The State of Missouri appeals from the district court's grant of habeas corpus relief to Laws. The district court vacated Laws' death sentence after concluding that Laws had received ineffective assistance of counsel during the punishment phase of his trial. We reverse and reinstate the sentence, finding no valid claim of ineffectiveness of trial counsel.

We deal here with a defendant who conspired to and did commit three separate and distinct robberies and murders on elderly and helpless victims. He claims ineffectiveness here only in the penalty phase of the trial for the second set of murders.

The claim of ineffectiveness involves a criminal defense attorney whom the record describes as excellent and experienced. He had been licensed to practice law in Missouri since 1975. After serving as a law clerk in the Missouri Court of Appeals and an assistant prosecuting attorney in St. Louis County for two years, he entered private practice, where, according to his estimate, ninety-five percent of his work was in criminal law. He handled approximately fifty felonies a year, and at the time of the hearing, had tried between twenty-five and thirty jury trials, as well as having briefed and argued criminal appeals. He represented Laws in both trials: for the first murder (Elliott) and the two later murders (Williams). He spoke with Laws for six or more hours before the Elliott trial, and spent additional time with him before the Williams trial. He further estimated that he spent about four hundred hours preparing for the two murder trials, which involved substantial overlap in witnesses and preparation. The chief trial attorney for the St. Louis Public Defender's office testified at the ineffectiveness hearing that Laws' trial counsel had "an excellent reputation" as a defense lawyer in the community.

I. BACKGROUND.

In October of 1980, Leonard Laws was living in a two-room house trailer with George Clifton Gilmore, Gilmore's brother Norman, and several others in the Gilmore family. All three men were unemployed and impoverished. George Gilmore suggested to his brother and Laws that they could make money easily by robbing old people and then killing them to prevent identification. The trio bought shotguns and a rifle on October 8, 1980. This was one day after the Elliott murder (described in fn. 1), when they were flush with $4,600 cash from decedent Elliott's mason jar.

In the early morning of October 29, 1980, they went to the home of Clarence and Lottie Williams, whom the Gilmores had known through an uncle. Mr. Williams was eighty-three years old and his wife was eighty-one. The assailants entered the Williams' home after Laws cut the telephone line, and tied Mr. and Mrs. Williams up in chairs. Laws threatened to cut off their fingers if they would not reveal where they kept their money. They complied, and the three men ransacked the house. Mr. and Mrs. Williams were then untied and taken into the bedroom. George Gilmore prepared to kill them with his twelve-gauge shotgun, and Laws suggested instead: "Let me hit them in the head with a ball bat." Gilmore, however, told Laws to go stand outside and see if he could hear the shotgun blasts, and Laws complied. Gilmore then shot Mrs. Williams, reloaded the weapon, then shot Mr. Williams. The first shot failed to kill Mr. Williams, and he tried to run toward the front door, so Gilmore shot him a second time, killing him. Laws then reentered the house. After removing property, the trio poured fuel oil on the floor, which Laws ignited. The fire substantially destroyed the house.

II. PRIOR PROCEEDINGS.

Laws received a four-day trial in St. Louis County, Missouri Circuit Court in July of 1982. A jury found him guilty of two counts of capital murder. The punishment phase of the case began on the morning of July 23, 1982, before the same jury. During this phase the State introduced evidence, through certified copies of judgments, that Laws had been convicted of two separate capital murders in Missouri1 (for which he received life sentences), two armed robberies in Arizona (concurrent five to six-year sentences), and one aggravated assault in Mississippi (eleven-year sentence). Laws was imprisoned in March of 1982 in the Missouri State Penitentiary for his first Missouri murder conviction, and he has been incarcerated ever since. Laws had previously spent over six years in prison on the armed robbery and aggravated assault convictions before being paroled. Laws' lawyer introduced no evidence of mitigating circumstances.2 The trial court nonetheless gave the jury an instruction which directed it to consider mitigating circumstances.3 The jury returned a verdict calling for the death penalty. Laws' motion for a new trial was overruled on September 17, 1982, and he was sentenced to death on each count of capital murder. Laws appealed, represented by new counsel, but the conviction and sentence were affirmed in full. State v. Laws, 661 S.W.2d 526 (Mo.banc 1983), cert. denied, 467 U.S. 1210, 104 S. Ct. 2401, 81 L. Ed. 2d 357 (1984). Rehearing was denied on December 20, 1983. Laws then filed a pro se motion pursuant to Mo.S.Ct.R. 27.26, alleging ineffective assistance of counsel.4 New counsel was appointed for him, and on November 16, 1984, an evidentiary hearing was held on the Rule 27.26 motion before a different judge from the one who presided over Laws' trial. The court found no showing of incompetence by Laws' trial counsel, and Laws appealed. The Missouri Court of Appeals affirmed the denial of relief. Laws v. State, 708 S.W.2d 182 (Mo.App.1986). The court denied Laws' motions for rehearing and for transfer on March 25, 1986 and May 13, 1986. He then unsuccessfully petitioned the Supreme Court for certiorari. Laws v. State, 479 U.S. 871, 107 S. Ct. 246, 93 L. Ed. 2d 171 (1986).

Thus, after eight unsuccessful attempts in the state court system, four of them alleging ineffective assistance of counsel, Laws began the federal attack upon his sentence. He filed a pro se petition for writ of habeas corpus in the district court pursuant to 28 U.S.C. Sec. 2254. Counsel was appointed for him, and he raised two arguments in his amended petition.

He first argued that his court-appointed attorney's failure to offer evidence in mitigation during the punishment phase of his trial5 denied him effective assistance of trial counsel in violation of the sixth and fourteenth amendments. In support of this contention, he argued that significant mitigating evidence should have been investigated and presented. For instance, Laws had an honorable service record in Vietnam. Also, hospital psychiatric records and family testimony showed that his experiences in Vietnam caused him severe psychological damage.

Laws also argued that the imposition of the death penalty violated his eighth and fourteenth amendment rights because he was an accomplice with no direct role in the killing. Laws sought a new trial of the punishment phase of his trial only. He did not raise issues regarding his counsel's handling of the culpability phase of his trial. In an unpublished memorandum opinion, the district court granted Laws' habeas corpus petition and vacated his two consecutive death sentences, giving the State the opportunity to retry the punishment phase of the case.6

III. STANDARD OF REVIEW.

When considering an appeal from the granting of habeas corpus relief on the ground of ineffective assistance of counsel, this court may engage in its own independent review of the district court's conclusion, because the issue of ineffective assistance of counsel presents a mixed question of law and fact. Martin v. McCotter, 796 F.2d 813, 817 (5th Cir.1986), cert. denied, 479 U.S. 1057, 107 S. Ct. 935, 93 L. Ed. 2d 985 (1987); Reiger v. Christensen, 789 F.2d 1425, 1427 (9th Cir.1986). If a state court has rendered specific predicate factual findings, those findings should be presumed correct unless conditions exist which cast those findings into doubt. 28 U.S.C. Sec. 2254(d); Sumner v. Mata, 449 U.S. 539, 101 S. Ct. 764, 66 L. Ed. 2d 722 (1981); Martin, 796 F.2d at 817. The district court's findings of fact, however, are reviewable under the clearly erroneous standard. Fed.R.Civ.P. 52(a); Morrow v. Parratt, 574 F.2d 411, 413 (8th Cir.1978); see also Martin, 796 F.2d at 817.

IV. DISCUSSION.

The State points to a number of flaws in the district court's opinion. We find two of these flaws sufficient to mandate reversal.

First, although it is unclear whether the district court intended to do so, it appeared to set out an erroneous legal test that departs from the standard for determining ineffective assistance of counsel that the Supreme Court set out in Strickland v. Washington, 466 U.S. 668, 104 S. Ct. 2052, 80 L. Ed. 2d 674 (1984). Second, the district court's conclusion that Laws received ineffective assistance of counsel was not supported by the state court factual findings.

A. Erroneous Legal Test for Ineffectiveness of Counsel.

The State argues that the district court wrongly established a per se rule by holding that trial counsel is ineffective whenever a defendant receives the death penalty after counsel offers no evidence in mitigation. The State argues that such a per se rule flies in the face of both Strickland v. Washington, 466 U.S. 668, 104 S. Ct. 2052, 80 L. Ed. 2d 674 (1984), and Darden v. Wainwright, 477 U.S. 168, 106 S. Ct. 2464, 91 L. Ed. 2d 144 (1986), because in both cases no mitigating evidence was offered in the punishment phase, yet the Court upheld both death sentences. Having examined the district court's analysis and the applicable cases, we agree with the State.

The district court found that counsel's decision not to present mitigating evidence was based on: (1) his belief that a jury would not impose the death penalty since Laws had not directly participated in the killings; (2) his discussion with a juror from Laws' earlier capital murder case indicating that the jury had decided not to impose the death penalty because of a deal, also applicable to this case, the State had made with Gilmore; and (3) his phone conversations with one of Laws' brothers and a woman he thought was Laws' stepmother. The court found that these three factors did not represent:

a reasonably thorough investigation into petitioner's character and background or into other information potentially relevant to mitigating circumstances. The record does not reflect counsel's investigation of petitioner's military or psychiatric background for purposes of the punishment phase of the trial. Nor does the record reflect a responsible effort to glean available relatives' own knowledge of and relationship to petitioner * * *. Mem.Op. at 22.

Calling counsel's decision to offer no evidence at the penalty phase "a bold tactic," the court stated that: "Had it succeeded, its brilliance may have been unquestioned. Having failed, it must be considered beyond the bounds of professional norms." The court concluded:

Trial counsel may well believe he had reasonable grounds for the decision he reached in the death penalty phase of this cause. Nonetheless, that decision was wrong. * * * To a certain extent, we risk developing a body of law that says at the death penalty phase defense counsel must introduce some evidence or be found ineffective, while it is certainly possible and sometimes probable that putting on and offering evidence may worsen the client's chances. Perhaps this ruling stresses only the importance of fully investigating each case on its merits before deciding that as a matter of reasonable strategy, no evidence should be introduced. Mem.Op. at 25.

Although the district court's conclusion may purport not to create a new rule in this area of the law, we nonetheless think that the court's reasoning is inconsistent with Supreme Court precedent, specifically, the Strickland and Darden decisions.

The district court's analysis is incorrect because the test set out in Strickland for determining whether counsel has functioned effectively is not outcome-determinative. The fact of conviction alone does not necessarily indicate that a defendant received ineffective assistance of counsel. As Strickland and Darden make clear, the decision not to present evidence at the penalty phase is well within the range of practical choices that are not to be second-guessed, as long as they are based on informed and reasoned judgment.

Strickland involved a challenge to counsel's handling of the sentencing portion of a death penalty capital murder case. To prepare for the sentencing hearing before the judge, counsel spoke to the defendant about his background. He also spoke on the telephone with defendant's wife and mother, though he did not follow up on one unsuccessful effort to meet with them. He did not otherwise seek out character witnesses. Nor did he request a psychiatric examination, since his conversations with the defendant gave no indication that the defendant had psychological problems. Counsel decided not to present and hence not to look further for evidence concerning the defendant's character and emotional state. Counsel's rationale was, in part, that by foregoing the opportunity to present new evidence on these subjects, he prevented the State from cross-examining the defendant on his claim and from putting on psychiatric evidence of its own. At the sentencing hearing, counsel's strategy was based primarily on the trial judge's prior remarks and his reputation as a sentencing judge who thought it important for a convicted defendant to own up to his crime. Counsel argued that the defendant's remorse and acceptance of responsibility for his crime, his total lack of prior criminal conduct, the fact that he surrendered, confessed, offered to testify against a co-defendant, and the fact that he was a good person gone wrong under stressful circumstances all combined to justify sparing him from the death penalty. The strategy failed, and the trial judge sentenced the defendant to death.

The Supreme Court found that defendant's counsel had not been ineffective. Counsel had made:

a strategic choice to argue for the extreme emotional distress mitigating circumstance and to rely as fully as possible on respondent's acceptance of responsibility for his crimes. * * * Counsel's strategy choice was well within the range of professionally reasonable judgments, and the decision not to seek more character or psychological evidence than was already in hand was likewise reasonable. * * * The aggravating circumstances were utterly overwhelming. Trial counsel could reasonably surmise from his conversations with respondent that character and psychological evidence would be of little help. * * * On these facts, there can be little question, even without application of the presumption of adequate performance, that trial counsel's defense, though unsuccessful, was the result of reasonable professional judgment.

Strickland, 466 U.S. at 699, 104 S.Ct. at 2070-71. (Emphasis added.)

Darden v. Wainwright, 477 U.S. 168, 106 S. Ct. 2464, 91 L. Ed. 2d 144 (1986), presented a similar scenario. In Darden, a Florida defendant who had been sentenced to death for murder argued that he was denied effective assistance of counsel at the sentencing phase of trial. He argued that his team of trial counsel did not delve sufficiently into his background, and as a result were unprepared to present mitigating evidence at the sentencing hearing. Counsel knew that they were free to offer mitigating evidence, yet they chose not to do so. The Court, noting that Strickland requires judicial scrutiny of counsel's performance to be highly deferential, stated:

In this case, there are several reasons why counsel reasonably could have chosen to rely on a simple plea for mercy from petitioner himself. Any attempt to portray petitioner as a non-violent man would have opened the door for the State to rebut with evidence of petitioner's prior convictions. * * * In addition, * * * the State could have responded with a psychiatric report * * *. In sum, petitioner has not "overcome the presumption that, under the circumstances, the challenged action 'might be considered sound trial strategy.' " * * * Petitioner has failed to satisfy the first part of the Strickland test, that his trial counsels' performance fell below an objective standard of reasonableness.

Darden, 477 U.S. at 186-87, 106 S.Ct. at 2474-75, quoting Strickland, 466 U.S. at 689, 104 S.Ct. at 2065.

The Supreme Court recently reaffirmed this standard in Burger v. Kemp, 483 U.S. 776, 107 S. Ct. 3114, 97 L. Ed. 2d 638 (1987). In Burger, a Georgia jury found the defendant guilty of murder and sentenced him to death. The Supreme Court, noting the high degree of deference given to counsel on a claim of ineffectiveness, rejected the defendant's contention that his trial counsel had been ineffective for failing to develop and present any mitigating evidence at either of the two sentencing hearings.

In Burger, a search for mitigating evidence would have disclosed that the defendant had an unhappy and unstable childhood. Trial counsel was aware of some, but not all, of this family history before trial. After speaking with the defendant's mother, his psychologist, and others, counsel "made the reasonable decision that his client's interest would not be served by presenting this type of evidence." Burger, 107 S. Ct. at 3124. Counsel decided not to put the defendant's mother on the stand because her testimony might have been counterproductive, and the Court noted that "it was surely not unreasonable for [counsel] to have concluded that cross-examination might well have revealed matters of historical fact that would have harmed his client's chances for a life sentence." Id. at 3125. The Court concluded:

The record at the habeas corpus hearing does suggest that [counsel] could well have made a more thorough investigation than he did. Nevertheless, in considering claims of ineffective assistance of counsel, "[w]e address not what is prudent or appropriate, but only what is constitutionally compelled." United States v. Cronic, 466 U.S. 648, 665, n. 38, 104 S. Ct. 2039, 2050, n. 38, 80 L. Ed. 2d 657 (1984). We have decided that "strategic choices made after less than complete investigation are reasonable precisely to the extent that reasonable professional judgments support the limitations on investigation." Strickland, 466 U.S., at 690-691, 104 S.Ct. at 2066. Applying this standard, we agree with the courts below that counsel's decision not to mount an all-out investigation into petitioner's background in search of mitigating circumstances was supported by reasonable professional judgment. It appears that he did interview all potential witnesses who had been called to his attention and that there was a reasonable basis for his strategic decision that an explanation of petitioner's history would not have minimized the risk of the death penalty. Having made this judgment, he reasonably determined that he need not undertake further investigation to locate witnesses who would make statements about [the defendant's] past.

Id. at 3125-26.

Applicable circuit court decisions are in accord. See Lightbourne v. Dugger, 829 F.2d 1012, 1025-26 (11th Cir.1987); Robison v. Maynard, 829 F.2d 1501, 1504-05 (10th Cir.1987); Bell v. Lynaugh, 828 F.2d 1085, 1089-90 (5th Cir.), cert. denied, --- U.S. ----, 108 S. Ct. 310, 98 L. Ed. 2d 268 (1987); Brock v. McCotter, 781 F.2d 1152, 1160 (5th Cir.), cert. denied, 476 U.S. 1153, 106 S. Ct. 2259, 90 L. Ed. 2d 704 (1986).7

Laws points to a number of cases in which counsel was found ineffective for having failed to present mitigating evidence. In these cases, however, the key to the findings of ineffectiveness was not that mitigating evidence was not presented, but that counsel, as a result of inadequate preparation, had failed to discover the evidence. Contrary to the district court's apparent reasoning in this case, the absence of mitigating evidence does not inexorably lead to a conclusion of ineffective counsel. One must look further and determine why the mitigating evidence was not produced. If counsel has through neglect failed to discover such evidence, then counsel will be found ineffective. If, however, counsel did not produce the mitigating evidence because, after a reasonable investigation and exercise of professional judgment, he has determined that withholding such evidence is the more strategically sound course, then there has been no ineffectiveness.8

Kimmelman v. Morrison, 477 U.S. 365, 106 S. Ct. 2574, 91 L. Ed. 2d 305 (1986), cited by Laws, supports this view. In Kimmelman, the defendant was indicted for rape. Among the items of evidence used at trial was a stained bed sheet, which was obtained from defendant's apartment without a search warrant. At trial, counsel objected to the introduction of the sheet and testimony concerning it. State law, however, required that suppression motions be filed within thirty days of the indictment. Because the time limit had long since passed, the trial judge denied the motion. Ultimately, defendant was convicted. The Supreme Court observed that trial counsel "failed to file a timely suppression motion, not due to strategic considerations, but because, until the first day of trial, he was unaware of the search and of the State's intention to introduce the bedsheet into evidence." Kimmelman, 106 S. Ct. at 2588. Counsel was unaware of the search and seizure because he had failed to conduct any pretrial discovery. Again, this failure was not based on strategy, but on counsel's erroneous belief that the State was obligated to turn over inculpatory evidence to the defense before trial and that the victim's preferences would determine whether the State proceeded to trial after an indictment had been returned. The Court found that counsel's failure to conduct any discovery was "unreasonable, that is, contrary to prevailing professional norms. * * * Respondent's lawyer neither investigated, nor made a reasonable decision not to investigate, the State's case through discovery." Counsel had a "duty to make reasonable investigations or to make a reasonable decision that makes particular investigations unnecessary." Kimmelman, 106 S. Ct. at 2588-89.

Laws also cites Morrow v. Parratt, 574 F.2d 411 (8th Cir.1978); Eldridge v. Atkins, 665 F.2d 228 (8th Cir.1981), cert. denied, 456 U.S. 910, 102 S. Ct. 1760, 72 L. Ed. 2d 168 (1982); and Woodard v. Sargent, 806 F.2d 153 (8th Cir.1986). As in Kimmelman, however, these cases stand for the proposition that inadequate trial preparation and investigation equal ineffective assistance of counsel, not the fact, standing alone, that potentially mitigating evidence is not introduced. In Morrow, the court found that the defendant's lawyer had failed adequately to investigate the facts before advising his client to plead guilty and had failed to interview eyewitnesses to the crime. The court concluded: "In summary, the evidence described above could have been uncovered by a reasonable investigation by Morrow's attorney." Morrow, 574 F.2d at 414.

In Eldridge, defense counsel did not interview any of the eyewitnesses to the robbery. "Counsel did not investigate because he conclusorily decided that Eldridge was guilty and that further investigation would not do any good." Eldridge, 665 F.2d at 235. Noting also that counsel's failure to use certain witnesses was not so much trial strategy as it was an accommodation to his own inadequate trial preparation, the court concluded that: "Counsel need not attain perfection, but he must exercise reasonable diligence to produce exculpatory evidence." Id.

Finally, in Woodard, a capital murder case, counsel was found ineffective for failing to request a jury instruction on a new mitigating circumstance which had been added to the pertinent statute shortly before the case was tried. The jury found two aggravating circumstances and no mitigating circumstances. Accordingly, the court found:

A finding of a mitigating circumstance should have been an important objective in Woodard's case, and the failure to seek the inclusion of this obvious mitigating circumstance certainly fell below the threshold of reasonably competent assistance. * * * [W]e can conceive of no possible tactical reason for such an omission.

Woodard, 806 F.2d at 157.

As these cases show, it is not the failure to present mitigating circumstances, but rather the reasons for the failure to present mitigating circumstances, that are the proper subject for judicial scrutiny. Therefore, we hold that the district court erred in setting out a per se rule in this regard.

B. Whether the District Court's Assessment of the Evidence was Erroneous.

Even if the district court had not created a framework for analysis that departs from established Supreme Court precedent (see supra section IV(A) of this opinion), we think the district court's assessment of the evidence, even under the correct standards, would mandate reversal. The district court concluded, applying the Strickland standards, that counsel had made an inadequate investigation and was thus ineffective. We believe this conclusion was contrary to the evidence. Presuming as we must the state court's factual findings to be correct, as required by 28 U.S.C. Sec. 2254(d),9 we are compelled to conclude that Laws received effective assistance of counsel.

Laws' arguments which the district court considered are as follows:

(1) Trial counsel failed adequately to investigate or introduce any evidence concerning Laws' mental state;

(2) Trial counsel failed to contact members of Laws' family so that they could testify on his behalf; and

(3) Trial counsel failed to present testimony from a statistician, priest and/or psychologist in opposition to the imposition of the death penalty.

We consider each issue in turn. Our analysis is governed by Strickland v. Washington, 466 U.S. 668, 104 S. Ct. 2052, 80 L. Ed. 2d 674 (1984). Under Strickland, a petitioner alleging ineffective assistance of counsel must prove two independent components:

First, the defendant must show that counsel's performance was deficient. This requires showing that counsel made errors so serious that counsel was not functioning as the "counsel" guaranteed the defendant by the Sixth Amendment. Second, the defendant must show that the deficient performance prejudiced the defense. This requires showing that counsel's errors were so serious as to deprive the defendant of a fair trial, a trial whose result is reliable. Unless a defendant makes both showings, it cannot be said that the conviction or death sentence resulted from a breakdown in the adversary process that renders the result unreliable.

Strickland, 466 U.S. at 687, 104 S.Ct. at 2064.

Turning first to whether counsel's performance was deficient, we note that the defendant must show that counsel's representation fell below an objective standard of reasonableness, considering all the circumstances. We recognize that our scrutiny of counsel's performance must be highly deferential, in view of the wide latitude counsel must have in making tactical decisions. We must "indulge a strong presumption that counsel's conduct falls within the wide range of reasonable professional assistance; that is, the defendant must overcome the presumption that, under the circumstances, the challenged action 'might be considered sound trial strategy.' " Strickland, 466 U.S. at 689, 104 S.Ct. at 2065.

With these standards in mind, we examine Laws' three claims.

1. Failure Adequately to Investigate Laws' Psychological Background.

Laws makes the argument, maintained throughout the state court post-conviction proceedings, that his trial counsel failed adequately to investigate his past psychiatric treatment record and to explore a possible psychologically-based defense. In this regard, Laws claims that his psychiatric condition was heavily influenced by his service in Vietnam, and argues that his trial counsel was ineffective for failing to present mitigating evidence relating to his Vietnam experience and its effect on him.10 Laws had undergone a psychiatric examination in the Mississippi State Hospital in 1974-75 while serving time in the Mississippi Penitentiary on an aggravated assault charge. Trial counsel, however, relied on a psychiatric examination of Laws conducted on September 22, 1981, at the Fulton State Hospital in connection with one of the two other capital murder charges which had been brought against Laws. This exam was much closer in time to the acts of murder and trials than the five-year old Mississippi exam. Further, the doctor at Fulton took into consideration the Mississippi exam. The Fulton State Hospital examination, which took into account Laws' prior Mississippi examination, concluded that Laws was mentally capable of standing trial, and that he was able to know and appreciate the nature, quality and wrongfulness of the murders when he committed them and had been capable of conforming his conduct to the requirements of the law. The record regarding this issue at the Rule 27.26 hearing is as follows:

Q. What were the results of this examination that you were aware of before the trial?

A. [Laws' trial counsel] The findings were, from Roman Numeral VIII, that the accused had no mental disease or defect within the meaning of Section 552.010, and that Leonard had the capacity to understand the proceedings against him and knew and appreciated the nature and wrongfulness of his conduct.

Q. And do they also, on page four of this document, talk about an evaluation in Mississippi and what the diagnosis there was?

* * *

* * *

A. It states that he was given the diagnosis of anti-social personality and situational reaction of adult life.Q. So the doctor in the examination you reviewed also considered this Mississippi evaluation, is that right?

A. Yes.

Q. Now, you said besides this examination in Defendant's Exhibit B, you also considered Leonard Laws, his background, and your conversations with him. And what did you glean from that, if anything?

A. Well, in my conversations with him I didn't detect--again, its a lay opinion. But I didn't detect any mental disease or defect that would interfere with our preparing to go to trial, or for his standing trial.

Q. All right. And so was that ever a plan, or a possible defense in your mind, a viable defense in any way to claim any mental disease or defect, based on your evaluation of this case?

A. Well, I considered it, but I rejected it.11

(Emphasis added.)

Q. Did Leonard at any time ever insist that he should be evaluated, or that he was insane in any way?

A. I don't recall that, no.

[Rule 27.26 Tr. at 73-74.]

* * *

* * *

Q. [D]id you read through that psychiatric report when you represented Mr. Laws?

A. While I represented him, yes.

Q. And in that report, do you recall the term "situational reaction," that Mr. Laws suffered from problems with situational reactions?

A. Yes.

Q. And did you speak to the psychiatrist who prepared that report about what a situational reaction was?

A. No. I didn't.

Q. And at this time, do you know what a situational reaction is?

A. A situational reaction does not rise to the level of mental disease or defect. I know that.

Q. But you did not investigate, at the time you represented Mr. Laws, exactly what a situational reaction was, what the scope of that was?

A. What the scope of it was, I have an understanding what it is. But I don't know the actual scope of it. And again, I read it. It was in the context of the individual's finding that there was no mental disease or defect.

(Emphasis added.)

Q. Right. But when you represented Mr. Laws and read this report, did you at any time talk to a psychiatrist or psychologist about what types of incidents aggravate situational reactions?

A. No, I didn't.

[Rule 27.26 Tr. at 83-84.]

The Rule 27.26 court, after hearing this testimony, found as fact that Laws:

never gave any indication to his trial attorney or the trial court that he was mentally incompetent at the time of the trial herein or at the time of the offenses. In fact [Laws] had been examined on September 22, 1981, prior to the trial with respect to a third murder case against [Laws] in Washington County. The results of that examination show that [Laws] had no neurosis or psychosis.

The court then concluded that "[t]here was no indication or evidence which required a new mental examination of [Laws]. Mental examination conducted pursuant to Chapter 552, further shows that [Laws] had no significant mental problems."

In reviewing the Rule 27.26 transcript and findings and the conclusions of the Rule 27.26 court, the Missouri Court of Appeals found that "[c]lose examination of the transcript reveals that [trial counsel] made a reasoned strategic decision based on all the circumstances to refrain from putting on evidence at trial of [Laws'] psychiatric state." Laws v. State, 708 S.W.2d 182, 185 (Mo.App.1986). The court concluded:The evidence at movant's 27.26 hearing convinces us that [trial counsel] made a careful and conscious choice against putting on psychiatric evidence during trial or at the penalty stage. Nothing made known to counsel during his representation of movant, and extensive interviews with him, suggested to counsel that presenting evidence of movant's psychiatric state would be of any benefit in his defense. Counsel had a duty to make reasonable investigations or to make a reasonable decision that makes a particular investigation unnecessary. Strickland v. Washington, 466 U.S. 668, 104 S.Ct. at 2067. Counsel could have well reasoned that the jury would be more harsh with movant as a result of learning that his behavior could be erratic and violent. Contrary to movant's contention that counsel failed to ascertain the meaning and import of a "situational reaction," our review of the record leads to the conclusion that counsel was fully cognizant of the nature of this disorder and decided as a matter of sound trial strategy that it would not be helpful.

[Counsel's] informed choice of trial strategy to refrain from presenting psychiatric evidence at trial or during the penalty stage rendered it unnecessary to engage in further inquiry into movant's psychiatric state.

Id. at 186.

The district court found, however, that trial counsel's decision not to present psychiatric evidence "was not based on a reasonably thorough investigation into [Laws'] character and background or into other information potentially relevant to mitigating circumstances. The record does not reflect counsel's investigation of [Laws'] military or psychiatric background for purposes of the punishment phase of the trial." Mem.Op. at 22.

We believe, however, that the record reveals that counsel acted well within the bounds of professional competency. The state court findings of fact were entitled to a presumption of correctness pursuant to 28 U.S.C. Sec. 2254(d). We accept the state courts' legal conclusion of effectiveness based on those facts.

Counsel could easily have concluded that the decision to present psychological testimony would have disastrous consequences. Skillful cross-examination could have revealed that although Laws suffered from no mental impairment that would negate responsibility for the murders he had committed, he was nonetheless a maladjusted man with a propensity for violence. In view of the fact that Laws' psychological history revealed to counsel that he had no mental disease or defect, as well as the fact that in their interviews, Laws appeared normal and made no mention of any psychological impairment, for counsel to have nonetheless raised the issue of Laws' mental state at trial could have induced cross-examination which would make counsel appear devious or unprepared. Such cross-examination could have alienated the jury against his client.

The record further reveals that counsel was well aware of Laws' Vietnam service. In an affidavit prepared in support of his federal habeas petition, Laws stated: "During my trial preparation for the Williams case, [trial counsel] and I discussed very briefly my service in the army and in Vietnam from 1967 to 1970. I told him I had an honorable discharge. He never asked me about it again." During his closing argument in the punishment phase of Laws' trial, Laws' counsel mentioned that Laws had spent time in Vietnam: "when I was in Vietnam, I killed somebody. Okay, I killed somebody. I had to. Leonard Laws was in Vietnam. I killed somebody in those circumstances and I can remember the day, March 17th." Tr. Vol. II at 808.

We also note that the Missouri trial judge who sentenced Laws was aware that Laws had been in Vietnam. The trial judge had before him the report of the presentence investigation on Laws, which revealed that Laws had spent time in Vietnam and had been honorably discharged from the United States Army. The sentencing judge was therefore able to take into account Laws' Vietnam experience when rendering his decision.

Although counsel was cognizant of Laws' time spent in Vietnam, it was nonetheless reasonable for him to forego presentation of this issue. Laws argues that testimony could have been elicited, as a result of psychological tests and family interviews, as to the effect of Vietnam upon his mental state, including facts about his personality before and after his military service. Introduction of these facts, however, would have opened the door to cross-examination as to whether Laws' disturbed mental state resulted not from his time in Vietnam but from his more than six years in prison. In sum, family members claimed testimony would be essentially limited to the fact that he had changed after Vietnam. Laws also argues that mitigating evidence could have been presented concerning his service record, including facts about his combat experience, his citations received for duty, and his good discharge record. As the State points out, however, to introduce evidence of Laws' record in Vietnam would have opened the door to cross-examination revealing prejudicial facts about Laws' drug usage while in the service, an auto theft charge, an AWOL conviction, and the lack of substantiation of his claimed war injury.

In sum, we conclude that under Strickland, counsel conducted an adequate investigation into a possible psychologically-related defense. In view of the fact that the clinical evidence did not support a conclusion that Laws had a mental disease or defect, there was no reason for counsel to investigate further into Laws' psychological background. Moreover, it was not unreasonable for counsel to conclude that the potential risks of a defense based on Laws' Vietnam experience outweighed the potential benefits. Counsel's performance in this regard cannot be considered deficient.

2. Failure to Contact and Interview Relatives.

Laws argues that his counsel erred in failing adequately to pursue an investigation of relatives who would be willing to testify on Laws' behalf. Again, this argument is contradicted by the facts established in the Rule 27.26 hearing. The hearing transcript reveals the following:

Q. Now, other than George Gilmore, did Leonard Laws also request that you call some members of his family to testify, at least to the death penalty phase of that, if it came to that?

A. [Laws' trial counsel] Yes, he did. He gave me a phone number of an individual which was either his father's trailer or his stepmother's trailer. I am not sure which. And it was in rural Missouri. I am not sure where it was. And I called numerous times.

Q. Did you get ahold of anyone?

A. I finally did, yes. I finally got ahold of the woman. And I don't recall her name right now, I'm sorry. But the woman who Leonard gave me her phone number. And it was late at night one night when I happened to catch her. And I explained to her that we wished her to testify perhaps in the death phase of the case.

Q. What was her response?

A. She refused.

(Emphasis added.)

Q. Did she give you any reason why she or Leonard Laws' family might be upset with him and why they wouldn't want to testify?

A. She stated that she didn't wish to help Leonard in this case, and she didn't--she knew that the other family members didn't wish to help. I believe I spoke with one of his brothers, too. I am not sure of that. I just saw that that was amended today, and I didn't have an opportunity to look over my records. But I believe I spoke with one of his brothers.

(Emphasis added.)

Q. But in any case, you made some efforts, but you couldn't get any of them to testify?

A. In the death phase, they were most adamant about not testifying.

(Emphasis added.)

Q. But you pursued that as being a possible advantage to have?A. In any death case, you wish to have someone, a relative to testify, to show that he has people who care enough to come in and testify.

[Rule 27.26 Tr. at 77-78.]

The Rule 27.26 court found as fact that trial counsel "contacted relatives of [Laws] by calling a phone number [Laws] supplied. Two different persons told [trial counsel] that the relatives would not testify on Leonard Laws's behalf." The court concluded that there was no deficiency on this ground, and the Missouri Court of Appeals agreed:

After multiple attempts to obtain relatives to testify for movant at the penalty hearing [counsel] was able to contact a woman he believed to be movant's step-mother and a brother. Both persons stated unequivocably that they would not testify at the penalty hearing and that other relatives would also not testify. [Laws] himself testified that he personally wrote his brothers and father requesting them to testify on his behalf, but they failed to appear for trial. Furthermore, [Laws] did not present family members as witnesses at the evidentiary hearing. Under the circumstances, we find that counsel engaged in an adequate investigation of movant's relatives, and that no favorable information appeared as a result. * * * Counsel was well aware of the helpfulness of favorable testimony by relatives during the penalty stage, as the record discloses, but unfortunately was faced with a situation where reasonable efforts revealed that movant's relatives were strongly opposed to testifying. The decision not to call them as witnesses was thus a matter of trial strategy.

Laws, 708 S.W.2d at 187.

Again, we find the state courts' conclusions compelling.12 The district court, however, concluded that the record did not "reflect a responsible effort to glean available relatives' own knowledge of and relationship to petitioner." Mem.Op. at 22. We disagree with the district court's conclusion. We think that counsel's efforts in this regard were objectively reasonable.

Trial counsel could have reasonably determined that calling Laws' relatives to the stand, after what they had told him of their feelings for Laws, would have been destructive to Laws' defense. In view of the relatives' adamant refusal to testify favorably on Laws' behalf, cross-examination would not even have been necessary in order to hurt Laws' case. The simplest questions asked on direct examination would have revealed the relatives' total lack of support for Laws. As a result, the jury more likely than not would have developed animosity toward Laws and would have distrusted and disrespected counsel's efforts. Counsel's decisions were within the wide range of professionally competent assistance approved by Strickland.

3. Failure to Elicit Testimony of Community Members.

Laws argues that his trial counsel was derelict in failing to call to the stand, or consider calling to the stand, a clergyman, statistician or professor to testify against the death penalty. The State first responds that this claim of error is subject to an adequate and independent state procedural bar precluding review by this court, because Laws failed to present this claim on appeal in the state courts. Laws presented the claim in his amended Rule 27.26 motion, but the Missouri trial court held it to be without merit. Laws, however, when appealing from the denial of Rule 27.26 relief, did not brief this allegation of error as a ground for appeal. We nonetheless conclude that this claim is properly before us for the following reasons. Laws filed a pleading with the Missouri Court of Appeals entitled "Appellant's Issues Expected To Be Raised On Appeal," which stated that "The trial court erred in failing to grant appellant's 27.26 motion for failure to investigate and prepare defense witnesses for the penalty stage." Laws argues, and we accept, that trial counsel's failure to elicit testimony from a clergyman, statistician or scholar was included in this general assignment of error. This pleading provided notice to the court of appeals that Laws sought review of this particular point. Furthermore, the court of appeals reviewed the findings of fact and conclusions of law with regard to this point in accordance with Missouri Supreme Court Rule 27.26(j). The court stated in relevant part:

On appeal, movant raises two points of error. First, movant contends that the trial court failed to make specific findings of fact and conclusions of law regarding his claim of ineffective assistance of counsel based on defense counsel's decision to refrain from putting on evidence during the penalty stage in mitigation of the death penalty.

* * *

* * *

Movant presented the testimony of several witnesses, including his trial counsel, concerning the advisability of putting relatives and clergymen on the stand during the penalty stage of a capital murder trial to testify in mitigation of the death penalty. This testimony was received without objection by the state. Under the circumstances, we find that the issue was tried by the implied consent of the parties, and thus, will be treated as amending the Rule 27.26 motion to conform to the evidence.

Laws v. State, 708 S.W.2d at 184. (Emphasis added.)

Therefore, Laws did not waive his right to raise the failure of trial counsel to elicit testimony from a clergyman, statistician or professor in support of his ineffective assistance of counsel claim. In addition, the issue has been exhausted at the state level. "The exhaustion doctrine requires only that the state courts have one full and fair opportunity to decide a question which is properly presented to it." Mucie v. Missouri State Department of Corrections, 543 F.2d 633, 636 (8th Cir.1976), citing Picard v. Connor, 404 U.S. 270, 92 S. Ct. 509, 30 L. Ed. 2d 438 (1971). Since the Missouri Court of Appeals ruled adversely to Laws on this point and the Missouri Supreme Court denied a motion for rehearing and/or transfer, Laws' state remedies as to this point have been exhausted, and it is properly before this court.

At the Rule 27.26 hearing, the following colloquy occurred as to this issue:

Q. Did you ever consider putting on any type of priest or some religious personage to testify?

A. [Laws' trial counsel] Yes, I considered that. We had used that, two individuals of that ilk in the first trial [for the murder of Woodrow Wilson Elliott] to testify in the death phase.

Q. And what about the second trial? Why was that not--did that not come to pass?

A. After the first trial I spoke with the jurors in the case. And they told me that they had disregarded totally the testimony of these individuals. I believe the gentleman's name is Mr. Gilsonin.

Q. He was a statistician, or professor?

A. And I don't recall the other. I don't recall the other gentleman's name right now.

Q. But the one was a priest and one was a statistician?

A. Right. And the jurors said that they rejected out of hand that testimony and instead centered on the deal between Norman Gilmore and the State in their finding of fifty years rather than death in the first case.

Q. And so that influenced you as far as your trial strategy in the second case?

A. It did. And the fact that the facts in the second case were much different than the facts in the first case.

Q. You mean as far as who the actual killer was?

A. Right.

[Rule 27.26 Tr. at 78-79.]

In considering this testimony, the Rule 27.26 court made the following factual findings:

[Trial counsel] was fully informed at the time of trial of the possibility of calling clergy persons and professors to give their views on the death penalty. [Counsel] decided after due consideration and after calling such persons in the first trial of Leonard Laws that their testimony would not be beneficial to the case.

The court concluded in this regard: "Failure to call witnesses * * * was an appropriate trial strategy."

The district court noted that this issue was not as significant as the others, in part because the district court recognized that counsel had used two "community members" at Laws' prior criminal trial and thus had the opportunity to weigh their credibility and presentation in deciding whether or not to use them again. The district court also noted that the Missouri Supreme Court had recently found such testimony irrelevant. State v. Gilmore, 681 S.W.2d 934, 941 (Mo.banc 1984). The district court concluded, however, that "a discussion with one of twelve jurors as to why the jury reached its decision seems an insufficient basis for a tactical decision in a trial before a different jury--particularly where the consequence of the decision may be fatal." Mem.Op. at 23 n. 2.

We note that the record reflects that counsel spoke with more than one of the prior jurors. In any event, we believe that talking to jurors regarding the success of a strategy suggests not deficient representation but keen lawyering. Trial counsel was far better able to evaluate the potential impact of community member testimony after having seen how successful it had been than if he had made the decision to use or reject such testimony in a vacuum.

Counsel's choice was solidly grounded in practical concerns and experience. A trial lawyer does not try each new case with his mind a blank slate; rather, he brings to the case the sum of his knowledge and experience gained through previous trials. This is what distinguishes the veteran litigator from the recent law school graduate. Laws' trial counsel had previously defended Laws for the Elliott murder, see supra note 1, and he took the opportunity to find out what had been effective and what had not. He was told directly that the jurors had put no credence in the testimony of community members. Relying on this information and his practical experience in the previous case, counsel made the reasonable decision not to present what he considered to be useless testimony in this case.

Counsel had pursued substantially the same strategy during the punishment phase of the Elliott trial as he used here. In the Elliott trial, the strategy was successful, because Laws received a life sentence rather than the death penalty. Laws had no complaint with counsel's strategy at the Elliott trial. Counsel then used the essential ingredients of his previously successful strategy in this trial, while refining the strategy to omit the testimony of community members based on the lessons learned from the prior trial. In short, for counsel to repeat a winning strategy in a later trial for the same type of crime, involving the same defendant, strikes us as eminently reasonable.

Accordingly, we hold that trial counsel's choice not to present the testimony of community members was intelligent, tactically sound trial strategy. Laws' claim on this ground thus fails the first Strickland prong, which requires that counsel's performance be shown deficient.

V. CONCLUSION.

In examining counsel's performance, we do not use 20-20 hindsight. As this court stated in Blackmon v. White, 825 F.2d 1263 (8th Cir.1987):

[T]he courts must resist the temptation to second-guess a lawyer's trial strategy; the lawyer makes choices based on the law as it appears at the time, the facts as disclosed in the proceedings to that point, and his best judgment as to the attitudes and sympathies of judge and jury. The fact that the choice made later proves to have been unsound does not require a finding of ineffectiveness. The petitioner bears the burden of successfully challenging particular acts and omissions of his attorney which were not the result of reasonable professional judgment; it is not enough to complain after the fact that he lost, when in fact the strategy at trial may have been reasonable in the face of an unfavorable case.

Blackmon, 825 F.2d at 1265 (Emphasis added.)

We believe that counsel's strategy was indeed reasonable in the face of this highly unfavorable case. Accordingly, Laws has failed to show that his trial counsel's performance deprived him of his sixth amendment right to effective assistance of counsel.

For the foregoing reasons, the judgment of the district court is reversed.

McMILLIAN, Circuit Judge, dissenting, with whom LAY, Chief Judge, and HEANEY and ARNOLD, Circuit Judges, join.

I respectfully dissent. I would affirm the district court's judgment granting the writ of habeas corpus because appellant did not receive effective assistance of counsel at the penalty phase of his trial. Laws v. Armontrout, No. 86-2214C(3), Slip op. at 23 (E.D.Mo. Dec. 15, 1986) (memorandum opinion). I would therefore remand the case to the state court for a new trial on the penalty phase only.

This is a death penalty case. It is readily acknowledged that "the representation of a person charged with a capital offense imposes the heaviest professional responsibility known to the practice of law." Johnson v. Cabana, 818 F.2d 333, 340 (5th Cir.1987). Yet appellant's counsel failed to conduct a reasonable investigation into appellant's background--an investigation that would have led to admissible mitigating evidence at the penalty phase.

After an extended hearing, an experienced district court judge found that counsel's representation fell far below the standard of reasonableness and prejudiced appellant. Specifically, the district court found that counsel (1) failed to investigate and present evidence about appellant's military background and the psychological effect his service in Vietnam had had on him, (2) failed to make reasonable efforts to contact appellant's family members and to present relevant evidence from them, (3) failed to investigate and present evidence of appellant's prior psychological evaluation, and (4) failed to present any evidence as to the inappropriateness of the death penalty. Relying on its finding and applying the standards set forth in Strickland v. Washington, 466 U.S. 668, 686, 104 S. Ct. at 2052, 2063-64, 80 L. Ed. 2d 674 (1984) (Strickland ), the district court concluded that there was a reasonable probability that, but for counsel's ineffective representation, the result of the proceedings would have been different. Id. at 23. I agree.

In Strickland the Supreme Court articulated the following standard for determining whether assistance of counsel has been ineffective: "the benchmark for judging any claim of ineffectiveness must be whether counsel's conduct so undermined the proper functioning of the adversarial process that the trial cannot be relied on as having produced a just result." 466 U.S. at 686, 104 S.Ct. at 2063-64.

To succeed on a claim of ineffective assistance of counsel, a convicted defendant must show both that the counsel's performance was deficient and that the deficient performance prejudiced the defendant. Id. at 687, 104 S. Ct. at 2064. This is the same test used by the Eighth Circuit Court of Appeals in cases decided before Strickland. See Pickens v. Lockhart, 714 F.2d 1455, 1460 (8th Cir.1983). The test of deficient performance is one of reasonableness. We do not deal in absolutes, but rather test counsel's performance by looking to "the skill and diligence that a reasonably competent attorney would exercise under similar circumstances." Thomas v. Lockhart, 738 F.2d 304, 307 (8th Cir.1984). Likewise, the test for prejudice is not absolute.

Strickland defines "prejudice" as "a reasonable probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different. A reasonable probability is a probability sufficient to undermine confidence in the outcome." 466 U.S. at 694, 104 S.Ct. at 2068.

The majority faults the district court for relying on the wrong legal test--the outcome determinative test. This assessment of the district court's analysis is wrong for two reasons. First, the district court clearly recognized and relied on the Strickland test. Slip op. at 20. Second, the district court found "that the outcome of the punishment phase of the case would have been different absent counsel's failure to investigate fully and to present mitigating evidence." Id. at 23. Thus, the district court concluded that appellant's evidence met an even more stringent test than that required by Strickland.

Strickland recognized that the outcome determinative test places a heavier burden on an appellant than the test it articulated: "with regard to the prejudice inquiry, ... the strict outcome determinative test ... imposes a heavier burden on defendants than the tests laid down today." 466 U.S. at 697, 104 S.Ct. at 2069-70. Therefore, because appellant's evidence met the outcome determinative test as the majority acknowledges, the evidence surely met the lesser Strickland test of "probability sufficient to undermine confidence in the outcome." Id. at 694, 104 S. Ct. at 2068.

The majority believes that the district court wrongly established a per se rule that defense counsel is ineffective whenever a defendant receives the death penalty after counsel offers no evidence in mitigation. However, the majority misreads the opinion of the district court. In fact, the district court specifically recognized that its opinion might be so construed and specifically rejected this approach:

To a certain extent, we risk developing a body of law that says that at the death penalty phase defense counsel must introduce some evidence or be found ineffective, while it is certainly possible and sometimes probable that putting on and offering evidence may worsen the client's chances. Perhaps this ruling stresses only the importance of fully investigating each case on its merits before deciding that as a matter of reasonable strategy, no evidence should be introduced.

Slip op. at 25. (emphasis added).

Where the defendant's life may be the penalty exacted by the state, I do not believe we ask too much of defense counsel that he or she first undertake a reasonable investigation so that he or she can make and put before the court the best defense available to the defendant. Strickland specifically recognized that:

[c]ounsel has a duty to make reasonable investigations or to make a reasonable decision that makes particular investigations unnecessary. In any ineffectiveness case, a particular decision not to investigate must be directly assessed for reasonableness in all the circumstances....

466 U.S. at 691, 104 S. Ct. at 2066 (emphasis added). The Supreme Court, in taking counsel to task for inadequate preparation, said:

[A] complete lack of pretrial preparation puts at risk both defendant's right to an "ample opportunity to meet the case of the prosecution" and the reliability of the adversarial testing process.

Kimmelman v. Morrison, 477 U.S. 365, 385, 106 S. Ct. 2574, 2588-89, 91 L. Ed. 2d 305 (1986) (citations omitted).

Applying the standards and holdings of Strickland and Kimmelman to the present case, it is clear that appellant received ineffective assistance of counsel and was prejudiced thereby. The duty to advocate the defendant's cause, including consulting with the defendant and bringing to bear such knowledge as will render the trial a reasonable adversarial process, Strickland, 466 U.S. at 668, 104 S.Ct. at 1052, mandates that defense counsel investigate and develop facts about the defendant's background in an attempt to develop mitigating evidence. Once counsel has completed such an investigation, then, and only then, would he or she be in a position to reasonably decide whether or not to present the evidence as a matter of trial strategy.

In the present case, trial counsel had a telephone conversation with someone claiming to be appellant's stepmother; he states that this person indicated that none of appellant's relatives wished to testify for appellant. Counsel also had a conversation with Laws' brother, whom he never even asked to testify. Based on these two sketchy conversations, counsel failed to contact other family members and failed to develop mitigating evidence he could have used at the penalty phase of Laws' trial.

The majority relies primarily on the presumption of correctness to be accorded a state court's finding of fact as the basis for its decision. While I do not dispute this proposition as a general legal proposition, the presumption of correctness does not apply to the state court's determination of the effectiveness of counsel. In Sumner v. Mata, 455 U.S. 591, 597, 102 S. Ct. 1303, 1306-07, 71 L. Ed. 2d 480 (1982), the Supreme Court held that the presumption of correctness accorded the factual findings of the state court under 28 U.S.C. Sec. 2254(d) applies only to historical facts underlying counsel's performance but not to the ultimate conclusion whether effective assistance has been rendered. This latter conclusion is not a finding of fact binding on a federal court because, by its very nature, the issue of effective assistance is a mixed question of law and fact. Thomas v. Lockhart, 738 F.2d at 307.

Thus, the district court had the duty to carefully examine the conclusion of the state court and to make "every effort ... to eliminate the distorting effects of hindsight, to reconstruct the circumstances of counsel's challenged conduct and to evaluate the conduct from counsel's perspective at the time." Id. at 309 (citations omitted); see Kellogg v. Scurr, 741 F.2d 1099, 1101 (8th Cir.1984). Moreover, defense counsel may not immunize his or her performance from sixth amendment scrutiny merely by labeling acts or omissions as trial strategy. Kellogg v. Scurr, 741 F.2d at 1102.

I do not take issue with the state court's determination of historical facts. I accept for purposes of this analysis that defense counsel in fact did what he testified he did: he reviewed the 1981 psychiatric examination but did not consider the 1974 psychiatric examination and did not speak to the psychiatrist to determine the significance of appellant's diagnosis; he spoke to a woman he believed to be appellant's stepmother and may have spoken to appellant's brother but could not remember their names; and he talked to one (or perhaps two) jurors to determine whether the testimony of members of the community on behalf of appellant was effective.

It is these facts admitted by counsel (and accepted by the state court) that the district court properly subjected to an independent evaluation to determine if there was effective assistance of counsel. It is this independent evaluation which the majority wrongly rejected. The majority incorrectly applies the presumption of correctness not only to the historical facts found by the state court, but also to the state court's conclusion that effective assistance was provided. See Sumner v. Mata, 455 U.S. at 597, 102 S.Ct. at 1306-07; Kellogg v. Scurr, 741 F.2d at 1102.

The majority begins its analysis with a recital of the many past accomplishments of defense counsel. However, the focus of inquiry concerning counsel's possible ineffectiveness must be "the reasonableness of counsel's challenged conduct on the facts of the particular case, viewed as of the time of counsel's conduct." Strickland, 466 U.S. at 690, 104 S.Ct. at 2066. For that reason, it is appropriate to judge counsel's effectiveness by looking solely to what he knew in this case and what he failed to find out because of his failure to investigate.

The majority credits defense counsel with being "well aware of Laws' Vietnam service." Laws had told counsel that he had served in Vietnam from 1968-1970 and that he had been honorably discharged. During closing argument, defense counsel even acknowledged that he himself had been in Vietnam. Far from excusing counsel's failure to further investigate, however, this knowledge makes his inaction all the more reprehensible.

Much has been written about the severe emotional impact suffered by some of the soldiers who fought for this country during the Vietnam Era. Congress has recognized the socialization and adjustment problems of Vietnam veterans by passing the Veterans Health Care Amendments Act in 1979. 38 U.S.C. Sec. 612 et seq. This act authorizes the Veterans Administration to provide services to Vietnam Veterans suffering from post-traumatic stress disorder after their return to civilian life. The legislative history of this act states:

According to testimony from psychiatrists and other physicians, psychologists, social workers, and counselors at hearings held in February 1976, June 1977, and January 1979, veterans of the Vietnam era have suffered significantly as the result of society's indifference and, in some cases, overt hostility to the sacrifices they made during their periods of military service. Those who served in Vietnam have also experienced anguish due to the unusual conditions of service in Vietnam--including the inability to differentiate between ally and enemy, the increased tension due to guerilla warfare, and the fact that, because service personnel often went to or returned from Vietnam singly rather than as part of entire combat units, they often lacked the mutual psychological support of long-known comrades during the critical periods of entering and leaving the conflict. Large numbers of returning Vietnam veterans have experienced guilt, bewilderment, alienation, pessimism, tension, restlessness, and other symptoms of low-grade readjustment problems. These problems frequently result in family difficulties, unemployment, alcohol or drug dependence or abuse, arrest records, and other forms of social and economic dislocation.

96th Congress, First Session, 1979 U.S.Code Cong. & Admin. News 169, 181-82. Given the overwhelming and well-known evidence of psychological problems among Vietnam veterans, it was unreasonable for counsel not to investigate Laws' military experience to see if any evidence of mitigating circumstances could be developed.

Had he investigated, counsel would have learned that Laws had earned the National Defense Service Medal, the Vietnam Service Medal, the Vietnam Campaign Medal and the Good Conduct Medal. Had he investigated, counsel would have learned that Laws' father, two sisters, and brother would have testified that Laws was a normal young man before his entry into the Army at age 17. Counsel would have learned that Laws had undergone a drastic personality change after the time he spent in Vietnam. He would have learned that Laws changed from a church-going young man who had never had any trouble with the law to a nervous and withdrawn loner constantly in trouble, in and out of prison. But counsel did not learn anything about Laws' background because he unreasonably relied upon the statement of one woman, who he thought might be Laws' stepmother, that no members of Laws' family would be willing to testify.

Counsel failed to investigate and discover the records of Laws' psychiatric treatment in the Mississippi State Hospital in 1974 and 1975. Had he done so, counsel would have learned even more evidence of appellant's personality change after his military service in Vietnam.

Granted, defense counsel put much store in his belief that the jury would never impose the death penalty because appellant was not the person who fired the gun, and therefore, he reasoned, did not deserve to die. Such a belief was totally unwarranted given the gruesome and calculated nature of the murders with which appellant was charged and his two prior capital murder convictions. To disregard the reality and seriousness of the situation was unreasonable and too risky to be condoned. His conduct was contrary to professional norms. It also evidenced a complete lack of pretrial preparation and put at risk appellant's right to offer evidence of mitigating circumstances. Given the totality of the circumstances in this case, counsel's representation of Laws was defective enough to undermine the reliability of the adversarial process. See Kimmelman, 477 U.S. at 386, 106 S.Ct. at 2589.

In the present case, the district court did not set Laws' conviction aside; the district court only ordered that the case be remanded for a new penalty hearing. Thus, even if we were to affirm the decision of the district court, appellant would not go free because defense counsel blundered. The only effect would be to grant appellant a new penalty hearing--one which would be in accord with our notion of the fair treatment that should be necessary before the state can take a person's life. I would affirm the decision of the district court.