan Court of Appeals reviewed Washington's case a second time. The court did not consider his claim for prosecutorial misconduct because Keston had not objected to the allegedly improper remarks. The appeals court then agreed with the trial court's conclusion that Keston's failure to object "was a matter of trial strategy," J.A. at 73, and that Keston's misstatement of the burden of proof was not prejudicial. The court also rejected other challenges to the conviction not brought before this court. Meanwhile, the court reversed Washington's sentence and remanded to the trial court for resentencing. The Michigan Supreme Court denied Washington leave to appeal. On December 10, 1996, Washington was resentenced to a term of seventeen to thirty years' imprisonment.
In 1998, Washington filed a habeas petition in the district court pursuant to 28 U.S.C. § 2254. On September 18, 1998, Magistrate Judge Donald Scheer issued a report and recommendation that the petition be denied, to which Washington filed objections on September 24. On October 7, 1998, Judge Gerald Rosen denied Washington's petition. Petitioner filed a claim of appeal on October 29, 1998. On April 5, 1999, this Court granted a certificate of appealability.
II.
Because Washington's petition was filed after April 24, 1996, the rules of the Antiterrorism and Effective Death Penalty Act of 1996 ("AEDPA") apply. See Tucker v. Prelesnik, 181 F.3d 747, 752 (6th Cir. 1999). This means that a writ of habeas corpus may not be granted unless the state court proceedings
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 proceedings.
28 U.S.C.A. § 2254(d) (West 1999). The Supreme Court recently instructed that § 2254(d)(1) defines two categories of cases in which a state prisoner may gain habeas relief. See Williams v. Taylor, 120 S.Ct. 1495, 1519-20 (2000) (O'Connor, J., concurring). To gain habeas relief under the first category, involving state decisions "contrary to" federal law, a defendant must show that "the state court arrive[d] at a conclusion opposite to that reached by [the Supreme] Court on a question of law" or that "the state court decide[d] a case differently than [the Supreme] Court has on a set of materially indistinguishable facts." Id. at 1523. Under the second category, involving the "unreasonable application of" federal law by a state court, a federal habeas court must ask whether the state court's application of clearly established federal law was "objectively reasonable." Id. at 1521. If the federal court finds that, viewed objectively, the state court has correctly identified the governing legal principle from the Supreme Court's decisions "but unreasonably applie[d] that principle to the facts of the prisoner's case," it may grant the writ. Id. at 1523. In Williams, the Supreme Court rejected the Fourth Circuit's definition of an "unreasonable application" of the law by reference to a "reasonable jurist." Id. at 1521-22. By implication, the standard for AEDPA review this Court set forth in Tucker is also overruled. See 181 F.3d at 753 (stating that a writ will issue "if the unreasonableness of the state court's application of clearly established precedent is not debatable among reasonable jurists").
III.
Neither party disputes that Washington's claim of prosecutorial misconduct is procedurally barred absent a showing of cause and prejudice, and both recognize that ineffective assistance of counsel can provide the necessary "cause" for the procedural default. But Washington's claim of Strickland ineffectiveness hinges on whether the prosecutor's misconduct was plain enough for a minimally competent counsel to have objected. For the sake of clarity, we therefore will address Washington's appeal in three steps. First, we will consider whether the prosecutor's conduct was plainly improper. Second, we will consider whether Keston was constitutionally ineffective for failing to object to the prosecutorial misconduct. Third, we will consider whether that ineffectiveness satisfies the "cause and prejudice" exception to procedural default, allowing us to ask whether the misconduct itself provides grounds for habeas relief. We find for Washington on all three questions.
A.
Before assessing whether Keston was ineffective for failing to object to the prosecutor's actions, we must first determine whether the prosecutor committed misconduct. See generally Cobb v. Perini, 832 F.2d 342, 347-48 (6th Cir. 1987) (rejecting claim that counsel's failure to object comprised ineffectiveness in part because it was unclear whether challenged evidence was improper); Barton v. Morris, No. 95-3848, 1996 WL 408504, at *2 (6th Cir. July 19, 1996) (unpublished decision) (concluding that counsel's failure to object to prosecutor's closing argument was not ineffective because "those comments did not amount to prosecutorial misconduct and would not have provided the basis for action by the trial judge"). Juxtaposing precedent from this circuit alongside the trial record convinces us that several aspects of the prosecutor's behavior clearly crossed the line into plain and prejudicial impropriety.
1.
First, we address the prosecution's emphasis on Washington's "bad character." A fundamental rule of evidence is that a defendant's "bad character" cannot be used to argue that the defendant committed the crime for which he is being tried, or had the propensity to commit that crime. See, e.g., Fed.R.Evid. 404(a) ("Evidence of a person's character or a trait of character is not admissible for the purpose of proving action in conformity therewith on a particular occasion. . . ."); Mich. R. Evid. 404(a) (same); Michelson v. United States, 335 U.S. 469, 476 (1948) (stating that improper character evidence "weigh[s] too much with the jury and . . . overpersuade[s] them as to prejudge one with a bad general record and deny him a fair opportunity to defend against a particular charge"); United States v. Vance, 871 F.2d 572, 575 (6th Cir. 1989) (providing that "bad acts evidence is not admissible to prove character or criminal propensity" under Fed.R.Evid. 404(b)); United States v. Ring, 513 F.2d 1001, 1004 (6th Cir. 1975) (stating that in jury trials, evidence of a criminal defendant's bad acts or prior misconduct is inadmissible to show criminal propensity because it "tends to confuse the issue of guilt or innocence of the specific offenses charged and to weigh too heavily with the jury"). When a prosecutor dwells on a defendant's bad character in this prohibited manner, we may find prosecutorial misconduct. See, e.g., Cook v. Bordenkircher, 602 F.2d 117, 120 (6th Cir. 1979) (noting that the "prosecutor's misconduct in this case is severe" due to his "persistent Ad homonym attack on the petitioner's character").
In this case, we find that while the evidence as to Washington's character was admissible for certain limited purposes, the prosecutor went far beyond the bounds of permitted conduct when presenting that evidence to the jury. Because Keston introduced much of this evidence as part of his defense strategy, see infra, and because aspects of Washington's character shed light on why Tamara had not informed others of the alleged acts, we do not find that the State's cross-examination of Washington constituted prejudicial misconduct on its own. However, the prosecutor's animated recitation of this character evidence during closing arguments was plainly improper. In his initial summation, the prosecutor improperly implied that the jurors should consider Washington's unseemly character when rendering their verdict; in his rebuttal, he explicitly urged them to do so. Meanwhile, he attacked Washington as a "self-serving, illogical selfish non-compassionate, no emotional interest in a family type of person," who acted irrational due to "drugs and alcoholism and a general not caring about other people." J.A. at 270-71. The crime, he implored to the jury, "[s]ure fits him." J.A. at 271. The prosecutor thus articulated perhaps the paradigm of the improper "bad character" argument — that the alleged criminal acts "fit" the evidence of Washington's character and lifestyle. Because this character attack pervaded the closing argument and rebuttal, we find that the prosecutor's misconduct was severe. See Cook, 602 F.2d at 120 (making the same conclusion after a pervasive character attack).
2.
We also agree with Petitioner that the prosecutor engaged in serious misconduct when he characterized Tamara's story as having been consistent over time when there was no evidence supporting that factual assertion.
Misrepresenting facts in evidence can amount to substantial error because doing so "may profoundly impress a jury and may have a significant impact on the jury's deliberations." Donnelly v. DeChristoforo, 416 U.S. 637, 646 (1974). For similar reasons, asserting facts that were never admitted into evidence may mislead a jury in a prejudicial way. See Berger v. United States, 295 U.S. 78, 84 (1935). This is particularly true when a prosecutor misrepresents evidence because a jury generally has confidence that a prosecuting attorney is faithfully observing his obligation as a representative of a sovereignty. See id. at 88.
Given this precedent, characterizing Tamara's conversations with different individuals as consistent comprised clear prosecutorial misconduct. The State suggested that Tamara had been consistent when it stated the following:
This child talked to her mother, this child talked to the doctor. This child talked to the social service worker in detail. She testified. This child talked to Sergeant Elford in detail. This child went through preliminary examination and cross examination where there was cross examination and this child testified before you and nowhere for the most part based upon what happened, has it changed.
J.A. at 255 (emphasis added). Yet apart from the doctor's notes — to which the parties stipulated — the prosecutor elicited no evidence on the specifics of Tamara's conversations with any of these individuals, establishing only that conversations had occurred. Surely, then, there was no evidence as to whether or not her story had changed.
When Tamara herself testified, she stated only that she had conversations with the referenced people, and that she had told them "what happened." The prosecutor did not ask her to describe the details of those conversations, nor did she volunteer them. Moreover, no other witnesses testified about what Tamara told them because such testimony would have been inadmissible hearsay. First, as Cora Beard was about to explain to the jury what Tamara had told her, Keston objected on hearsay grounds. She therefore testified only that, as a result of Tamara's statements, she did not move back in with Washington and she took Tamara to the hospital. During his examination of Sergeant Elford, the prosecutor only elicited that Elford interviewed Tamara Beard, and that he spoke to the prosecutor's office after that interview. Similarly, Woodson, the social service worker who examined Tamara, testified only that she had talked with Tamara four times, that Tamara had been alone with her for three of those conversations, and that as a result of their talks, she had contacted Sergeant Elford. She said nothing of the content of their conversations.
Q. Did you tell your mother something that happened to you then?
A. Yes.
Q. Why'd you tell her?
A. `Cause I didn't want to go back there. . . .
Q. Did you talk to some nurses or maybe one nurse[?]
A. I talked to a doctor and a nurse. . . .
Q. Did you tell them what happened to you?
A. Yes.
Q. Okay. And then . . . did a lady come out to see you at your school?
A. Yes.
Q. Did she talk to you alone?
A. Yes.
Q. And did you tell her what had happened to you?
A. Yes.
Q. She asked you, didn't she?
A. Yes.
Q. Okay. And then later on there was some policemen, Sergeant Elford, this gentleman right here?
A. Yes.
Q. Did he talk to you and ask you what happened?
A. Yes.
Q. Go over it with you?
A. Yes.
Q. And then there came a time a while ago that you testified across the street over here in the District Court building before a judge, didn't you?
A. Yes.
Q. And he asked about telling the truth and you were under oath, is that right?
A. Yes.
Q. And you told the court what happened?
A. Yes.
J.A. at 162-63.
Given this testimony, we find that the State committed plain misconduct by stating that Tamara's story had not changed as she talked to these different individuals. Not only did the prosecutor improperly refer to statements not in evidence, but it is clear that the prosecutor's purpose was to enhance Tamara's credibility in the eyes of the jury. See, e.g., J.A. at 255 ("You think that a ten year old child is going to go through all of that, fool everybody, talking about two instances."). Such bolstering is also improper. Cf. United States v. Francis, 170 F.3d 546, 551 (6th Cir. 1999) (stating that improper "[b]olstering occurs when the prosecutor implies that the witness's testimony is corroborated by evidence known to the government but not known to the jury"); United States v. Duval, No. 89-1891, 1990 WL 52371, at *2 (6th Cir. April 26, 1990) (unpublished decision) (stating that improper witness vouching occurs when a prosecutor alludes to evidence outside the record as supporting the witness's testimony).
Finally, we are unmoved by the State's feeble attempt to justify its plain misconduct. The State argues in its brief that in "this case, Petitioner was charged with first degree criminal sexual conduct involving penetration. Obviously Tamara must have given a statement to someone prior to trial in which she claimed penetration." Hofbauer's Br. at 23. In a similar vein, the State contended at oral argument that the prosecutor was simply pointing out to the jury that Tamara did not "recant" her story, a position the State argues was a reasonable inference given that the State brought the prosecution. This explanation is specious for two reasons. First, this justification simply sidesteps the impropriety at issue. The prosecution did far more than merely inform the jury that Tamara "must have" stated that penetration occurred at some point, or that she did not "recant" her story. Instead, it informed the jury that Tamara's story to each and every witness had never changed, when there was in fact no evidence to that effect. This argument was a clear attempt to boost the credibility of Tamara and the believability of her story. Second, the very premise of the State's justification on appeal is flawed. Indeed, if the State had been attempting to argue the "reasonable inference" it described at oral argument and in its brief, that effort itself would have constituted gross misconduct. "[I]t is always improper for a prosecutor to suggest that a defendant is guilty merely because he is being prosecuted or has been indicted." United States v. Bess, 593 F.2d 749, 754 (6th Cir. 1979). It is equally improper to imply to a jury that an underlying factual predicate of a crime must be true due to the fact of indictment or prosecution. For these reasons, we find the State's boosting of Tamara's credibility based on facts not in evidence to constitute clear misconduct.
B.
We now must assess whether Keston's failure to object to this clear misconduct violated Strickland. We find that it did. More importantly, we find that the trial court's application of Strickland was not simply incorrect, but was objectively unreasonable.
1.
An essential ingredient of the Sixth Amendment right to counsel is that counsel provide constitutionally effective assistance. See Powell v. Alabama, 287 U.S. 45, 57 (1932). In Strickland, the Supreme Court established that the benchmark of effectiveness "must be whether counsel's conduct so undermined the proper functioning of the adversarial process that the trial cannot be relied on as having produced a just result." 466 U.S. at 686. The Strickland Court set out a two-part inquiry to determine whether a counsel's assistance is constitutional ineffective.
First, a defendant must show that counsel's performance was "deficient," involving "errors so serious that counsel was not functioning as the `counsel' guaranteed the defendant by the Sixth Amendment." Id. at 687. This requires that counsel's conduct "fell below an objective standard of reasonableness," Tucker, 181 F.3d at 754, and that counsel's "identified acts and omissions were outside the wide range of professionally competent assistance." Strickland, 466 U.S. at 690. In making this determination, a court must "indulge a strong presumption that counsel's conduct falls within the wide range of reasonable professional assistance," and that it is the defendant who "bears the burden of overcoming the presumption that the challenged action might be considered sound trial strategy." Tucker, 181 F.3d at 754 (citation omitted). Courts must not view a trial in hindsight, but must evaluate the reasonableness of counsel's performance within the context of the circumstances at the time of the alleged errors. See McQueen v. Scroggy, 99 F.3d 1302, 1311 (6th Cir. 1996). This Court has on several occasions found that a counsel's failure to object to prosecutorial misconduct constitutes defective performance when that failure is due to clear inexperience or lack of knowledge of controlling law, rather than reasonable trial strategy. See, e.g., Gravley v. Mills, 87 F.3d 779, 785-86 (6th Cir. 1996); Rachel v. Bordenkircher, 590 F.2d 200, 204 (6th Cir. 1978).
Second, even if counsel's performance is deemed deficient, a defendant must show that those deficiencies were prejudicial to the defense. See Strickland, 466 U.S. at 692. To make this showing, the defendant must demonstrate 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." Id. at 694. The essential question is "whether better lawyering would have produced a different result." McQueen, 99 F.3d at 1311 (quoting Ward v. United States, 995 F.2d 1317, 1321 (6th Cir. 1993)).
Finally, because the AEDPA applies to Washington's petition, we can only consider Washington's challenge within the more limited assessment of whether the state court's application of Strickland to the facts of this case was objectively unreasonable. See Williams, 120 S.Ct. at 1521.
2.
We find that the trial court's application of Strickland in this case was not simply incorrect, but was objectively unreasonable, meeting even the high threshold required by the AEDPA. The prosecution's tactics and challenged statements amounted to unfair and prejudicial misconduct plainly meriting an objection and curative instruction, yet Keston sat silent. At the most pivotal moments, we conclude, his silence was due to incompetence and ignorance of the law rather than as part of a reasonable trial strategy. The trial court's assessment of these events was wholly inadequate, failing to appreciate or even consider the critical issues involved.
a.
First, the trial court did not reasonably apply Strickland to the facts of this case when it determined that Keston's choice not to object to the State's improper emphasis on Washington's "bad character" was acceptable. In its written order, the court merely concluded that the prosecutor's use of the bad character evidence had not been improper because the State was merely "seek[ing] to establish a rational factual backdrop" of the crime and explain Tamara's reaction to it. But as explained supra, the record demonstrates beyond a doubt that the State's closing argument went well beyond such a narrow purpose. The court thus failed even to scratch the surface of the required inquiry — whether Keston's failure to object to the plain misconduct stemmed from a reasonable trial strategy. It most clearly did not. Indeed, his failure to object to certain statements fell below an objective standard of reasonableness and was outside the wide range of professionally competent assistance.
The court elaborated that "[t]he child-victim's fear of defendant and the beatings of the victim's mother helped explain to the jury why the victim did not report defendant's sexual abuse . . . ." J.A. at 49.
First, however, we agree with the State and the trial court that Keston's failure to object during the State's cross-examination was generally consistent with his broader trial strategy. As Keston explained at the Ginther hearing, his trial strategy was to provide the jury a basis "for disbelieving" Tamara Beard's testimony. The trial record reveals that he indeed pursued this strategy throughout the trial. Although Keston did not mention this "protection theory" in his opening argument, he clearly pursued the theory in examining witnesses. He asked Cora Beard in detail about Washington's physical abuse, his drinking, his use of drugs, his spending of Beard's money, his lack of employment, and his general lack of self-control. He also asked Tamara about Washington's drinking, and whether she had seen Washington beat Cora; when she said that she had, he asked her: "And, of course, you wouldn't want to see that happen to your mother again, would you?" She answered "No." J.A. at 170. And during his direct examination of Washington, Keston elicited from Washington that the only reason that Tamara would have made up the story "is that she wanted to protect her mother." J.A. at 200. In closing argument, counsel stated that while Tamara's account could neither "be proved or disproved," there was one explanation. J.A. at 261. "[W]e know it means Rufus Washington won't be around anymore. [Tamara] can see that. She knows not why, but he won't be around anymore, he hadn't been around for ten months." J.A. at 262. Overall, therefore, Keston's choice to allow the prosecutor to criticize Washington's character during cross-examination stemmed at least partially from trial strategy.
Keston explained:
I have to position her as being the champion for her mother; a fighter for her mother, or even in her mother's place by no longer letting her mother allow Mr. Washington to be a member of their household. . . . So I have to make him not simply a sword in the mother's side, but a real bad guy. . . . I've got to make him so bad that in the young lady's eyes, he is to be kept away no matter what length she has to go to do it.
J.A. at 292-93.
But we cannot stop there, for we must also assess if this strategy itself was constitutionally deficient. See Martin v. Rose, 744 F.2d 1245, 1249 (6th Cir. 1984) (stating that "even deliberate trial tactics may constitute ineffective assistance of counsel if they fall outside the wide range of professionally competent assistance") (internal quotations and citation omitted); see also Lovett v. Foltz, No. 88-1682, 1989 WL 101522, at *4 (6th Cir. Sept. 5, 1989) (unpublished opinion) (stating that "the label `strategy' is not a blanket justification for conduct which otherwise amounts to ineffective assistance of counsel"). In doing so, we again agree with the State that Keston's "protection" strategy was not so deficient as to constitute ineffective assistance. Certainly, the strategy walked a fine and risky line. In addition to providing a coherent and plausible explanation for why Tamara may have contrived the incident, Keston's strategy was one of perhaps few ways to "spin" the evidence of Washington's unappealing character into a potentially exculpatory use. Of course, any minimally competent attorney would also have appreciated that this strategy risked swaying the jury's determination of guilt versus innocence with improper inferences based on his "bad character." Still, due its positive potentialities, we cannot in hindsight find that Keston's overall defense strategy was so poor as to be constitutionally defective.
Nonetheless, even this articulated strategy wholly fails to explain why Keston did not object to the prosecution's most egregious character attacks during closing argument. Indeed, the risk of prejudice inherent in Keston's articulated strategy would have made a reasonably competent attorney doubly cautious about the potential misuse of that evidence. Such an attorney would have also understood that while the prosecutor was permitted to use the "bad character" evidence to explain Tamara's silence and Keston was permitted to use it to explain Tamara's allegations, the prosecutor was still clearly prohibited from using Washington's "bad character" as a basis to argue his guilt of the crime charged. Yet this is precisely the argument the State made to the jury when it urged that Washington's character "fits" the alleged crime. While a minimally competent lawyer would have recognized these statements to be blatantly improper and highly prejudicial, requiring an objection and curing instructions, Keston said nothing.
Moreover, Keston's explanation at the Ginther hearing shows that his silence arose from incompetence and ignorance of the law, rather than strategy. Failing to appreciate that the evidence of Washington's character was susceptible to some uses and not others, Keston articulated his mistaken assumption that once he had introduced evidence as part of his defense, the State could use it in any manner it desired. To object to the State's "fit" rhetoric would simply have been "inconsistent," Keston believed. This basic misunderstanding of universal trial and evidence principles falls well below an objective standard of reasonableness. See Gravley, 87 F.3d at 786 (stating that when counsel failed to object because of a lack of awareness of the law, Strickland was violated); Rachel, 590 F.2d at 204 (concluding that the Sixth Amendment was violated because attorneys' inexperience, inattention or lack of knowledge of the law led to their failure to object to misconduct). For similar reasons, we find that the state trial court's analysis of this issue was objectively unreasonable. Once again, that court concluded that the prosecutor had only sought to provide a "factual backdrop" to the crime as well as an explanation for Tamara's silence after the alleged acts, J.A. at 48-49, when it is crystal clear from the record that the State went well beyond that limited use, proffering an argument that was a prototypical example of how character evidence should not be used. Not having recognized the clear predicate problem itself, the trial court's conclusion that Keston did not violate Strickland by failing to object to that problem is thus inherently flawed. To characterize this conclusion as an "objectively reasonable" application of Strickland would be to dilute our review under the AEDPA to a generous apology for the clearest of errors.
His full explanation is as follows:
Q[of Keston]: [Y]ou wanted to go into [Washington's bad character] for a different reason than I wanted to go into it, obviously. You wanted to go into it to show that it was so bad that this young lady would have a real reason to make up something of this nature.
A:To be able to sit in here and lie under oath to a jury, it has to be serious or it can't be sold.
Q:So after I went into it in general . . ., you then on cross-examination went into it in much greater detail; did you not?
A:That is correct.
Q:And you wanted to; didn't you?
A:That is correct.
Q:And then once I used that in my final summation — And I used a lot of things that you brought out in cross-examination; didn't I?
A:I expect so.
Q:All right. So once I started using what you had actually brought out in my closing summation, you had no reason to object to that because you wanted it brought out; didn't you?
A.That's correct.
Q:You were just going to use it in a different way than I used it.
A. It would be rather inconsistent of me to object.
J.A. at 311-12 (emphasis added).
Finally, we note that while the defense's strategy carried an inherent risk of some prejudice, that added risk did not diminish the far greater prejudice that resulted from Keston's inexplicable silence as the prosecutor misused that same evidence for patently improper purposes. Our rules addressing character evidence implicitly recognize the fine yet vital distinction between the risk of prejudice borne by evidence introduced for permissible reasons and the clear prejudice that results from an uncured and flagrantly improper use of that same evidence. Thus, even if some potential prejudice arises from the introduction of certain evidence, this Court generally presumes that if properly instructed by judges and guided by counsel, juries are capable of considering evidence for one purpose but not another. See generally Richardson v. Marsh, 481 U.S. 200, 211 (1987). In this case, an objection would have prompted the judge to inform the jurors that, counter to the prosecutor's suggestion, they could not convict Washington because he was the "type" of person who would commit the alleged crime; we then would presume that the jury heeded that instruction in rendering its verdict. On the other hand, Keston's silence allowed the prosecutor's improper use of that evidence, as well as its improper suggestions to the jury of how to consider that evidence, to go uncorrected. For this reason, and because this was a close case riding on Washington's credibility, see infra, Washington was prejudiced by his counsel's failure to object to the closing argument.
b.
We also find that Keston's failure to object to the State's improper characterization of Tamara's statements to others constitutes a second instance of constitutional ineffectiveness. Again, we find the trial court's analysis of Keston's ineffectiveness and "trial strategy" to have been objectively unreasonable.
Keston's explanation of why he did not object to the prosecutor's characterization of statements not in evidence is again unconvincing. At the Ginther hearing, he explained that he feared an objection would do more harm than good because it would focus the jurors' attention on the prosecutor's statement even if the court instructed them otherwise. Rather than risking these negative consequences, his choice was not to object. Keston also stated that he did not object because the jury knew the statement was "not true because there [was] an inconsistent statement to the doctor," which he had elicited at trial. J.A. at 297.
He stated:
[A] couple avenues can present itself [sic] at the moment [the prosecutor] makes this statement. . . . [I]f I object, the jury would want to know — [t]heir attention is suddenly drawn to this factor and the fact that there are prior consistent statements are emphasized whether they are instructed to ignore it or not.
J.A. at 297. Moreover, he explained, were he to have objected, he feared that the prosecutor might have retorted: "well, if there was an inconsistency, wouldn't you have heard it? Wouldn't Mr. Keston have brought it out?" J.A. at 298.
In its brief, the State defended Keston's rationale:
Every trial attorney knows that an objection in such circumstances is a double edged sword. An objection snaps the jury to attention, focusing on the very statement the attorney believes they should not have heard. Attorneys also disagree on the value of curative instructions. . . . An attorney in the courtroom is better equipped to gauge a particular jury's ability to ignore improper information than a reviewing court.
Hofbauer's Br. at 23. This is also the argument articulated by the magistrate court. J.A. at 96 ("[C]ounsel's decision not to object was a matter of trial strategy intended to avoid drawing further attention to the prosecutor's statement.").
We find both explanations unreasonable. Keston's explanation that he was worried the objection would render damage that an instruction could not cure is clearly not viable, for two reasons. First, accepting as a proper trial strategy a lawyer's doubts over the effectiveness of objections and curative instructions would preclude ineffectiveness claims in every case such as this, no matter how outrageous the prosecutorial misconduct might be. In other words, were we to accept the State's argument, no failure to object could ever comprise ineffective assistance of counsel, and no claim of prosecutorial misconduct, however egregious, could be raised on habeas review if not objected to. Second, we must presume that juries follow their instructions. See Richardson, 481 U.S. at 211 (stating that this presumption is "rooted less in the absolute certitude that the presumption is true than in the belief that it represents a reasonable practical accommodation of the interests of the state and the defendant in the criminal justice process"). Indeed, we are excused from making this presumption only when there is a "strong likelihood that the effect of the evidence would be `devastating' to the defendant" and that there is an "overwhelming probability that the jury will be unable to follow the court's instructions." United States v. Ford, 872 F.2d 1231, 1239 (6th Cir. 1989) (citations omitted). For both of these reasons, a court reviewing a failure to object must look at factors independent of the general effectiveness of objecting, such as other possible trial strategies, the degree of the purported misconduct, or the admissibility of the evidence in question.
Under this analysis, if Keston truly worried that the impropriety was too great for an instruction to overcome, he should have objected and called for a mistrial, rather than chosen not to object at all.
Circuit cases that have stated that a decision not to object was based in part on a lawyer's desire not to focus attention on a potentially damaging statement have done so only after concluding that it was unclear that the challenged conduct itself was improper in the first place. See, e.g., Cobb, 832 F.2d at 347-48 (stating that it was not clear whether the prosecutor's conduct was improper, and that Cobb had offered no basis on appeal for why the testimony should have been excluded). In this case, the conduct was clearly improper and prejudicial, so Keston's silence cannot be so justified.
The second explanation for Keston's failure to object is simply incoherent: that the jury knew that the prosecutor's statement was not true because there was an inconsistent statement to the doctor in the record. This justification is not strategy, but absolute folly. First, it overlooks the fact that the prosecutor's statement that the "story never changed" was not based on any evidence in the record. The jury therefore had no basis to conclude whether the characterization was true or not; similarly, Keston had no way to rebut the prosecutor's assertion without also referring to conversations not in evidence. Indeed, while Keston testified to the trial court that he had "intended" to argue to the jury that Tamara's statements to the doctor were inconsistent with her stories to others, he did not in fact do so. Although he referred to the doctor's notes that she had denied penetration, Keston certainly did not emphasize that this showed that Tamara's stories had been inconsistent. In fact, Keston himself echoed the prosecution's suggestion that the content of Tamara's discussions was in evidence when he argued that all of Tamara's statements came with Cora Beard by her side. "You heard from the witness stand, she said this, said that. She always identified and repeated what was said." J.A. at 264-65. But, contrary to Keston's words, the jury had not in fact heard any witnesses testify as to what Tamara had told them.
Finally, although the trial court attempted to rationalize this "impeachment defense," its reasoning is equally flawed. The trial court's explanation reads as follows:
[Keston] related that his lack of objection was grounded on trial strategy based on his awareness that the victim had given previous inconsistent statements during the investigative stage of the case. In proper cases, a decision not to object to the prosecutor's trial efforts may be considered sound trial strategy. Cf. Darden v. Wainwright, 106 S.Ct. 2464, 2471, 2472 (1986). . . . In this case defense counsel's strategy involved impeachment with inconsistent statements and the contrasting the same with the claim that the victim's reporting of crime was consistent.
J.A. at 50. Like Keston's own words, this explanation wholly fails to appreciate that "the claim" of consistency involved statements never admitted into evidence. Moreover, while the trial court cited Darden v. Wainwright, 477 U.S. 168, 182 (1986) in support of its argument that a "decision not to object to the prosecutor's trial efforts may be considered sound trial strategy," Darden's holding provides no support for its decision. The cited portion of Darden addresses prosecutorial misconduct that the Supreme Court found insufficient to constitute a due process violation; as part of the discussion, the court noted that "defense counsel made the tactical decision not to present any witness other than petitioner." Id. at 182 (emphasis added). The decision in no way condones a lawyer's failure to object to plain misconduct as legitimate trial strategy.
In short, Keston's failure to object fell below an objective standard of reasonableness and constituted an omission outside the wide range of professionally competent assistance. Washington has shown that the failure to object was based on simple incompetence, and not on sound trial strategy. Because the trial court's conclusion merely echoed Keston's deeply flawed justification, its application of Strickland was objectively unreasonable.
3.
In addition to finding constitutionally defective performance, we also believe that the failure to object to these statements prejudiced Washington's case. As both parties agree, this trial was a credibility contest. There was no evidence in the record indicating Washington's guilt outside of Tamara's own allegations. Thus, outside of the substance of Washington's and Tamara's testimony, nothing was more important to the case than the indicia that one story was more believable than the other. In such a close case, the prosecutor's twin acts of misconduct — improperly boosting Tamara's credibility while diminishing that of Washington — very likely tipped the scales in her favor. There is thus a reasonable probability that Keston's failure to have the judge halt the ongoing misconduct and provide curative instructions affected the outcome of the trial. The fact that it had been a month since the jury had heard any witnesses only magnified the prejudicial impact of the prosecutor's misconduct and Keston's failure to object to it.
4.
Keston's failure to articulate clearly the beyond a reasonable doubt standard of proof is yet another troubling indication that his performance in this case was severely wanting. Equally troubling is his suggestion, expressed at the Ginther hearing, that there is "no difference" between the beyond a reasonable doubt and clear and convincing standards of proof. J.A. at 300. Nonetheless, the record shows that his muddled recitation of the standard of proof did not prejudice Washington's case. In his rebuttal, the prosecutor himself corrected Keston's misstatement by reiterating the reasonable doubt standard. Likewise, the trial court instructed the jury of the correct standard. Thus, although Keston's discussion of the burden of proof was confused, Washington has not shown that the argument met the "prejudice" prong of Strickland.
C.
Washington argues that the State's actions at trial constituted prosecutorial misconduct that denied him due process. While the Michigan Court of Appeals held that this claim was defaulted due to Keston's failure to object at trial, we excuse procedural default if a criminal defendant can demonstrate cause for the procedural default and actual prejudice resulting from the alleged constitutional error. See Rust v. Zent, 17 F.3d 155, 160 (6th Cir. 1994). Here, our conclusion that Keston's failure to object comprised ineffective assistance of counsel provides the required "cause." See Gravley, 87 F.3d at 785 (citing Coleman v. Thompson, 501 U.S. 722, 753-54 (1991)). We also find that for reasons stated supra — the severity of the misconduct, the lack of substantial evidence against Washington, and the length of time between the witness examinations and the closing arguments — Keston's error infected the entire trial with error of constitutional dimensions. See Rust, 17 F.3d at 161. Thus, the "cause and prejudice" test is met with respect to Washington's claim of prosecutorial misconduct. We therefore may proceed to the merits of his argument.
In examining alleged prosecutorial misconduct on habeas review, this Court can only provide relief "if the relevant misstatements were so egregious as to render the entire trial fundamentally unfair to a degree tantamount to a due process violation." Caldwell v. Russell, 181 F.3d 731, 736 (citing Donnelly, 416 U.S. at 643-45). In assessing whether the alleged misconduct amounts to a constitutional deprivation, the court must view the totality of the circumstances. See Hayton v. Egeler, 555 F.2d 599, 604 (6th Cir. 1977). We must first determine if the comments were improper. See Boyle v. Million, 201 F.3d 711, 717 (6th Cir. 2000). We then must determine if the comments were sufficiently flagrant to warrant reversal by looking to four factors: 1) the likelihood that the remarks would mislead the jury or prejudice the accused; 2) whether the remarks were isolated or extensive; 3) whether the remarks were deliberately or accidentally presented to the jury; 4) whether other evidence against the defendant was substantial. See id.; United States v. Carroll, 26 F.3d 1380, 1385 (6th Cir. 1994). Because Keston did not object to any of the statements made, plain error is required. See United States v. Blandford, 33 F.3d 685, 709 (6th Cir. 1994); United States v. Morrow, 923 F.2d 427, 432 (1991).
As explained in Part III.A, the challenged aspects of the State's closing argument were clearly improper. We also find the improprieties to have been sufficiently flagrant to satisfy the four prongs of Boyle and warrant reversal. First, as stated above, there was a strong likelihood that the improper statements would have misled the jury and prejudiced the defendant, particularly considering the long delay since the actual testimony of the witnesses. Second, the comments were extensive, comprising part of the prosecutor's continuous effort to have the jury determine credibility based on improper considerations — either statements not in evidence or improper character assessments. Third, it is clear that the remarks were deliberately made, with the prosecutor repeating his "fit" theory throughout closing argument. And fourth, there was no evidence against Washington outside of Tamara Beard's account, which the prosecutor's misconduct impermissibly bolstered. In short, the prosecutor's misconduct was sufficiently flagrant to violate Washington's due process rights.
IV.
As the people's representative in our system of justice, a prosecutor must adhere to the rules and principles that ensure that a jury determines a defendant's guilt based on the evidence before it. In a close credibility contest such as this, with horrible acts alleged but scant hard evidence for the jury to weigh, a prosecutor must be doubly careful to stay within the bounds of proper conduct. See Martin v. Parker, 11 F.3d 613, 616-17 (6th Cir. 1993) (stating that because cases involving sexual abuse "turn on the relative credibilities of the defendant and the prosecuting witness . . ., a strict adherence to the rules of evidence and appropriate prosecutorial conduct is required to insure a fair trial"). One of defense counsel's most important roles is to ensure that the prosecutor does not transgress those bounds.
In this case, both attorneys failed to perform their respective duties. We find that their failure deprived Washington of his constitutional rights, and that the state courts' conclusions to the contrary were objectively unreasonable. We therefore REVERSE the district court and grant a conditional writ of habeas corpus, giving the State of Michigan ninety days in which to provide Washington a new trial or release him.