Miller v. Terhune

ORDER KARLTON, Senior District Judge. Petitioner, a state prisoner proceeding through counsel, has filed this application for a writ of habeas corpus pursuant to 28 U.S.C. § 2254. The matter was referred to a United States Magistrate Judge pursuant to 28 U.S.C. § 636(b)(1)(B) and Local General Order No. 262. This action proceeds from petitioner’s second amended petition filed May 21, 2002. The second amended petition raised two claims: jury instruction error and ineffective assistance of counsel (“LAC”) based on the failure to present evidence of voluntary intoxication. On April 16, 2003 the magistrate judge assigned to this case recommended that the petition be denied. On September 10, 2003, this court declined to adopt the findings and recommendations with respect to petitioner’s ineffective assistance of counsel claim and remanded the matter for an evidentiary hearing. See September 10, 2003 Order. An evidentiary hearing was held on November 8, 2004 and November 23, 2004. On November 7, 2006, the magistrate judge filed findings and recommendations which were served on the parties. The magistrate judge recommended that the petition be denied. Petitioner filed timely objections. In accordance with the provisions of 28 U.S.C. § 636(b)(1)(C) and Local Rule 72-304, the court has conducted a de novo *489 review of this case. Having carefully reviewed the file, the court declines to adopt the magistrate judge’s findings and recommendations and makes the following determination based on the record. I. BACKGROUND A. Facts of the Case Petitioner was convicted of second degree murder in the 1997 shooting death of his friend, Steven Faddis. The court adopts the factual summary as set forth in the opinion of the California Court of Appeal: The defendant and Steven Faddis were long time friends. On September 25, 1996, the defendant hosted a gathering at his house, celebrating Faddis’ release from prison. In attendance were Fad-dis, the defendant, and others. Kassy Goold, Faddis’ girlfriend arrived around 4 p.m. She and Faddis were in the process of breaking off their relationship. Everyone was drinking. Everyone except Goold, Faddis, and the defendant left around 8 p.m. Later that night, Goold and the defendant were in the livingroom discussing Goold’s breakup with Faddis. Goold told the defendant she was breaking up with Faddis because her family, in particular her brother, disapproved of the relationship. Faddis threatened to kill Goold’s brother. He shoved Goold onto the couch with his finger, then grabbed her by the hair and slammed her through the glass coffee table. The defendant intervened. During the defendant’s trial testimony, he claimed Faddis punched him and then left the premises through the front door. According to the defendant’s testimony, he grabbed the rifle and some ammunition and followed Faddis out the front door. Although the defendant did not see or hear anyone other than Faddis outside, he believed Goold was outside because he saw something underneath the truck. In fact, Goold was hiding inside the house. The defendant walked down the steps of his porch and yelled at Faddis to get off his property. Faddis yelled and came towards him. The defendant fired a shot at Faddis’ shoulder but missed. Faddis continued to walk toward the defendant and threatened him. The defendant then aimed at the left side of Faddis’ chest and shot again. The defendant claimed he was only trying to disable Faddis, but was not trying to kill him. After shooting Faddis, the defendant called 911 and began administering CPR until medical personnel arrived. The defense theory was self-defense. Evidence was presented that Faddis was exceptionally muscular and violent and had a history of abusing women. His ex-wife testified he had given her broken ribs, a collapsed lung, and head injuries during the course of their relationship. Answer, Ex. B, pp. 2-3. The parties do not dispute that petitioner’s trial counsel stipulated that his blood alcohol level was .30 approximately three hours after the shooting. Reporter’s Transcript (“RT”) 217. The jury was instructed on both second degree murder and voluntary manslaughter. A jury instruction on the lesser included offense of involuntary manslaughter was not given. Petitioner was convicted of second degree murder and his conviction was affirmed by the California Court of Appeal for the Third District on August 24, 1999 (Blease, J., dissenting). The only claim before the court is petitioner’s ineffective assistance of counsel (“LAC”) claim. Petitioner argues that his trial counsel performed unreasonably and prejudicially in failing to investigate and *490 develop evidence of petitioner’s extreme intoxication in order to persuade the jury that he did not have the requisite intent to murder Steve Faddis. Petitioner first raised his failure to investigate allegations in his habeas petition to the California Supreme Court, see Respondent’s Answer, Exh. E, at 10-11, which was rejected by postcard denial on April 17, 2002. See Resp’ts Answer, Exh. F. B. Summary of Evidentiary Hearing An evidentiary hearing on petitioner’s IAC claim was held on November 8, 2004 and November 23, 2004. What follows is a brief overview of the evidence presented. 1 1. Dr. Gregory Sokolov 2 Dr. Gregory Sokolov, M.D., testified as an expert as to alcohol’s effect on the brain and related mental states. Dr. Sokolov testified that a blood alcohol level (“BAC”) of .30 is clinically significant. Evidentiary Hearing Transcript (“EH”) at 20. With this BAC, it is physiologically possible for a person to be walking and talking, yet suffer significant cognitive impairment. EH 21. A person’s ability to reason, choose a course of action, and formulate a plan would be affected in significant ways. EH 24. A person’s ability to reason — ■ including understanding cause and effect, the ability to draw inferences from facts, and anticipation of likely consequences of actions — would also be significantly affected. EH 24. These are all executive brain functions. EH 24-25. Dr. Sokolov testified that petitioner’s BAC of .30 at 2:25 a.m. indicated a likely BAC between .33 and .34 at the time of the shooting. EH 31. This qualified as “extreme alcohol intoxication,” approaching the level at which coma and death can occur. EH 32-33. Dr. Sokolov also testified that there is not always a correlation between the degree of physical impairment and the degree of mental impairment. EH 26. In a person who is highly alcohol tolerant, there may be a disconnect between the degree of visible physical impairments and the degree of mental impairment. EH 30. As applied to petitioner, Dr. Sokolov testified that although petitioner was able to speak, walk and physically function, it does not follow that his mental functioning was similarly intact. EH 33, 72. Petitioner’s level of intoxication significantly affected his perceptions of sights and sounds, his ability to interpret data, and his general ability to accurately understand what was going on around him. EH 33-35. Dr. Sokolov also testified that petitioner’s ability to weigh choices and anticipate the consequences of his actions was also significantly impaired. EH 34. It was Dr. Sokolov’s opinion that petitioner’s intoxication precluded him from forming an intent to kill. EH 39. Dr. Sokolov reasoned that with petitioner’s BAC, petitioner was unable to accurately perceive whether his life or Ms. Goold’s life was in danger and was also unable to cognitively process that a shot to the chest could be lethal. EH 37. He also concluded that petitioner’s ability to appreciate the cause and effect relationship (between shooting and killing) was significantly impaired. EH 38. Dr. Sokolov believed that *491 petitioner’s statement that he hadn’t meant to kill Faddis was consistent with his state of inebriation. EH 39. Dr. Sokolov testified in detail about how the cognitive effects of intoxication explain, medically, how petitioner could not have intended to kill despite circumstances which would support an inference of such intent in a sober person. EH 39-41. Accordingly, it is credible that petitioner could have taken actions facially suggestive of specific intent to kill, without actually intending anything more than to “stop” a situation out of control. Id. Dr. Sokolov also testified that had he been contacted by petitioner’s trial counsel and had reviewed the same material (police reports, petitioner’s statements, blood alcohol test results, etc.), he would have opined that petitioner acted without specific intent to kill, without realizing that his actions could kill Faddis, and with the subjective belief that he was acting to save his own life and/or that of Ms. Goold. EH 47-47. 2. Testimony of Mark Cibula, Esq. Mr. Cibula represented petitioner at trial. At the time, he had been a lawyer for less than two years and had never represented a client charged with murder. EH 85-88. He had also never used an expert on the issue of intent or mental state. EH 88. He had handled several drunk driving cases, and was familiar with the type of blood alcohol evidence typically at issue in that context. EH 98-99. Mr. Cibula remembered that petitioner confessed to the shooting and that he stated that he had not meant to kill Faddis. EH 95. He also remembered that petitioner asserted that he was acting in self-defense and that petitioner was intoxicated. EH 96-97. Mr. Cibula recalls receiving a toxicology report which showed petitioner’s BAC. EH 97. Petitioner informed his attorneys that he wanted to pursue a theory of self-defense. EH 102. Petitioner was adamant that he had not intended to kill Faddis. EH 103. Mr. Cibula wrote the motion to set aside the charges pursuant to California Penal Code § 995, based on the preliminary hearing testimony. EH 104-015. In addition to self-defense, the following theories were asserted in the motion: sudden quarrel or heat of passion; provocation; imperfect self-defense; and intoxication. Id. Mr. Cibula did not recall when he and his co-counsel decided to abandon the voluntary intoxication defense. EH 106. They did not request a jury instruction on voluntary intoxication. EH 109. Prior to trial, Mr. Cibula and his co-counsel met with a psychologist, Dr. Marilyn Wooley. EH 119. Cibula did not recall any of the conversation, but assumes that they would have discussed blood alcohol issues. EH 110, 123-124. Dr. Wooley did not identify any mental state issues. EH 121. Mr. Cibula did not ask Dr. Woo-ley for a referral to someone who specialized in alcohol-related issues, EH 134, and did not utilize available DUI-related resources to locate an alcohol expert. Id. The only expert witness presented by the defense at trial was a pathologist, Dr. Sharon Van Meter, who testified regarding bullet trajectory issues. RT 390-421. Mr. Cibula testified that he did not recall the specific reasons that he had for not presenting evidence of petitioner’s blood alcohol level. EH 131. He did recall clearly that petitioner was adamant about having acted in self-defense. EH 101, 126, 127, 129, 131. Mr. Cibula also recalls feeling that emphasizing petitioner’s intoxication may have been inconsistent with a theory of self-defense. EH 145-146. 3. Testimony of Ronald Mclver, Esq. Mr. Mclver was co-counsel with Mr. Ci-bula at petitioner’s trial. He had tried *492 three prior homicide cases, one of which was a capital case. EH 174. He had never retained an expert on how blood alcohol levels affect a client’s state of mind. EH 175. Although Mr. Mclver was more experienced than Mr. Cibula, he did not consider himself to be “senior” or “lead” counsel. EH 175-176. Mr. Mclver would have recognized that at the time of the petitioner’s case, a .30 BAC was a “high blood alcohol level.” EH 180. He was aware of petitioner’s BAC test results. Id. Mr. Mclver did not recall when or why he and Mr. Cibula decided to not pursue a defense of voluntary intoxication. EH 184. He also did not recall stipulating to the petitioner’s blood alcohol level, or discussing with Mr. Cibula the pros and cons of such a stipulation. EH 193-194. He also remembered consulting Dr. Wooley, but cannot remember the substance of their discussion. He assumes they discussed the issue of intoxication. EH 184. Mr. Mclver did not recall thinking about, or discussing with Mr. Cibula, how evidence of intoxication might affect the jury’s reasoning as to malice and intent. EH 186. He also did not remember discussing how the jury might use intoxication evidence in assessing reasonableness in the self-defense context. EH 187. Mr. Mclver did not consult a medical doctor or forensic toxicologist regarding the effects of alcohol on petitioner. EH 185. Nor did either counsel consult any mental health professional whose expertise was in the area of alcohol abuse or intoxication. Id. Nor was any expert consulted on the question of whether a person’s blood alcohol level would necessarily affect physical and cognitive functioning to the same degree. EH 206-207. Finally, counsel never consulted any scientific literature, practice materials or other sources of information on the issue of alcohol’s effect on the formation of intent. EH 207-208. 4. Testimony of Dr. Marilyn Wooley 3 Dr. Wooley has no memory of consulting on this case. She has no files or materials related to the case. Dr. Wooley’s areas of expertise are child sexual and physical abuse, spousal abuse, child custody, eating disorders and trauma-related stress management. In her declaration she stated: I was not a substance abuse specialist at the time of the [case] meeting. As among clinical psychologists and comparable mental health professionals, I was not at the time an expert in the particular effects of alcohol on the brain and its functioning. II. Standard of Review Petitioner’s federal petition was filed after the effective date of the Anti-Terrorism and Effective Death Penalty Act (“AEDPA”), and accordingly, the standard of review set forth in the AEDPA applies in the instant ease. Generally, the AED-PA mandates that federal courts defer to the “state court’s determination of the federal issues unless that determination is contrary to, or involved an unreasonable application of, clearly established Federal law.” Pham v. Terhune, 400 F.3d 740, 742 (9th Cir.2005) (citations omitted). “For claims for which no adjudication on the merits in state court was possible ... AEDPA’s standard of review does not apply.” Killian v. Poole, 282 F.3d 1204, 1208 (9th Cir.2002). This includes claims such as petitioner’s IAC claim, in *493 which the “state courts could not have made a proper determination on the merits” because the evidence upon which adjudication must be based was adduced for the first time at a federal evidentiary hearing. Id. Here, as in Killian, there is no state court factual finding to which deference might apply under 28 U.S.C. §§ 2254(d)(2) or (e)(1). There is also no state court decision addressing questions of law, which would subject the claim to a reasonableness review pursuant to section 2253(d)(1) of AEDPA. Kesser v. Cambra, 392 F.3d 327, 334-35 (9th Cir.2004) (reversed on other grounds by Kesser v. Cambra, 465 F.3d 351 (9th Cir.2006)). In short, none of the limitations on relief set forth in § 2254 apply to this court’s review of petitioner’s IAC claim regarding the failure to investigate and develop evidence of intoxication. III. ANALYSIS A. Exhaustion As a threshold matter, the magistrate judge concluded that during the evidentia-ry hearing and in post-hearing briefing, “petitioner raised a new ineffective assistance of counsel claim that is not exhausted.” See Findings and Recommendations (“F & Rs”) filed November 2, 2006 at 6:16. As this court reads the second amended petition, as well as the F & Rs issued on April 16, 2003, and this court’s remand order, petitioner is not seeking to raise new, unexhausted claims, rather, he is making new arguments with respect to his original IAC claim. As previously mentioned, petitioner first raised this IAC claim in his habeas petition te the California Supreme Court, see Respondent’s Answer, Exh. E, at 10-11, which was rejected by postcard denial. See Respondent’s Answer, Exh. F. He asserted a violation of his Sixth Amendment right to effective assistance of counsel on the grounds that his counsel failed to investigate, develop and present voluntary intoxication evidence. The petitioner is not now alleging new claims, rather, he is presenting additional theories encompassed within his original IAC claim. Moreover, both the magistrate judge and this court have, on previous occasions, recognized the scope of petitioner’s IAC claim. In the first findings and recommendations, filed on April 16, 2003, the magistrate judge discussed the two claims presented by petitioner, the jury instruction error claim and the IAC claim. As explained by the magistrate judge, the essence of the IAC claim involved several allegations, namely, that counsel was ineffective for failing to investigate, develop and present evidence of intoxication on the issue of intent. The magistrate judge did not treat the separate allegations as separate claims. 4 Instead, the magistrate judge discussed the various components of the IAC claim, such as the failure to request an involuntary manslaughter instruction, and the failure to investigate and present a defense of voluntary intoxication. See F & Rs filed April 16, 2003 at 15. The magistrate judge did not parse out each allegation and classify them as separate claims. There is no reason to do so now. This court’s remand order also recognized that the IAC claim contained several allegations. The remand order provided in part, “[pjetitioner alleges that his counsel was ineffective for, inter alia, failing to *494 investigate and present a defense based on voluntary intoxication.” September 10, 2003 Order at 2:8-10 (emphasis added). The use of the words “inter alia” suggests that there were several allegations within the IAC claim. In remanding the case for an evidentiary hearing, this court did not intend for its order to be read as limiting the arguments that petitioner could raise regarding potential uses of intoxication evidence. Rather, the court’s remand order specifically called for an evidentiary hearing so that new facts could be ascertained. The court’s remand order provided: [B]y claiming that trial counsel failed to investigate the effects of petitioner’s high blood alcohol level, petitioner raises the possibility that there are facts not in the record that might show that a voluntary intoxication defense was viable— perhaps even more viable than an ordinary self-defense theory. September 10, 2003 Order at 3:16-21. The matter was referred back with the express purpose allowing petitioner to “develop ... facts not present in the record.” Id. at 4:7. By allowing petitioner the opportunity to develop new facts, it follows that the court was also allowing petitioner the opportunity to develop new arguments based on those facts. In short, the court finds that the magistrate judge has misconstrued the scope of petitioner’s IAC claim. Petitioner raised the same IAC claim before the state court and this court, namely, that counsel acted unreasonably and prejudicially in failing to investigate, develop and present intoxication evidence. Nothing in the evidentia-ry hearing and related briefing suggests that petitioner is now trying to impermissi-bly expand that claim to encompass other claims for relief. Rather, petitioner used the facts obtained at the evidentiary hearing to raise additional arguments to support his IAC claim. Petitioner’s supplemental arguments about the possible use of intoxication evidence are new theories about the evidence and are not new theories of constitutional injury. It is well settled that new factual allegations do not render a claim unex-hausted unless they “fundamentally alter the legal claim already considered by the state courts.” Vasquez v. Hillery, 474 U.S. 254, 258-260, 106 S.Ct. 617, 88 L.Ed.2d 598 (1986) (finding that supplemental extra-record evidence does not require further exhaustion when alleged constitutional violation is unchanged). Here, petitioner’s core IAC claim has not changed. Accordingly, there is no exhaustion problem. B. The Merits of Petitioner’s IAC claim The Sixth Amendment guarantees criminal defendants the right to effective assistance of counsel. Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). To prevail on a claim of ineffective assistance of counsel, petitioner must show: 1) his attorney’s performance was unreasonable under prevailing professional standards; and 2) there is a reasonable probability that but for counsel’s unprofessional errors, the results would have been different. Id. at 692, 104 S.Ct. 2052. Strickland defines a reasonable probability as ‘a probability sufficient to undermine confidence in the outcome.’ Id. For the reasons discussed herein, the court declines to adopt the recommendation that petitioner’s IAC claim be denied. 1. The “Unconsciousness Defense” As an initial matter, the magistrate judge limits his discussion of the IAC claim to the “unconsciousness defense.” As the magistrate judge explains, “to succeed on [his IAC] claim, petitioner must demonstrate that counsel acted unreason *495 ably in failing to pursue the unconsciousness defense.” Nov. 2, 2006 F & Rs at 11:14-15. The magistrate judge concludes that petitioner’s level of impairment did not rise to the level of unconsciousness and therefore, it was reasonable for trial counsel to not argue intoxication. “The total picture simply does not permit a finding of legal unconsciousness.” Id. at 21:22-23. This conclusion misconstrues both the law with respect to intoxication and the evidence ascertained at the evidentiary hearing. I begin by noting that the issue for the jury was not whether petitioner was “unconscious,” but whether he harbored malice and/or intended to kill. Under California law, evidence that a defendant was unconscious is sufficient but not necessary to negate intent: The critical factor in distinguishing the degrees of a homicide is thus the perpetrator’s mental state. If a diminished capacity renders him incapable of entertaining either malice or an intent to kill, then his offense is mitigated to a lesser crime. Although a finding that the perpetrator was unconscious would establish the ultimate facts that the perpetrator lacked both the ability to entertain malice and an intent to kill, the absence of either or both of such may nevertheless be found even though the perpetrator’s mental state had not deteriorated into unconsciousness. People v. Ray, 14 Cal.3d 20, 28, 120 Cal. Rptr. 377, 533 P.2d 1017 (1975). 5 At the time of petitioner’s trial, neither the statute governing voluntary intoxication evidence nor the CALJIC jury instructions suggest that a defendant must be rendered “unconscious” in order to negate intent. Voluntary intoxication is not a true defense. Rather, intoxication is a species of mental state evidence which, in a homicide case, can show that the defendant “did not in fact form the intent unlawfully to kill (i.e., did not have malice aforethought).” People v. Saille, 54 Cal.3d 1103, 1117, 2 Cal.Rptr.2d 364, 820 P.2d 588 (1991) (emphasis in original) (citation omitted). The magistrate judge relies on People v. Ochoa, in which the California Supreme Court stated that “when a person renders himself or herself unconscious through voluntary intoxication and kills in that state, the killing is attributed to his or her negligence in self-intoxicating to that point, and is treated as involuntary manslaughter.” People v. Ochoa, 19 Cal.4th 353, 423, 79 Cal.Rptr.2d 408, 966 P.2d 442 (1998). The case does not, however, hold that establishing unconsciousness is a necessary prerequisite to negating intent. Whether or not petitioner was unconscious has no bearing on the likely success of a defense based on intoxication. Complete impairment is not needed for alcohol to render a person incapable of forming the requisite mens rea. Second, Dr. Sokolov, petitioner’s expert on the effects of intoxication, testified that petitioner’s ability to physically function did not therefore mean that his cognitive abilities were similarly intact. See EH 33, 72. As petitioner explains: The diverging levels of physical and cognitive impairment that Mr. Miller exhibited on the night of the shooting are to some degree, petitioner acknowledges, counter-intuitive. That is precisely why expert testimony was needed at the trial. Without education about the effects of alcohol on the brain, the jury was all too willing to follow the prosecutor’s *496 lead in attributing to Mr. Miller the cognitive processes of a sober person. Pet’rs post-hearing br. at 41. As discussed infra, Dr. Sokolov testified extensively about how the state of petitioner’s memory was not inconsistent with intoxication. It cannot be said that simply because petitioner could recollect certain parts of the night, that he was sober enough to form intent. Similarly, just because petitioner was able to physically function does not also mean that petitioner was able to cognitively function. As petitioner suggests, these inconsistencies are exactly why expert testimony would have been helpful. For these reasons, the court cannot adopt the recommendation that since petitioner’s level of impairment did not rise to the level of unconsciousness, it was reasonable for trial counsel to not investigate petitioner’s level of intoxication. 2. Relevánt California Law at Time of Petitioner’s Trial The magistrate judge concluded that “counsel could not have been ineffective for not urging voluntary intoxication as a form of voluntary manslaughter because at the time of his conviction such was not legally possible.” F & Rs filed Nov. 2, 2006 at 7:16-18. The magistrate judge explained that: “voluntary intoxication could not reduce murder to voluntary manslaughter, as both require an intent to kill at the time; but if intoxication could negate that state of mind, the resulting crime would be, at most, involuntary manslaughter.” Id. at 8:17-20 (citing to People v. Saille, 54 Cal.3d 1103, 1116, 2 Cal.Rptr.2d 364, 820 P.2d 588 (1991)). I cannot agree. Petitioner 'is not arguing that the only cognizable theory of IAC is one predicated on the possibility of an involuntary manslaughter conviction. 6 Petitioner concedes that involuntary manslaughter should have been the primary theory at trial. Evidence of intoxication could have negated malice and resulted in a verdict no greater than involuntary manslaughter. 7 Petitioner argues that as to acquittal, California law at the time of his conviction was clear: “if this evidence [of intoxication sufficient to negate malice] is believed, the only supportable verdict would be involuntary manslaughter or an acquittal. ” Saille, 54 Cal.3d at 1116-1117, 2 Cal.Rptr.2d 364, 820 P.2d 588 (emphasis added). The Saille court explained: A defendant ... is ... free to show that because of his mental illness or voluntary intoxication, he did not in fact form the intent unlawfully to kill (i.e., did not have malice aforethought). In a murder case, if this evidence is believed, the only supportable verdict would be involuntary manslaughter or an acquittal. If such a showing gives rise to a reasonable doubt, the killing (assuming there is no implied malice) can be no greater than involuntary manslaughter. Saille, 54 Cal.3d at 1117, 2 Cal.Rptr.2d 364, 820 P.2d 588 (internal citations omitted). In short, the law at the time of petitioner’s trial was settled: evidence of voluntary intoxication in a homicide case could be admitted to show that the defen *497 dant did not form the intent to kill. Saille, at 1117, 2 Cal.Rptr.2d 364, 820 P.2d 588. 8 Moreover, evidence of intoxication would also be relevant to a defense theory of voluntary manslaughter based on imperfect self-defense. Even if the jury did not find that petitioner’s intoxication negated the formation of intent, the evidence of cognitive impairment could have lead to a finding that petitioner genuinely perceived Faddis to pose an imminent life-threatening danger. To be exculpated on a theory of imperfect self-defense, a trier of fact must find that the defendant killed another person because the defendant actually— but unreasonably — believed he was in imminent danger of death or great bodily injury. People v. Flannel, 25 Cal.3d 668, 674, 160 Cal.Rptr. 84, 603 P.2d 1 (1979). The defendant is deemed to have acted without malice and thus can be convicted of no crime greater than voluntary manslaughter. 9 In re Christian S., 7 Cal.4th 768, 782, 30 Cal.Rptr.2d 33, 872 P.2d 574 (1994). In sum, at the time of petitioner’s trial, evidence of intoxication would have been admissible and relevant to defense theories based both on involuntary and voluntary manslaughter. 3. Whether Trial Counsel’s Performance was Deficient The central issue before the court is whether petitioner’s trial counsel rendered ineffective assistance when they failed to investigate and present evidence as to how petitioner’s level of intoxication likely affected his perceptions, intentions and actions on the night of the shooting. Under Strickland, the court determines first whether trial counsel’s performance was deficient. Strickland, 466 U.S. at 687, 104 S.Ct. 2052; Harris v. Wood, 64 F.3d 1432, 1435 (9th Cir.1995). A counsel’s performance is deficient if, considering all the circumstances, it falls below an objective standard of reasonableness measured under prevailing professional norms. Id. Even if counsel’s decision “could be considered one of strategy, that does not render it immune from attack—it must be a reasonable strategy.” Jones v. Wood, 114 F.3d 1002, 1010 (9th Cir.1997). a. Failure to Investigate & Develop Evidence of Intoxication While there is a “strong presumption that counsel’s conduct falls within the wide range of reasonable professional assistance,” and “[jjudicial scrutiny of counsel’s performance must be highly deferential,” Strickland at 689, 104 S.Ct. 2052, defense counsel must, “at a minimum, conduct a reasonable investigation enabling him to make informed decisions about how best to represent his client,” Sanders v. Ratelle, 21 F.3d 1446, 1456 (9th Cir.1994) (emphasis in original). Before a decision *498 can be considered strategic under Strickland, counsel must conduct a “thorough investigation.” Williams v. Taylor, 529 U.S. 362, 396, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000). With respect to counsel’s failure to investigate, the magistrate judge concluded: Counsel had at least explored the possibility of presenting an intoxication theory to the jury by visiting a mental status expert [Dr. Wooley]. While not a forensic alcohol expert herself, this person certainly knew enough about alcohol impairment to direct counsel to other witnesses if she thought the facts required. F & Rs filed Nov. 2, 2006 at 22. While it may be that bad facts are bad facts, the record from the evidentiary hearing does not support the conclusion that the trial counsel investigated the facts surrounding petitioner’s intoxication. At the evidentiary hearing, both of petitioner’s trial attorneys testified that they were aware of the fact that petitioner’s blood alcohol level was .30 three hours after the shooting occurred. Nonetheless, as discussed supra, neither counsel consulted a medical doctor or forensic toxicologist regarding the effects of alcohol on petitioner. EH 185. Counsel did not consult any mental health professional whose expertise was in the area of alcohol abuse or intoxication. Id. Nor was any expert consulted on the question whether a person’s blood alcohol level would necessarily affect physical and cognitive functioning to the same degree. EH 206-207. Finally, counsel never consulted any scientific literature, practice material or other sources on the issue of alcohol’s effect on the formation of intent. EH 207-208. This testimony is undisputed by respondent. Contrary to the finding of the magistrate judge, counsel’s consultation with Dr. Wooley does not constitute “investigation” as required by Strickland and related case law. It is undisputed that Dr. Wooley was consulted for a casual, one-hour brain storming session on possible psychological issues in the case. EH 119, PX 663, 664. Mr. Cibula testified that he did not recall any of the substance of the conversation, but assumes that they would have discussed the case broadly, including blood alcohol issues. He did not recall anything that Dr. Wooley might have said specifically about intoxication. EH 122. Dr. Wooley was also unable to recall any opinion she rendered in the case. She stated in her declaration that she has no specialized training in the effects of alcohol on cognitive processes. At the time, Dr. Wooley’s areas of expertise were child sexual and physical abuse, spousal abuse, child custody, eating disorders and trauma-related stress management. In light of this testimony, the court cannot agree with the magistrate judge that the consultation with Dr. Wooley constitutes exploring the possibility of presenting an intoxication defense. Neither counsel nor Dr. Wooley remember what issues were discussed, and Dr. Wooley conceded that she was not an expert on how alcohol affects the brain, nor an expert on substance abuse. Accordingly, counsel’s “consultation” with Dr. Wooley—who admits she is not qualified to render an opinion on intoxication—cannot be considered investigation upon which to rest a strategic decision. See Caro v. Calderon, 165 F.3d 1223, 1226-27 (9th Cir.1999) (finding that counsel performed unreasonably in failing to investigate impact of brain damage due to pesticide exposure, having consulted a medical doctor, psychiatrist, and psychologist about the case but not a neurologist or toxicologist). Nor can counsel’s consultation with Dr. Van Meter be considered investigation into petitioner’s blood alcohol level. Dr. Van *499 Meter’s testimony was limited to bullet trajectory evidence and counsel did not question Dr. Van Meter, either before trial or on the stand, about intoxication. RT 390-425, EH 191. In sum, the record reflects that counsel failed to investigate the effects of intoxication on petitioner. Accordingly, counsel was in no position to make a reasoned or strategic decision regarding the use of intoxication evidence. It is well settled that under Strickland, “attorneys have considerable latitude to make strategic decisions about what investigations to conduct once they have gathered sufficient evidence upon which to base their tactical choices.” Jennings v. Woodford, 290 F.3d 1006, 1014 (9th Cir.2002). In the instant case, there is simply no indication that defense counsel gathered any evidence upon which to base their decision to not investigate or present evidence of intoxication. See Williams, 529 U.S. at 369, 120 S.Ct. 1495(counsel must conduct a “thorough investigation” before decision can be considered strategic under Strickland); Sanders, 21 F.3d at 1457 (citing United States v. Gray, 878 F.2d 702, 711 (3rd Cir.1989)) (finding that “... [cjounsel can hardly be said to have made strategic choice when he has .not obtained the facts on which a decision could be made.”). The ABA Standards for Criminal Justice provide guidance as to the obligations of criminal defense attorneys in conducting an investigation. Rompilla v. Beard, 545 U.S. 374, 125 S.Ct. 2456, 162 L.Ed.2d 360 (2005); Williams, 529 U.S. at 396, 120 S.Ct. 1495. The standards in effect at the time of petitioner’s trial clearly described the defense lawyer’s duty to investigate: (a) Defense counsel should conduct a prompt investigation of the circumstances of the case and explore all avenues leading to facts relevant to the merits of the case and the penalty in the event of conviction. The investigation should include efforts to secure information in the possession of the prosecution and law enforcement authorities. The duty to investigate exists regardless of the accused’s admissions or statements to defense counsel of facts constituting guilt or the accused’s stated desire to plead guilty. ABA Standards for Criminal Justice, Defense Functions, Standard 4-4.1 (3d Ed.). When trial counsel is on notice that his client may have a particular mental impairment relevant to the case, the Ninth Circuit has repeatedly held that failure to investigate the mental state constitutes deficient performance under Strickland. See, e.g., Douglas v. Woodford, 316 F.3d 1079, 1085 (9th Cir.2003) (citing Bean v. Calderon, 163 F.3d 1073, 1078 (9th Cir.1998) (holding that “[tjrial counsel has a duty to investigate a defendant’s mental state if there is evidence to suggest that the defendant is impaired.”); see also Caro v. Woodford, 280 F.3d 1247, 1254 (9th Cir.2002); Hendricks v. Calderon, 70 F.3d 1032, 1043 (9th Cir.1995)). In such circumstances, counsel must undertake at least “a minimal investigation in order to make an informed decision regarding the possibility of a defense based on mental health.” Seidel v. Merkle, 146 F.3d 750, 756 (9th Cir.1998) A preliminary psychological assessment does not constitute investigation into mental state defenses sufficient to support a reasonable decision not to pursue the possible defense further. Jennings, 290 F.3d at 1013. In Jennings, the petitioner took methamphetamine and consumed alcohol on the night of the murder. The court concluded that trial counsel was ineffective for failing to investigate mental health and drug abuse issues: Because Mr. Jennings’ alibi defense was weak and uncorroborated, and given the *500 wealth of mental health and drug abuse evidence at the ready, effective counsel almost certainly would have made an effort to raise reasonable doubt as to Mr. Jennings’ intent and his ability to undertake a “willful, deliberate, and premeditated killing” and his ability to act with “malice.” Jennings, 290 F.3d at 1019. In the instant case, petitioner’s trial counsel failed to even obtain a preliminary psychological assessment much less conduct a more meaningful investigation into petitioner’s intoxication the night of the crime. See also Rios v. Rocha, 299 F.3d 796 (9th Cir.2002) (finding ineffective assistance where counsel made choice of defense theories based on police reports and one psychological report, without investigating whether there was factual basis for alternative theory); People v. Mozingo, 34 Cal.3d 926, 196 Cal.Rptr. 212, 671 P.2d 363 (1983) (noting that a possible conflict between mental state defense and alibi defense does not excuse counsel’s failure to initially investigate the potential strengths of a mental state defense). Respondent’s strongest argument, one which the magistrate judge agreed with, is that trial counsel, faced with pursuing two seemingly inconsistent theories of defense, made a strategic decision to not present evidence of intoxication. See F & Rs filed November 2, 2006 at 22:11-15. At the evidentiary hearing, Mr. Cibula testified that he was worried about “making two different sorts of arguments to the jury.” EH 124. Mr. Cibula was concerned that a voluntary intoxication theory would have been inconsistent with a theory of self-defense. Id. Based on this testimony, the magistrate judge concluded that trial counsel had to choose “one of two problematic defenses,” and that “bad facts are bad facts.” F & Rs filed November 2, 2006 at 22:11-15. While initially appealing, respondent’s argument is ultimately without merit. First, even if there were two inconsistent theories of defense, a matter far from clear, the choice between the two theories was made without adequate investigation. That there were two possibly conflicting theories does not excuse the failure to investigate. As previously discussed, choosing a defense strategy without prior investigation into the alternative theories constitutes unreasonable performance on the part of the trial attorneys. See Jennings, 290 F.3d at 1015-1016 (ineffective assistance where counsel chose an alibi defense without first investigating mental health and drug use effects on intent); Rios, 299 F.3d at 796. Second, presenting evidence of intoxication at trial would operate to negate malice and there is nothing inconsistent in arguing that petitioner lacked the requisite malice and that he was acting in self-defense. It is undisputed that petitioner’s BAC was extremely high and that he had told police responding to the scene that he hadn’t meant to kill Faddis. Presenting evidence of intoxication would have provided context for the jury to understand how petitioner may have not had the requisite intent and/or falsely believed that he needed to defend himself. Expert testimony would have helped explain how someone with a blood alcohol level of .30 would respond in a situation in which they felt threatened. While petitioner’s physical abilities may not have been impaired, as Dr. Sokolov testified, it is possible that petitioner’s cognitive abilities were impaired and thus, negated his intent to kill. In support of its position, respondent relies on Turk v. White, in which the court rejected a failure to investigate claim. 116 F.3d 1264 (9th Cir.1997), cert. denied, 522 U.S. 1125, 118 S.Ct. 1073, 140 L.Ed.2d 132 (1998). The court explained that defense *501 counsel does not have an obligation to pursue an alternative, conflicting defense once he reasonably selects the defense to present at trial. Turk, at 1267. The key point in Turk, and what makes it inapplicable to the case at bar, is that counsel reasonably selected the self-defense theory before making the decision to not investigate an alternative theory. Here, the decision to not investigate petitioner’s intoxication was unreasonable precisely because counsel failed to conduct any preliminary investigation. See Phillips v. Woodford, 267 F.3d 966, 980 (9th Cir.2001) (holding that “defense counsel did not reasonably select the alibi defense used at trial ... [because it] was not selected on the basis of a reasonable investigation or strategic decision.”) Third, respondent also argues that counsel’s decision to not investigate intoxication was reasonable in light of petitioner’s adamant desire about wanting to present a self-defense theory. EH 101-102. According to trial counsel, petitioner was not interested in a theory of defense that would reduce his culpability to manslaughter, “he wanted a perfect self-defense.” EH 148. It is well established that “going against the wishes” of a client is an “unreasonable basis not to investigate ...” Avila v. Galaza, 297 F.3d 911, 921 (9th Cir.2002). Even when a defendant insists that he is innocent, counsel has a duty to investigate possible mental state defenses. See Jennings, 290 F.3d at 1011. Finally, respondent contends that the evidence available to counsel indicated that petitioner was acting in ways inconsistent with being intoxicated. For example, petitioner performed CPR on Faddis, called 911, and directed officers at the scene. It is respondent’s position that “[a]ll these facts which must have been known to counsel at the very early stages of their investigation were wholly inconsistent with someone who is so intoxicated that they did not actually intend to kill.” Resp’t Br. at 16. As discussed supra, had counsel actually investigated the effects of intoxication on petitioner, they would have found that these actions were not necessarily inconsistent with petitioner being intoxicated. Every fact that respondent cites in this regard was addressed by Dr. Sokolov, who explained that in light of petitioner’s intoxication, these actions would be consistent with a lack of intent. Moreover, counsel’s failure to investigate renders any argument based on strategy inapposite. See Jennings, 290 F.3d at 1018-18. See also Hart v. Gomez, 174 F.3d 1067, 1070 (9th Cir.1999) (holding that “[a] lawyer who fails adequately to investigate, and to introduce into evidence, [information] that demonstrate^] his client’s factual innocence, or that raise[s] sufficient doubt as to that question to undermine confidence in the verdict, renders deficient performance.”). In conclusion, it is apparent that trial counsel never investigated the effects of alcohol on petitioner’s perception, cognition, or ability to form the requisite intent. Nor did counsel consult any expert who could have educated the jury as to the effects of alcohol. Instead, counsel acted on their own assumptions about intoxication evidence. Accordingly, their choice of strategy was not based on any investigation and was unreasonable under Strickland. b. Failure to Present Evidence Counsel also performed unreasonably when they failed to retain an expert on the effects of intoxication and failed to present such evidence to the jury. This evidence would have directly supported petitioner’s claim that he had been intending to only stop Faddis, and had not intended to kill him. *502 Neither of petitioner’s trial attorneys could recall at which point it was decided to abandon a defense based on voluntary intoxication. It is undisputed that an intoxication theory was “on the table” as of November 1996, and that subsequently a decision was made to abandon it. See EH 27-38. Trial counsels’ stated justifications for why they did not present evidence of intoxication are without merit. First, trial counsel testified that they pursued a theory of “perfect self-defense” because that is the theory petitioner wanted to pursue. Putting aside the problem of counsel’s obligation to make an independent judgment, see discussion supra, the facts of the case demonstrate that a perfect self-defense theory would be very difficult to establish. Petitioner was charged with murder on the basis of a shooting that he subjectively believed to be justified, even though the facts were difficult to reconcile. Petitioner was armed, Faddis was not. The beatings of Goold was over and she was safe inside the house, though petitioner thought she was underneath the truck. Finally, petitioner was approaching Faddis when he fired the shots. These facts do not support a theory of perfect self-defense. Most importantly, the “reasonable person” standard is impossible to reconcile with a defendant who has been drinking all day and has a BAC of at least .30. In light of these facts, it was not reasonable for counsel to defer to petitioner’s desire to pursue a perfect-self defense theory. Second, trial counsel did not present evidence of intoxication because they believed that evidence of intoxication would have contradicted their self-defense theory. See EH 124-125, 145-146. As previously discussed, just the opposite was true. In order to argue perfect self-defense, the jury would have had to conclude that a person with a stipulated .30 BAC was acting as a “reasonable person.” Evidence regarding petitioner’s alcohol-related impairments would have explained how he could have genuinely, even if unreasonably, believed himself to be in life-threatening danger. As testified to by Dr. Sokolov, the effects of intoxication explain how petitioner could have not intended to kill despite facts which would support an inference of such intent in a sober person. For example, the prosecutor argued that the fact that petitioner shot a gun directly at Fad-dis’s chest meant he must have deliberately shot, to kill. This reasoning assumes that the person shooting is able to appreciate the likely result of his actions, understand cause and effect, and make rational decisions in light of risks. As Dr. Sokolov explains, none of these assumptions hold true when considering petitioner’s high BAC. See EH 39-41. Petitioner argues that it is credible that he “could have taken actions facially suggestive of specific intent to kill, without actually intending anything more than to ‘stop’ a situation out of control.” Pet’rs Closing Postr-Hrg. Br. at 14. Respondent relies on White v. Single-tary, in which the Eleventh Circuit found no unreasonable performance in counsel’s failure to present an intoxication defense because “it was inconsistent with the deliberateness of White’s actions during the shootings.” 972 F.2d 1218, 1221 (11th Cir.1992). White is clearly distinguishable from the case at bar. First, the facts are inapposite. The defendant in White drove to a grocery store, entered with a gun, and shot two customers in the back of the head. Here, the shooting was, as petitioner puts it, “a spontaneous response to dramatic interpersonal conflict fueled by alcohol.” Pet’rs Closing Post-Hrg. Br. at 15. Second, in White, petitioner tendered no expert testimony regarding the intoxi *503 cation defense which could have been presented in his case. In contract, petitioner in the instant case has presented expert medical testimony on the relationship between intoxication and the formation of malice as applied to the facts of the case. For these reasons, White is inapplicable. Because counsel failed to investigate and present evidence on the effects intoxication on petitioner’s ability to form the requisite intent, the court holds that counsel’s performance was constitutionally deficient under Strickland. 4. Whether Counsel’s Deficient Performance Prejudiced Petitioner Having determined that trial counsel acted unreasonably in failing to investigate and present evidence of petitioner’s intoxication, the court turns next to the question of whether petitioner was prejudiced by his counsels’ unreasonable performance. Strickland, 466 U.S. at 687, 104 S.Ct. 2052. To show prejudice under Strickland, a 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, 104 S.Ct. 2052. The focus of the prejudice test is on “whether counsel’s deficient performance renders the result of the trial unreliable or the proceeding fundamentally unfair.” Lockhart v. Fretwell, 506 U.S. 364, 372, 113 S.Ct. 838, 122 L.Ed.2d 180 (1993). As the Ninth Circuit explained: A reasonable probability does not mean that we must determine that the jury more likely than not would have returned a verdict for something beside first degree murder, but only that Mr. Jennings has shown “a probability sufficient to undermine confidence in the outcome.” Jennings, 290 F.3d at 1016. For the reasons discussed herein, the court concludes that this standard is satisfied. a. Intoxication Could Negate Express & Implied Malice Petitioner was charged with second degree murder, which is the unlawful killing of a human being with malice aforethought. Cal.Penal Code §§ 187, 189. Malice may be either express or implied. Express malice is the deliberate intention to kill. Cal.Penal Code § 188. Malice may be implied “when [the] defendant does an act with a high probability that it will result in death and does it with a base antisocial motive and with a wanton disregard for life.” People v. Watson, 30 Cal.3d 290, 300, 179 Cal.Rptr. 43, 637 P.2d 279 (1981) (citations omitted). A finding of implied malice thus “depends on a determination that the defendant actually appreciated the risk involved, i.e., a subjective standard.” Watson, 30 Cal.3d at 296-297, 179 Cal.Rptr. 43, 637 P.2d 279. At the túne of petitioner’s trial, it was well settled under California law that evidence of intoxication, like that of mental illness, could be considered in assessing a defendant’s subjective state of mind related to the presence or absence of malice. This includes the factual question of whether a defendant unreasonably believed that he faced life-threatening danger. People v. Wells, 33 Cal.2d 330, 202 P.2d 53 (1949). Because the prosecutor’s theory focused on express malice, evidence of voluntary intoxication would have also been admissible under California Penal Code § 22(b). 10 *504 The prosecutor argued to the jury that petitioner’s denial of intent to kill was inconsistent with the facts. RT 907. For example, the prosecutor deemed “incredible” petitioner’s perception that Faddis, who was drunk himself, presented a real danger and that petitioner truly wanted to only stop Faddis, not kill him. Id. Dr. Sokolov established, however, that petitioner’s brain functioning was impaired and accordingly, assumptions about intent that might apply to a sober person, did not apply to petitioner. If a sober person pointed a gun directly at a person and shot, a jury may well be justified in concluding that the shooter, a rational person, was making an intentional decision. The same cannot be said with someone who has a BAC of .30. Petitioner’s ability to walk and talk similarly does not indicate that petitioner also had the ability to form intent. As Dr. Sokolov testified, in an alcohol-tolerant individual, significant impairment to executive functioning may coexist with largely intact speech and motor skills. See EH 49-50. This includes CPR, which petitioner performed on Faddis. Petitioner was a certified EMT, RT 636, who had been trained to perform CPR, id. at 660. According to Dr. Sokolov, an experienced EMT such as petitioner would not need much executive functioning to perform CPR. EH 50, 81. Likewise, loading and firing a gun can be rote behavior for a person experienced with guns, such as petitioner. EH 73. Petitioner used his shotgun regularly to shoot squirrels, woodpeckers, and other small animals on his rural property. Faddis’ father testified that petitioner was a “good shot” with the gun. RT 126-128. Accordingly, petitioner’s ability to handle a gun while extremely intoxicated does not necessarily indicate that he was thinking rationally. Because petitioner’s levels of physical and cognitive impairments were arguably divergent (for example, he could perform CPR and shoot a gun, yet still be mentally impaired), expert testimony was needed to explain the effects of alcohol on the brain. It was incumbent upon petitioner’s trial counsel to raise reasonable doubt as to malice. The medical evidence, illustrated by Dr. Sokolov’s testimony, raises sufficient doubt as to petitioner’s intent. Dr. Sokolov opined that with petitioner’s BAC, petitioner was unable to accurately perceive whether his life or Ms. Goold’s life was in danger and was also unable to cognitively process that a shot to the chest could be lethal. EH 37. He also concluded that petitioner’s ability to appreciate the cause and effect relationship (between shooting and killing) was significantly impaired. EH 38. It was Dr. Sokolov’s opinion that petitioner’s intoxication precluded him from forming an intent to kill. EH 39. Dr. Sokolov explained that the formation of intent to kill depends on executive functions performed by the frontal lobe. EH 37. Intoxication, however, impaired petitioner’s ability to understand cause and effect, to think ahead and to identify a specific desired outcome. EH 38-39. At the evidentiary hearing, Dr. Sokolov testified that had he been contacted by petitioner’s trial counsel and had reviewed the same material he would have concluded that petitioner acted without specific intent' to kill, without realizing that his actions could kill Faddis, and with the subjective belief that he was acting to save *505 his own life and/or that of Ms. Goold. ED 47_47. 11 For all these reasons, evidence of intoxication would have likely created a reasonable doubt about petitioner’s intent to “undermine confidence in the outcome.” Strickland, 466 at 649, 104 S.Ct. 2052 A jury informed as to the effects of intoxication would have found that petitioner did not appreciate the risk of death and the likely result would have been acquittal or a involuntary manslaughter conviction. Saille, 54 Cal.3d at 1116-17, 2 Cal.Rptr.2d 364, 820 P.2d 588. b. Imperfect Self-Defense Even if the jury did not conclude that intoxication negated malice, evidence of petitioner’s cognitive impairment could have lead to a finding that he genuinely perceived Faddis to pose an imminent life-threatening danger. That conclusion would have resulted in a voluntary manslaughter verdict under People v. Flannel, 25 Cal.3d at 669, 160 Cal.Rptr. 84, 603 P.2d 1 (1979). The California Supreme court explained the central premise of “imperfect self-defense”: When the trier of fact finds that a defendant killed another person because the defendant actually but unreasonably believed he was in imminent danger of death or great bodily injury, the defendant is deemed to have acted without malice and cannot be convicted of murder. In re Christian S., 7 Cal.4th 768, 783, 30 Cal.Rptr.2d 33, 872 P.2d 574 (1994). 12 Dr. Sokolov’s testimony supports a finding of imperfect self-defense. As previously discussed, Dr. Sokolov testified that alcohol significantly impaired petitioner’s perceptions and his ability to reasonably evaluate the degree of danger posed by Faddis. EH 45-46, 52-53. The facts are consistent, Dr. Sokolov explained, with petitioner subjectively believing that his actions were necessary to protect himself. Medical evidence about intoxication would have explained, for example, petitioner’s statements to the police that he believed he and Ms. Goold were in immediate danger when he shot. “[WJhether the defendant actually held the required belief is to be determined by the trier of fact based on all the relevant facts.” In re Christian S., 7 Cal.4th at 783, 30 Cal.Rptr.2d 33, 872 P.2d 574. Dr. Sokolov explained that an evaluation of petitioner’s actual state of mind was not possible without specific consideration of his intoxication. For these reasons, even if the jury did not return a verdict of involuntary manslaughter on the basis of intoxication, it is reasonably likely that, in *506 the alternative, they would have found petitioner liable of no more than voluntary manslaughter based on imperfect self-defense. Clearly, intoxication evidence would have been relevant in .undercutting the prosecution’s theory about intent in three different ways. The jury could have acquitted or convicted petitioner of involuntary manslaughter on the grounds that petitioner was too intoxicated to have the requisite mens rea for a murder conviction. In the alternative, the jury could have convicted petitioner of voluntary manslaughter on the theory of imperfect self-defense. For these reasons, the court concludes that “there is a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.” Strickland, at 694, 104 S.Ct. 2052 IV. Conclusion After careful review of the record, particularly the facts of the crime and the evidence concerning petitioner’s intoxication on the night of the crime, the court finds there is a reasonable probability that, absent the deficiencies, the outcome of the trial might well have been different. See Strickland, 466 U.S. at 695, 104 S.Ct. 2052. Because petitioner’s level of intoxication was clearly relevant to create reasonable doubt as to the mens rea element of the offense, confidence in the outcome is seriously undermined by trial counsel’s failure to investigate and present that evidence. Accordingly, the application for a writ of habeas corpus is GRANTED. Petitioner shall be released from custody unless a new charging document is filed within sixty [60] days of the date of this order and petitioner is tried on that document in due course. IT IS SO ORDERED. Footnotes [1] The summary of the evidence is adopted from the parties’ papers and after careful review of the record. [2] Respondent did not present an expert on petitioner’s state of intoxication. Accordingly, this court adopts the finding of the magistrate judge that "in matters of medical expertise ... the expert's testimony will be the factual finding of the court.” Findings and Recommendations at 13: 6-8. [3] The parties stipulated that Dr. Wooley's declaration would be accepted as testimony. Dr. Wooley’s declaration was filed on November 8, 2004. [4] It is common for an IAC claim to have several components. Prejudice from ineffective assistance of counsel is analyzed cumulatively. See Harris v. Wood, 64 F.3d 1432, 1438-39 (9th Cir.1995). [5] The Ray decision was overruled in 2000 by People v. Blakeley, 23 Cal.4th 82, 89, 96 Cal. Rptr.2d 451, 999 P.2d 675 (2000). However, Ray was the law at the time of petitioner’s trial. [6] To be clear, the jury was instructed on murder in the second degree, and voluntary manslaughter. A jury instruction on the lesser included offense of involuntary manslaughter was not given. [7] Involuntary' manslaughter is an unintentional killing under circumstances which make it nonetheless unlawful. CALJTC 8.45 (6th ed.1996) (unlawful killing without malice and without intent to kill is involuntary manslaughter). [8] The court notes that the People v. Blakeley, 23 Cal.4th 82, 96 Cal.Rptr.2d 451, 999 P.2d 675 (2000), discussed at length by the magistrate judge, was decided well after petitioner was convicted and thus, is not pertinent. It is well established that an IAC claim must be evaluated in light of the state of the law at the time of the petitioner's trial. Strickland, 466 U.S. at 689, 104 S.Ct. 2052. An attorney is only charged with knowing the law as it existed at the time the action or inaction was taken. Lowry v. Lewis, 21 F.3d 344, 346 (9th Cir.1994). [9] Voluntary manslaughter encompasses intentional killing, under circumstances which nonetheless negate malice. People v. Rios, 23 Cal.4th 450, 454, 97 Cal.Rptr.2d 512, 2 P.3d 1066 (2000) (intentional but non-malicious killing is voluntary manslaughter); CALJIC 8.40 (6th ed.1996) (every person who unlawfully kills without malice but with intent to kill is guilty of voluntary manslaughter). [10] California Penal Code § 22(b) (rev. 1995) provides: Evidence of voluntary intoxication is admissible solely on the issue of whether or not *504 the defendant actually formed a required specific intent, or, when charged with murder, whether the defendant premeditated, deliberated, or harbored express malice aforethought. [11] It is important to note that Dr. Sokolov would not have been permitted to testify at trial as to whether he believed petitioner harbored malice. Pursuant to California Penal Code § 29, "in the guilt phase of a criminal action, any expert testifying about a defendant’s mental illness, mental disorder, or mental defect shall not testify as to whether the defendant had or did not have the required mental states ... The question as to whether the defendant had or did not have the required mental states shall be decided by the trier of fact.” Cal.Penal Code § 29. However, the opinions expressed at the evi-dentiary hearing would have been provided to counsel and could have formed the basis for a defense theory that accounted for petitioner’s level of intoxication. At trial, counsel could have presented physiological and bio-chemical expert testimony, asked hypothetical questions based on the facts of the case, and then invited the jury to draw the ultimate conclusion as to intent. [12] This approach would not fall within the scope of Penal Code § 22 and instead would be governed by the standards for imperfect self-defense, which allows for evidence of intoxication. See Christian S., 7 Cal.4th at 783, 30 Cal.Rptr.2d 33, 872 P.2d 574.