Having been convicted of first degree murder and sentenced to death by a jury, Alvin Scott Loyd petitions for federal habeas corpus relief claiming ineffective assistance of counsel in the penalty phase of his capital murder trial. The district court denied his petition. Finding ineffective assistance of counsel in the penalty phase, we reverse, render, and remand.

Loyd was charged with the capital murder of three-year-old Brandi Giovanetti.1 Local law enforcement officials apprehended Loyd when he returned to his residence. Loyd was held in custody at the Feliciana Forensic Facility following a Sanity Commission determination that he was not competent to stand trial. After four months Loyd was deemed competent to stand trial.

A jury found Loyd guilty and imposed the death penalty. The Louisiana Supreme Court affirmed the conviction but vacated the death sentence, remanding for a new sentencing trial because a faulty instruction to a hesitant jury violated the integrity of the unanimous jury verdict. At the second sentencing trial the jury again imposed the death penalty. The sentence was affirmed on appeal.2

The state trial court denied Loyd's first petition for post-conviction relief but the Supreme Court of Louisiana granted a stay of execution and remanded the case to the state trial court for an evidentiary hearing on four issues, including the ineffective assistance of counsel claim.3 After a hearing, the state court concluded that the performance of counsel at the second sentencing trial was deficient; however, the state court denied habeas relief on the ground that counsel's deficient performance did not prejudice Loyd. The Louisiana Supreme Court denied Loyd's application for review, assigning no reasons.

After exhausting state court remedies, Loyd sought federal habeas relief. The district court granted a stay of execution but ultimately denied Loyd's requested relief. In regard to the ineffective assistance claim, the district court concluded that, contrary to the finding by the state court, counsel's performance was not deficient. We vacated the district court finding on the ground that proper deference had not been given to the state court's findings of fact as required by 28 U.S.C. § 2254(d).4 We directed the district court to conduct an evidentiary hearing if it concluded that the record was not fully developed.

The district court did not conduct a hearing but reviewed the state habeas court findings, adopted some, rejected others, and reached conclusions of its own. The court again held that the performance of counsel was not deficient and additionally found that any hypothetical deficiency did not prejudice Loyd.

State Proceedings: Sentencing and Post-Conviction Hearing

After conducting an evidentiary hearing, the state habeas court concluded that the professional performance of Loyd's defense counsel in the 1985 sentencing trial fell below reasonable professional standards. Loyd's defense team was composed of three attorneys. Court-appointed counsel Gordon Hackman and Randy Lewis had represented Loyd in the 1983 proceedings and had asked permission to withdraw as counsel three weeks before the second sentencing trial. This request was denied but William Allison was added to, and headed, the defense team. At that time Allison's law practice was, as described by him, "ninety percent civil, various mix and ten percent criminal." Allison had practiced law for 14 years and had participated in approximately six criminal jury trials, including one capital case. Hackman, who had been lead counsel at the guilt/innocence phase, had a practice composed primarily of civil litigation, although his firm had accepted a number of criminal cases in the mid 1970s. Lewis was his law partner. At the state habeas hearing, all three attorneys expressed dissatisfaction with their representation of Loyd in the 1985 sentencing trial.

At trial the defense called three doctors, all of whom had been retained by the state to determine Loyd's competence to stand trial. Also in evidence at the penalty phase were sanitarium admission papers reciting an initial diagnosis of "Antisocial Personality Disorder," a Psychological report, a Neuropsychiatric Examination report, a Neurological Examination report, and the report of a social worker. Most of the psychological testimony was presented by Dr. Cox, Loyd's treating physician at the Feliciana Forensic Facility, where Loyd was held during the four months that he was diagnosed as incompetent. Allison spoke with Dr. Cox for the first time on the day of the trial, for 45 minutes during the lunch hour. Also testifying were Dr. St. Martin, the Feliciana coroner and a member of the Sanity Commission that found Loyd initially incompetent, and Dr. Ritter, the other Sanity Commission member.

Although Dr. St. Martin described "an in-depth exam" with Loyd, Dr. Ritter emphasized that he "did not do a detailed personality inventory on Mr. Loyd." Dr. Ritter also described the role of the Sanity Commission as limited:

When you evaluate someone in a prison setting and someone who is depressed sometimes that's very difficult to get any details. Besides you're there for two specific purposes. Not to get a personality inventory, to make a detailed study of personality, but to determine if there are any mental diseases or defect which could impair his ability to proceed to trial or could impair his ability to tell the difference between right and wrong.

No independent psychiatrist or psychologist testified on Loyd's behalf in the sentencing phase despite the fact found by the state court that Loyd's sanity was a critical issue. Allison had initially requested $1,000 to hire an independent, neutral psychiatrist to testify in Loyd's defense; $600 was approved. Thereafter, according to the state court, "Allison made several half-hearted attempts to procure [independent, psychiatric] services, but eventually abandoned those efforts." Hackman had $1,250 which had been given to him by Loyd's mother to be used for Loyd's defense. Hackman did not inform Allison of the availability of these funds. Hackman did not pursue further psychiatric testimony because he believed that such an attempt would have been futile.

The state court made the factual finding that Hackman's decision was based upon a failure to understand the difference between the McNaughten test for sanity and the Louisiana mitigating factors of "mental or emotional disturbance," or "mental disease or defect."5 The state court also found that the aggregate funds available were sufficient for an independent psychiatric analysis of Loyd. The court concluded that "for counsel not to have sought such an evaluation, where funds were available to do so, was an error which fell below the professional standards of conduct required to constitute proper representation."

Loyd's new habeas counsel sought the services of doctors whose testimony in the state habeas evidentiary hearing presented a more detailed explanation of Loyd's mental impairments. Dr. Kenneth Perkins, a clinical psychologist, reviewed the raw data from the Feliciana tests and determined that to a large extent the data had been misinterpreted in such a way that Loyd did not receive additional necessary testing.6 Dr. Stephen Honor, also a clinical psychologist, repeated the psychological tests; the raw data from the new psychological tests mirrored the Feliciana raw data. Dr. Honor, like Dr. Perkins, found that the Feliciana staff's interpretation of the data understated the extent of Loyd's mental impairments. Similarly, Dr. Barry Scanlon examined Loyd, reviewed his records from Feliciana, and disagreed with the Feliciana conclusions. Dr. Sanchez, the forensic psychiatrist requested by counsel for the state, examined Loyd after both the state and the defense accepted him as qualified. Because of the opinion formulated after this review, Dr. Sanchez was called to testify by Loyd's counsel.

The evidence presented at Loyd's sentencing trial--evidence based on the Feliciana conclusions--markedly differs from the new evidence.

At the penalty phase, Drs. St. Martin and Ritter testified that at the time of the crime, Loyd knew the difference between right and wrong. Dr. Cox concurred that Loyd was "sane" at the time of the crime. In the state habeas hearing, Dr. St. Martin testified that he based his diagnosis of Loyd's sanity on "how he was feeling, [he was not] hallucinating, his general psychological condition at the time, and the things that he was able to do at the site of the crime and after the crime." Dr. St. Martin stated that he did not believe it was possible for individuals to "not show ... any psychosis any other time but just have a short burst of psychosis."

The experts presented by Loyd's habeas counsel all expressed doubts regarding Loyd's sanity at the time of the crime. Dr. Honor reported that:

While Mr. Loyd did apparently form an intent to kill the child, the evidence of his mental and emotional state, based upon his self report and the examination results, strongly suggests that he would have been unable to appreciate the significance of this act or the probable consequences of this act. Therefore, at that point in time it can be said that Mr. Loyd could not distinguish between the concept of right and wrong.

Dr. Honor opined that there was an 80 to 90 percent probability that Loyd did not understand his actions at the time of the crime. Similarly, Dr. Scanlon reported that:

I think a convincing case can be made that Loyd was incapable of distinguishing between right and wrong in view of my conclusion that the quality of Loyd's acts were delusional (psychotic) or delirious (impaired by physical factors).... Loyd may have had a rather limited awareness that he was drowning, raping and sodomizing a little girl (the "nature" of his acts) but it is unlikely in view of his delusional (psychotic) or delirious (impaired by physical factors) state that he could have understood the moral significance of his acts (the "quality" of his acts).

Dr. Honor described how Loyd could move in and out of a psychotic state:

Loyd represents an individual who is a borderline psychotic. He has a number of personality traits that certainly appear to be [of] very long standing duration.... [I]ndividuals [who] get that designation ... under reasonable circumstances are able to hold themselves together. There is some kind of personality integrity they can hold together. When individuals like that are exposed to very stressful circumstances the great likelihood is that there is going to be a deterioration.... [I]t may in fact precipitate the development of a frank psychosis. But probably in many, many cases they become temporary episodes where the person spontaneously regenerates so to speak after the stressors have been ... by-passed.

The sources of information "most pertinent" to Dr. Honor in making this diagnosis were his clinical interviews with Loyd and Loyd's letters.

Dr. Sanchez, an expert forensic psychiatrist frequently appointed to state sanity commissions, considered Loyd to have experienced a "psychotic episode" with "sensations of losing control over his mind." Loyd knew right from wrong at some points, but not at others.

The Sanity Commission determined that Loyd was initially incompetent to stand trial because "he was severely depressed and he could not best assist his counsel at that time." Dr. Ritter, a member of the Sanity Commission, confirmed the Commission's finding that Loyd was depressed:

[Loyd] was crying, expressed remorse, ... best thing I could say, is he, he was suffering from depression, he was feeling blue, remorseful, down in the dumps, and as a result of that he was more difficult than usual to talk to because his thoughts were not coming very rapidly, he was slowed up, like most people who are depressed experience.

Dr. Cox, Loyd's treating physician at Feliciana, also stated that Loyd had been depressed and despondent, but stated that Loyd's incarceration for the crime was a major part of his depression. Similarly, Dr. St. Martin acknowledged that it was very possible that remorse at being caught could have caused Loyd's depression.

Dr. Honor disagreed with "[t]he conclusion by the forensic experts that Mr. Loyd's level of disturbance was likely to be a result of his reaction to his crime and his arrest rather than a representation of his general psychological state."

Dr. Ritter said little regarding Loyd's personality traits because he did not do a detailed personality inventory on Loyd.7 Dr. Cox had ordered a neurology exam and other psychological tests. At the time of trial, Dr. Cox stated: "I cannot explain why this happened." Defense counsel questioned Dr. Cox regarding specific findings in a psychological evaluation report performed at Feliciana. At the onset, Dr. Cox stated that he agreed with parts of the report yet had doubts regarding other conclusions. This report was admitted into evidence. The report recited the probable etiology of the crime:

Mr. Loyd was himself a victim of child abuse and he had a very poor model for adult masculine behavior. He was not able to integrate the antisocial patterns of behavior he learned from his father with the excessively moralistic standards he was taught by his mother. One consequence of this confusion was that he had not obtained a stable adult sexual adjustment. And another consequence was that he developed a pattern of substance abuse.

Regarding this conclusion, Dr. Cox testified: "I think it's very clear that he developed a pattern of substance abuse. As far as stable adult sexual adjustment, I can't comment on that." Dr. Cox testified that the following excerpts from the Feliciana report were "consistent with Mr. Loyd's personality and makeup":

On the day of his crime, his intoxication rendered his inhibitions ineffective. Apparently he committed the crime cognizant of the wrongness of it but unwilling or unable to stop himself.... Later overwhelmed by horror over his actions he avoided memory of the events.

Regarding Loyd's history of child abuse, Dr. Cox testified that a very high percentage of people who commit violent acts have a history of abuse as children. He further explained that the kind of stimulus that could trigger Loyd into blowing up included stress, pressure, substance abuse, or sleep deprivation.

Dr. Honor reviewed the raw test scores from the Feliciana tests, performed his own tests, and found many of the raw test scores to be consistent with the Feliciana results. Pursuant to his interpretation, the test results evidence somatic delusions, disordered thinking, chronic psychological maladjustment with chronic disorientation, alienation and withdrawal. Dr. Honor found Loyd to be "of a borderline psychotic nature [with] very clearly paranoid ideology and schizophrenic personality characteristics." "Diagnostically he is seen as manifesting a schizoid or schizotypal personality.... The personality profile indicates extremely high elevations in overall emotional disturbance, depression, and schizophrenia." Dr. Honor explained the circumstances of the crime:

Loyd's description of his subjective perceptions from the time of being in the bar until after committing the crime strongly suggests an altered state of consciousness, part of which seems to be rooted in several somatic delusions.... Mr. Loyd reports strong feelings of confusion through the night and a dreamlike quality to his state of consciousness.... The overwhelming feelings of being threatened, which appear to have been significantly exacerbated by his diminished state of cognition and his underlying paranoid ideation, seems to have triggered a response of rage and fear that became displaced onto this child, who was, at the time, seen as a significant part of this sense of threat. The decision to kill the child seems to have been perceived as a means of eliminating the threatening circumstances.... Loyd's sexual attack of the child is seen as a violent weapon aimed at silencing the child.

Dr. Perkins likewise concluded that the Feliciana interpretation of Loyd's raw test data tended to overlook or understate the extent of Loyd's chronic emotional disturbance. Some of the patterns are "typically regarded as the classic paranoid-schizophrenia profile." Drs. Perkins, Scanlon, and Sanchez each found no basis for the Feliciana admissions report diagnosis that Loyd has an antisocial personality disorder.8

Dr. Scanlon testified that the Feliciana psychological report "was insufficient in terms of trying to ... come up with an explanation for Loyd's behavior." Dr. Scanlon believed that physical impairment of thought was the ultimate cause of the crime:

I personally do not see Loyd's crime as primarily determined by emotional or psychological factors.... [T]he "form" his actions took (the killing of a little girl after raping and sodomizing her) was probably determined by "emotional or psychological factors" rooted in Loyd's childhood experience. I believe, however, that what triggered his actions was largely an organic (or physical) impairment of thought.

A neuropsychiatric report mentions scarring of the right temple, weakness on Loyd's left side, and some decreases in sensory response.9 The report concludes that Loyd "exhibits no evidence of overt psychosis or brain syndrome" and that the "focal neurological findings in terms of weakness and sensory deficit are fleeting and would appear at this time to be hysterical in etiology." A neurological exam, although finding some inconsistencies in sensory responses, did not report neurological impairment or neuromuscular disease.

Testing for Organic Brain Damage was not extensive. Some evidence of problems coordinating movement in his left hand was revealed, thus lending support to his complaints of weakness in this area. It is not possible to say when the damage might have been sustained. Other than that, no evidence of brain damage was found.... [Loyd] may have sustained some minor brain damage. This appears to be unrelated to either the commission of the crime or his memory loss.

Dr. Perkins reported that the Feliciana tests inadequately investigated the possibility of brain damage. According to Dr. Perkins, the raw test scores from the Feliciana psychological evaluation contained inconsistencies associated with underlying organic factors, such as brain dysfunction. Dr. Perkins found that the following facts point to the possibility of brain damage: (1) that Loyd's mother sustained paint poisoning one month before his birth, (2) family history of epilepsy, (3) family history of dyslexia, (4) Loyd's history of head injuries, (5) the radiation therapy Loyd received as an infant for the open exposure of blood vessels in his temple, (6) family history of alcoholism, (7) Loyd's personal history of chemical abuse, (8) Loyd's history of headache experience, and (9) the possibility of a head injury while boating on the day of the crime. Regarding the neurological exam performed at Feliciana, Dr. Perkins stated:

The neurological exam at Feliciana observed the left-handed weakness in [Loyd's] behavior as well as significantly lower sensation to touch or pin prick on the left side of his body, including arm and leg but was regarded as probably hysterical symptomatology. That certainly would be consistent with Dr. Fain's comment that [Loyd] may tend to express anxiety through somatic complaints, but it also may be indicating, again, a nervous system disorder which, like the possibility of his headaches, may be somatic reactions with neurological basis which are activated by times of intense emotion which would include anxiety, stress/tension, etc.

Dr. Honor reported that the tests of neuropsychological functioning "while not definitive, are consistent with a diagnosis of frontal lobe dysfunction. Such dysfunction would be expected to have an impact on judgment, and higher levels of cognitive functioning such as organization of thought and processing of complex ideas."

An ineffective assistance of counsel claim requires a two-part showing that: (1) Counsel's performance was deficient, meaning that "counsel made errors so serious that counsel was not functioning as the 'counsel' guaranteed the defendant by the Sixth Amendment"; and (2) the deficient performance prejudiced the defendant by depriving him "of a fair trial, a trial whose result is reliable."10 A state court conclusion regarding effective assistance of counsel is a mixed question of law and fact. State court findings of fact made in the course of deciding a sixth amendment ineffectiveness claim are subject to the deference requirement of section 2254(d).11

In determining whether counsel's representation passes constitutional muster, we must make an independent evaluation based on the state court's subsidiary fact findings; we need not defer to the district court's conclusion.12

"The proper measure of attorney performance remains simply reasonableness under prevailing professional norms.13 All of the circumstances must be considered. In this consideration, however, we must guard against the temptations of hindsight and second-guessing. Accordingly, a reviewing court must grant counsel's decisions a reasonable presumption of reasonability.14 Applying these guidelines, the state habeas court concluded that the performance of counsel in Loyd's 1985 sentencing trial fell below reasonable professional standards. The state court found that funds for an independent psychiatric expert were available, that Loyd's sanity was a critical issue, that counsel were aware of the critical nature of Loyd's sanity, and that lead counsel made only "half-hearted" attempts to obtain independent evidence regarding this sanity and abandoned those efforts for no strategic purpose. The court a quo acknowledged these factual findings as binding pursuant to section 2254(d), but by discounting other state findings as being erroneous or irrelevant, and by supplementing the findings, the district court avoided the state court conclusion that counsel's performance was not within the range of professional reasonability. The district court opined that calling Drs. Cox, Ritter, and St. Martin "was more than reasonable and constitutionally adequate under the circumstances prevailing at the time of trial."

The district court discounted the state habeas court finding that: "[T]he most compelling fact in the performance of counsel in this matter is that they simply failed to develop independent psychiatric evidence of mental disease or defect in a death case where this line of investigation was clearly indicated." Regarding this statement, the district court held that "to the extent that the state court implied ... that no statutory mitigating circumstances were presented on Loyd's behalf at trial, ... such a conclusion is not supported by the record and is clearly erroneous." The district court made the supplemental conclusion that Loyd's counsel presented significant statutory mitigating psychiatric testimony. We cannot agree, but accept as binding the state court finding that defense counsel's failure to pursue a crucial line of investigation in a capital murder case was not professionally reasonable.

The state court found that Hackman did not appreciate the distinction between the McNaughten rule and the mitigating circumstances of mental disease or defect. Hackman, who was court-appointed and did not wish to represent Loyd, testified at the state hearing that despite his general experience with criminal trials, he did not feel competent to handle the case, he was convinced that the defense should have conducted additional investigation and should have obtained the assistance of a forensic psychiatrist. Allison, who had participated in at least six criminal jury trials, testified that he did not believe that his representation of Loyd contributed to the defense.15 The state court discounted this testimony, stating: "[I]n the case of Mr. Allison, his personal abhorrence of the death penalty might well have affected his later assessment of his conduct." The district court accepted the state court decision to discount Allison's testimony, but dismissed as inconsequential that court's fact-finding that Hackman misunderstood the law regarding mitigating circumstances. According to the district court, Hackman's knowledge was irrelevant because Hackman was not lead counsel.16 Aside from noting our consistent insistence that effectiveness of counsel requires familiarity with current state law,17 we refrain from further dissection of Hackman's role. The essence and significance of the state court's finding of fact is that no member of the defense team made the strategic decision referred to in Strickland regarding the critical issue in this death case.

The contrast between the data and testimony presented in the 1985 proceeding and the new evidence warrants our rejection of the district court's supplemental finding that Loyd's defense counsel introduced significant mitigating evidence. The marked disparity in testimony establishes that reasonable professional standards require that counsel should have used readily available funds to hire an independent psychiatrist to put Loyd's mental condition in proper focus.

"[C]ounsel has a duty to make reasonable investigations or to make a reasonable decision that makes particular investigations unnecessary.18 Counsel's professional duty may be met in more than one way:

[S]trategic choices made after thorough investigation of law and facts relevant to plausible options are virtually unchallengeable; and strategic choices made after less than complete investigation are reasonable precisely to the extent that reasonable professional judgments support the limitations on investigation. In other words, counsel had a duty to make reasonable investigations or to make a reasonable decision that makes particular investigations unnecessary.19

The state court's factual findings make clear that the decision of defense counsel not to pursue an independent psychological analysis of Loyd was neither a strategic choice made after investigation nor a strategic choice made in light of limits on investigation. There were no limitations; funds were available. According to the state factual findings, Allison's decision had nothing to do with strategy; he wrongly assumed that funds were unavailable and he abandoned what he knew to be an important pursuit. Hackman's decision had not been made after thorough investigation of the law; Hackman was unaware of the law.20

Whether counsel's omission served a strategic purpose is a pivotal point in Strickland and its progeny.21 The crucial distinction between strategic judgment calls and plain omissions has echoed in the judgments of this court.22 For example, in Nealy we found counsel's performance deficient and stressed that at a post-conviction hearing, the defense counsel "did not testify that such efforts would have been fruitless, nor did he claim that the decision not to investigate was part of a calculated trial strategy. He simply failed to make the effort."23 Counsel " 'did not choose, strategically or otherwise, to pursue one line of defense over another. Instead, [he] simply abdicated his responsibility to advocate his client's cause.' "24

Finding a void of available evidence on a critical issue in a death penalty case, we hold that counsel's performance was not professionally reasonable.

The prejudice component of the Strickland test requires a showing

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. In making the determination whether the specified errors resulted in the required prejudice, a court should presume ... that the judge or jury acted according to law.25

The state court and the district court held that Loyd had not demonstrated prejudice.

The state court found that the opinions of the experts testifying in the habeas proceeding were, to some extent, based on three unproven hypothetical facts: that Loyd was involuntarily drugged, that he worked unusually long hours in excessive heat in the days preceding the crime, and that he suffered a concussion on a boat ride on the day of the crime. The state court declared that, "the opinions of Drs. Honor, Scanlon, and Sanchez, to the extent that they were based on these assumptions, would probably have had no influence on the jury's assessment of petitioner's psychological condition on the night of the crime."26 Regarding frontal lobe brain damage,27 the state court observed that "no more than a suspicion was indicated, even by his own experts, that this irregularity explained or excused the crime." The court considered whether there was "a reasonable probability that the remaining overall testimony of Drs. Honor, Scanlon and Sanchez as to petitioner's diminished capacity or mental disease or defect, would have been found by at least one juror, to be a mitigating factor sufficient to change his or her verdict of death." The court accepted the doctors' conclusions that Loyd suffered from a mental disease or defect that in all probability predated the crime. Yet, according to the court, "[t]he enormity of the crime committed was so great that any mental disease or defect, short of legal insanity, would simply not, in this court's opinion, have tipped the balance in any juror['s] mind so as not to warrant a sentence of death."

Giving deference to the state habeas decision, we accept as unproved the theories that Loyd was drugged and exposed to excessive heat, that he suffered a head injury on the day of the crime, or that the clinical signs of physical abnormality in his frontal lobe brain functions explained the crime. The state court did not explain what portions of the experts' opinions rested on these assumptions; however, it is apparent that the court found a mental disease or defect predating the crime but did not find legal insanity at the time of the crime. If the record were incomplete with regard to this prejudice issue, we would remand for an evidentiary hearing.28 Our review of the state habeas hearing, however, discloses significant testimony and documentation regarding the factors the experts used in assessing Loyd's mental capacity. It is abundantly clear that significant portions of the experts' testimony were not based on the suggested unproven factors.

When questioned regarding the basis of his diagnostic conclusions, Dr. Honor stated that he based his analysis on a combination of his examination of Loyd in person, his meeting with Loyd's mother, and all of the available test and diagnostic data. Dr. Honor reported that the tests he performed indicated somatic delusions, disordered thinking, chronic psychological maladjustment with chronic disorientation, alienation, and withdrawal. Regarding the immediately preceding events, Dr. Honor assumed that Loyd had been drinking, but of other drugs he discussed only biochemical precipitators. Dr. Honor listed contributing factors: that Loyd apparently had little sleep, had been working long hours in the heat, had been arguing with his wife over financial difficulties, had encountered rejection, and had been depressed. He emphasized that no individual factor was determinative of his analysis, but that based on all of the data he diagnosed Loyd as a borderline psychotic with problematic personality traits of very longstanding duration. Dr. Honor also gave substantial independent weight to Loyd's description of events, as this description specifically matched what is known clinically about psychotic states. Dr. Honor testified that the information most pertinent to his diagnosis consisted of the letters Loyd wrote and his clinical interviews of Loyd.

Dr. Scanlon likewise gave significant weight to his interview with Loyd. He discussed the unproven factors as being "presumed" or "possible" and indicated his awareness that these factors might not be supported by the evidence.

Dr. Sanchez stated that he did not know whether Loyd was drugged. He recited hypothermia and fatigue as part of his factual basis, stating that no one trigger set Loyd off; rather, the stresses were cumulative. Both Dr. Scanlon and Dr. Sanchez testified that in their opinion Loyd was not faking and, in fact, that he could not fake his mental defect.

Discounting the unproven factors, the experts' fundamental opinions are largely based on Loyd's descriptions of the crime and on test scores and interviews. These experts, including Dr. Sanchez, the psychiatrist first selected by the prosecution, found that during the crime Loyd was experiencing the effects of substantial pre-existing mental defects. The absence of this mitigating evidence undermines our confidence in the outcome of Loyd's penalty phase. We are charged with "assess[ing] ... the likelihood of a result."29 The weighing of the defendant's mental condition was for the jury, which "must resolve differences in opinion within the psychiatric profession on the basis of the evidence offered by each party."30 As we noted in our earlier opinion,31 because of counsel's inadequacy the jury could not perform this function.

We now hold that there is enough evidence proving mental disease and defect that the balance of aggravating and mitigating factors in this case must be weighed by a jury in a new sentencing hearing.

The judgment of the district court is REVERSED and judgment is now RENDERED granting the requested writ of habeas corpus. The case is REMANDED for entry of an appropriate judgment issuing the Great Writ and directing the State of Louisiana, at its own option, to sentence Loyd to life imprisonment or to retry the sentencing phase of his trial within a reasonable period.