concurring.
The court’s opinion, which I join, concludes that Brewer did not receive the sort of legal assistance that was his due at the sentencing hearing. Counsel invested all his time in attempting to get Brewer off and treated sentencing as an afterthought — a blunder, because counsel should have appreciated from the beginning that there was not much chance of acquittal. Sentencing was to be the main event.
Perhaps a canny lawyer would have proceeded just as Brewer’s did, trying to maximize the chances of acquittal while counting on the courts to protect his client from execution in the event of conviction. In capital cases, the best defense at sentencing may be no defense, leading to an order annulling the death sentence. Once guilt is established the options are death or extended imprisonment. The lack of a stirring defense at the sentencing phase increases the likelihood that a capital sentence will be converted to a life sentence, while an impeccable performance may doom the client to the gallows.
Deliberately sub-par performance is unethical, but some lawyers are willing to *861break rules to prevent capital punishment, which they view as a sin greater than any they could commit in the client’s behalf. Brewer’s lawyer disregarded his legal obligations in order to assist his client: the lawyer submitted perjured testimony. That maneuver backfired. Maybe the lackadaisical performance at sentencing was just another stratagem — properly treated, when discovered, as forfeiting any right to a new sentencing hearing. Indiana does not contend, however, that counsel was trying to pull this stunt, and if we take things at face value we must conclude that counsel botched the job.
Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984), holds that even in a capital case the defendant must establish that his lawyer’s shortcomings led to prejudice. This means “a reasonable probability that, absent the errors, the sentencer ... would have concluded that the balance of aggravating and mitigating circumstances did not warrant death.” 466 U.S. at 695, 104 S.Ct. at 2069. In Indiana the sentencer is the judge; juries recommend but do not impose sentences. Ind.Code § 35-50-2-9. Indiana naturally contends that counsel’s failure to present the psychological evidence to the jury was immaterial, because before imposing sentence the judge obtained the information that Brewer says his lawyer should have furnished.
The state would have a good argument, if the judge made an independent decision — if the recommendation of the jury were no different from the recommendation of the judge’s law clerk. Indiana’s brief depicts it so. Yet Martinez Chavez v. State, 534 N.E.2d 731, 735 (Ind.1989), holds that “to sentence a defendant to death after the jury has recommended against death, the facts justifying a death sentence should be so clear and convincing that virtually no reasonable person could disagree that death was appropriate in light of the offender and his crime. A trial court cannot override the jury’s recommendation unless the facts meet this standard.” In denying the petition for rehearing, the Supreme Court of Indiana rejected the argument that a judge could sentence a defendant to death if the jury’s recommendation for mercy is “unreasonable”. 539 N.E.2d 4 (1989). The brief filed in this court by the Attorney General of Indiana, asserting that the judge may freely reject the jury’s recommendation, cites no cases.
Reasonable persons could believe death an inappropriate penalty for Brewer, so the judge could not have imposed the death sentence in the face of a contrary recommendation from the jury. That leaves only the question whether there is a “reasonable probability” that the jury would have recommended against death had it known of Brewer’s limited intellect and passive personality. This is an empirical inquiry. How do juries react to such information? On the one hand, it shows the defendant to be less culpable; on the other, it shows the defendant to be less deterrable. These cut in different directions. Jurors who see capital punishment as the just desert of the wicked will be swayed in favor of lenience; jurors with more instrumental views will incline toward execution as the only way to incapacitate such a person.
Brewer’s current lawyers, like those representing the state, offer up confident (and divergent) assertions about how juries respond to claims of diminished mental capacity. Neither of these incompatible beliefs has any visible means of support. Lawyers see but a few capital cases during their lifetimes. They acquire anecdotes, not data. You need to study hundreds of similar cases to learn the probable effects of presenting different kinds of evidence to juries. As it turns out, social scientists have carried out such studies — studies neither side bothered to consult, each preferring asseveration to fact.
Trying to persuade the jury that the accused is mentally ill is worse than no defense at all. Jurors distrust insanity defenses, believe that the defendants are trying to bamboozle them; if persuaded that the defendants are indeed nutty, jurors believe that death is the only sure way to prevent future crimes. Lawrence White, Juror Decision Making in the Capital Penalty Trial: An Analysis of Crimes and Defense Strategies, 11 L. & Human *862Behavior 113, 122-25 (1987). Accord, Project, Standardless Sentencing, 21 Stan. L.Rev. 1297, 1361-63 (1969). Drawing to the jury’s attention an organic problem such as mental retardation, though, cuts the other way; jurors are more likely to credit these claims and to express sympathy. Ellsworth, Bukaty, Cowan & Thompson, The Death-Qualified Jury and the Defense of Insanity, 8 L. & Human Behavior 45 (1984). Whether such defenses actually help the accused is a close question. The Stanford study finds no effect, 21 Stan.L.Rev. at 1383, and the Ellsworth study a small one.
Brewer has an organic intelligence problem, no one doubts. His “passivity” too may have an organic source, although a jury also might think this so much psychiatric mumbo-jumbo. Presenting to the jury the congeries of facts and diagnoses laid before the judge could not have done much harm, and might have helped if Ells-worth and colleagues are right. The impetus for death may have been so strong that Brewer had little to lose. I therefore agree with my colleagues that there is a “reasonable probability” that the jury would have recommended against death had it known of Brewer’s limited intellect and passive personality. Indiana might have been able to make a contrary showing by analyzing the results of defenses presented to Indiana juries. It did not try; as I have emphasized, the prosecutors thought that they could rescue this sentence by thumping on the table and hoping that our gestalt would match theirs. Intuition is a poor substitute for data. Before sending a man to his death a state should have more regard for both law and fact than Indiana has shown.