486 So.2d 476 (1985)
Ex parte STATE of Alabama.
(Re Jerome Vincent Berard v. State of Alabama).
Ex parte Jerome Vincent BERARD.
(Re Jerome Vincent Berard v. State of Alabama).
84-309, 84-310.
Supreme Court of Alabama.
December 20, 1985.
Rehearing Denied February 7, 1986.
Charles A. Graddick, Atty. Gen., and Ed Carnes, Asst. Atty. Gen., for petitioner/cross-respondent.
Ira De Ment and Ronald W. Wise, Montgomery, for respondent/cross petitioner.
PER CURIAM.
This is an appeal following retrial. On this retrial, the defendant, Jerome Vincent Berard, was convicted in the Circuit Court of Montgomery County of a capital offense based on Code 1975, § 13-11-2(a)(10) (intentionally killing two or more human beings by one or a series of acts). Following the conviction by a jury, a sentencing hearing was held in the presence of the same jury, which returned a recommendation that the defendant be sentenced to death. A sentencing hearing was then held before the trial judge, who also found that the aggravating circumstances outweighed the mitigating circumstances and that the defendant should be sentenced to death. The Court of Criminal Appeals, pursuant to the mandates of Beck v. State, 396 So.2d 645 (Ala.1980), reviewed the record and the issues presented by the defendant and upheld the conviction of the capital offense, but remanded the case to the trial court for a new sentencing hearing. Berard v. State, 486 So.2d 458 (Ala.Crim.App.1984). On rehearing, the court denied the applications for rehearing from both the defendant and the State, but extended its original opinion in order to state that, on remand to the trial court, "this Court has clearly expressed its disapproval of a sentence to death in this case." Berard, supra.
Both the defendant and the State filed petitions for writ of certiorari to this Court. The defendant's petition for the writ is basically a request for this Court to review the issues concerning his conviction. The State's petition is a request for this Court to review the Court of Criminal Appeals' decision to remand the case for a new sentencing hearing. Because we are reversing the defendant's conviction of the capital offense and remanding his case for a new trial, we do not address the issues raised by the State with respect to the sentencing phase of the trial.
A complete statement of the facts in this case can be found in the opinion of the Court of Criminal Appeals following the first trial in this case, Berard v. State, 402 So.2d 1044 (Ala.Crim.App.1980). In that opinion, the Court of Criminal Appeals affirmed the defendant's conviction, but remanded the case for a new sentencing hearing. Before further action was taken, the conviction was reversed on authority of Beck v. State, supra, and that reversal necessitated a second trial, from which the present appeal was taken.
The defendant was charged with the capital murder of two young boys, ages 14 and 16, outside the Skatehaven skating rink in Montgomery in April 1978. At trial, the defendant entered pleas of not guilty and not guilty by reason of insanity, and he presented several expert witnesses on the issue of whether he was insane at the time of the crime. One of those witnesses was Dr. Chester Jenkins, a psychologist, who testified that the defendant was probably having a psychotic episode at the time he shot the two boys and was suffering from latent schizophrenia. On cross-examination, the district attorney questioned Dr. Jenkins in the following manner:
"Q. Is he [defendant] capable then of having another psychotic episode?
"A. Certainly.
"Q. Not unlike the one you say he had on April the 14th and 15th of 1978?
"MR. DE MENT [Defendant's attorney]: Same objection.[1]
"THE COURT: Overruled.
"Q. Sir?
"A. Do I think he is capable of having recurrent episodes? Yes, I do.
"Q. Recurring episodes, sir, let me ask you this: Is he capable of shooting somebody else?
"MR. WISE [defendant's attorney]: Same objection, Your Honor.
"THE COURT: Overruled.
"Q. Again?
"THE COURT: While having a psychotic episode?
"Q. While having a psychotic episode? Sir?
"A. Yes, I think so." (Emphasis added.)
The defendant contends that the question of whether he would shoot somebody else was very prejudicial to his defense and was meant solely to inflame the passions of the jury. The State argues, on the other hand, that this is valid cross-examination of an expert witness on the issue of the defendant's insanity plea and, therefore, that the trial court did not err in allowing this testimony into evidence.
The State has cited several cases stating that both the defense and the State have a "wide latitude" in introducing evidence of a defendant's acts, declarations, or conduct, occurring both prior and subsequent to the crime, when the defendant has interjected the issue of insanity. See, e.g., Nichols v. State, 276 Ala. 209, 212, 160 So.2d 619, 621 (1964); Barbour v. State, 262 Ala. 297, 303, 78 So.2d 328, 333 (1954); George v. State, 240 Ala. 632, 637, 200 So. 602, 606 (1941). It is the State's contention that the question asked by the district attorney was used to shed light on the defendant's mental state at the time of the crime and to test the value and accuracy of the witness's opinion. See, Nichols, supra; George, supra. The State's argument appears to rest on the fact that there was a difference in the definition of the term "latent schizophrenia" as given by Dr. Jenkins and the definition of that term as found in standard psychiatric books. Thus, the State argues, the question asked by the district attorney was relevant for the jury's determination of the value to be placed on Dr. Jenkins's testimony.
Although the district attorney's question may have had some relevance in testing the value to be given to Dr. Jenkins's testimony, we must disagree with the State's interpretation of our prior cases. While the acts, declarations, and conduct of a defendant subsequent to the crime are admissible on the issue of insanity, the cases cited by the State make it clear that that principle relates to events occurring after the crime but before the trial. See, Nichols, supra (the defendant's answers to questions three hours after the murder of his wife were admissible to shed light on his state of mind at the time of the murder); George, supra (the defendant's attempted suicide while awaiting trial was admissible).
We have not been cited to any case in Alabama that approves of a prosecutor's asking a question about what the defendant is capable of doing in the future. We additionally note that this Court has previously stated that, as long as a prosecutor does not comment on the possibility that the defendant will commit future illegal acts, he may legitimately argue to the jury the need for law enforcement as a deterrent to crime. Ex parte Waldrop, 459 So.2d 959, 962 (Ala.1984); Cook v. State, 369 So.2d 1251, 1255 (Ala.1979). It would seem to be much more prejudicial to a defendant to allow the prosecutor to elicit this kind of testimony from a defense witness than it would be for the prosecutor to merely comment on that possibility in closing argument.
Our research has uncovered only one case which deals with this kind of situation. In State v. Barksdale, 590 S.W.2d 931 (Tenn.1979), the defendant had been charged with the crime of rape and had entered the plea of not guilty by reason of insanity. The psychiatrist called by the defense testified that the defendant was suffering from a mental disease which manifested itself in his inability to control his actions when he committed the rape. On cross-examination by the prosecution, the following took place:
"Q. If this jury should find this man not guilty by reason of insanity, can you assure us that he won't go out this afternoon and do it again?
"A. MR. SCHATZ [Defense Attorney]: Your Honor, if the Court please, I think that is improper, and I would(Interrupted)
"MR. AXLEY [State's Attorney]: He's come as an expert, Your Honor, and this is cross-examination.
"MR. SCHATZ: If the Court please, though there is an instruction that will be given as the Court well knows, and as Mr. Axley knows that would cover that event should it come up.
"THE COURT: I'm going to overrule you, Mr. Schatz.
"BY MR. AXLEY (Continuing Cross-examination):
"Q. Would you answer that? Do you want me to ask it again?
"...
"A. I know that I would be reluctant to release him today, because I feel he has a serious sexual problem, and I don't think he has received the treatment for that problem in the past, and I think he does need the treatment for that problem. And, I think if he does not receive the treatment for that problem, the odds are that he would get involved in this sexual behavior again, unless he was treated for it."
590 S.W.2d at 932. (Emphasis added.)
The Tennessee Supreme Court reversed the defendant's rape conviction, relying on a previous case that held impermissible a prosecutor's comment in closing argument to the effect that "the defendant could `go right out of here and kill again.'" 590 S.W.2d at 933 (quoting Covey v. State, 504 S.W.2d 387, 393 (Tenn.Crim.App.1973)). The court decided that this kind of testimony was highly prejudicial and "would be even more devastating than the advancement of that thesis by the prosecution as in the Covey case, supra." Barksdale, 590 S.W.2d at 933.
We, too, believe that this kind of question is unfairly prejudicial to the defendant. First, the cases cited by the State do not support the proposition that an expert may be questioned on the future conduct of a defendant. Therefore, there was no proper basis for the district attorney to ask such a question. Second, the central issue in the guilt phase of a capital murder trial is whether the State has satisfied its burden of proving beyond a reasonable doubt that the defendant is guilty of the crime charged. Beck, 396 So.2d at 662. This kind of question could have easily shifted the focus of the jury's attention to the issue of punishment, which is an improper consideration at the guilt phase of the trial.
For these reasons, the defendant's conviction for the capital murder of two persons is reversed and this cause remanded to the Court of Criminal Appeals for that court to order a new trial.
REVERSED AND REMANDED.
TORBERT, C.J., and FAULKNER, JONES, ALMON,[2] SHORES, BEATTY, ADAMS and HOUSTON,[2] JJ., concur.
MADDOX,[2] J., dissents.
MADDOX, Justice (dissenting).
I must respectfully dissent from the majority opinion, which concludes that the trial court erred in permitting the district attorney to ask Dr. Chester Jenkins, a psychiatrist hired by the defendant, whether the defendant's "latent schizophrenia" could recur.
As the majority opinion correctly states, there is limited authority on the exact question presented, but I believe that our evidentiary rule, which allows a searching cross-examination of witnesses, should be applied in this case. In fact, because of the questionable reliability of expert testimony involving mental disorders, I believe the searching cross-examination rule is uniquely applicable to the testimony of psychological experts.
In a comment in the Maryland Law Review, the author examines the history and the development of the use of the psychologist as an expert witness and suggests that the reliability of expert testimony is suspect. He writes:
"Studies of the reliability of psychological diagnoses are rare; however, a number have been conducted regarding psychiatric diagnoses. In a review of six major studies of the reliability of psychiatric diagnosis conducted between 1956 and 1975, two researchers calculated interrater reliability, correcting for change agreement. Summarizing their results, they concluded:
"`There are no diagnostic categories for which reliability is uniformly high. Reliability appears to be only satisfactory for three categories: mental deficiency, organic brain syndrome (but not its subtypes), and alcoholism. The level of reliability is no better than fair for psychosis and schizophrenia and is poor for the remaining categories.'
"The authors also observed that in five of the six studies the diagnosticians were of similar background, and in some instances, special efforts were made to have the participants arrive at some consensus on diagnostic principles prior to beginning the researchfactors that should have increased interrater reliability. Contrary to their expectations, however, these elements did not significantly contribute to high reliability. They theorized: `One can only assume ... that agreement between heterogeneous diagnosticians of different orientations and backgrounds, as they act in routine clinical settings, is even poorer than is indicated in this review.' Their study also revealed no essential change in diagnostic reliability over time. The authors' conclusion was less than sanguine about the reliability of psychiatric diagnoses:
"`The reliability of psychiatric diagnosis as it has been practised since at least the late 1950's is not good. It is likely that the reasons for diagnostic unreliability are the same now as when Beck et al. (1962) studied them. They found that a significant amount of the variability among diagnosticians was due to differences in how they elicited and evaluated the necessary information, and that an even larger amount was due to inherent weakness and ambiguities in the nomenclature.'
"Overall interrater reliability for specific diagnoses ranges from about thirty-two percent to sixty-three percent agreement, which means that psychiatrists tend to disagree on specific diagnoses a significant percentage of the time. To date, there are no studies that demonstrate that the rate of agreement among psychologists is any better or worse than that of psychiatrists. One danger arising from such a lack of reliability is that patients may be inappropriately classified; these classifications may then directly influence the type of treatment they receive.
"Despite a salutary trend toward the creation of an empirically based nomenclature, the accuracy of psychological diagnoses based thereon may still be suspect. If the specific criteria for inclusion in a particular diagnostic category remain too subjective, then their value as the basis for an operational nomenclature may be significantly diminished. For example, if one element for inclusion in a diagnostic category is evidence of memory impairment, the individual psychologist's concept of the degree of impairment necessary for inclusion may vary widely. Second, patients with a few, but not all, of the symptoms necessary for a particular diagnosis, or who manifest symptoms that are not discrete but overlap significantly, may be placedtemporarily or permanently depending upon the combination of symptoms observedin the residual category of `undiagnosed psychiatric illness.' The implication is that an individual may be labeled mentally disordered, albeit tentatively, even though a specific diagnostic category does not exist. Such flexibility may lead to overinclusiveness. Finally, depending on how carefully the symptom criteria are drafted, they still may not account for the relative frequency of such behavior in `normal' populations."
Comment, The Psychologist as Expert Witness, 38 Md.L.Rev. 539, 577-81 (1979).
The author concludes:
"The probative value and admissibility of each level of psychological testimony is suspect. Although psychological expert witnesses may be properly qualified in terms of their education and experience, these criteria do not assure that the scientific underpinnings of such testimony are valid. Psychological judgments are not as accurate as the courts presume them to be, and diagnoses based upon a psychologist's observations are, at the very least, questionable. For reasons that remain obscure, the courts have failed to apply to psychology the standard used to evaluate the admissibility of novel types of scientific evidence. Because the courts have tacitly taken judicial notice of psychology as a bona fide scientific discipline, there is no opportunity at trial to assail the underlying accuracy of such judgments in terms of their reliability and validity.
"Psychological expert testimony should be excluded or at least severely curtailed for two reasons. First, psychological diagnoses may not be reasonably accurateaccuracy greater than chance and, consequently, cannot meet the prima facie showing of logical relevancy. Those judgments that are reasonably accurate should be admitted subject to the usual course of rigorous cross-examination. Second, even if psychological diagnoses are reasonably accurate, psychological expert testimony should be excluded because such evidence cannot meet the higher standard of accuracy required by the courts in light of the evidentiary counterweights, and the jury is likely to exaggerate its significance given its aura of scientific objectivity. Psychological opinion testimony regarding the causal connection between a mental disorder and criminal or tortious conductthe issue of ultimate factshould be excluded on similar grounds, at least until it can be satisfactorily demonstrated that psychologists possess special expertise in evaluating the relationship between the scientific `fact' of mental disorder and the applicable legal standard. The test should be whether such an opinion would be more helpful to the trier of fact than the thoughtful opinion of the average layperson."
Id. at 598-99.
Although I disagree with the author of the article in the conclusion that psychological expert testimony should be excluded, I do agree that when it is admitted it should be subject to "the usual course of rigorous cross-examination."
Why was this rigorous cross-examination proper in this case? Dr. Jenkins's conclusions were diametrically opposed to those of the State's expert, and during cross-examination Dr. Jenkins admitted that the defendant did not meet the definition of "latent schizophrenia" contained in the American Psychiatric Association's Diagnostic and Statistical Manual II. Consequently, the district attorney's questions about the likelihood that the defendant would have a recurring episode of what Dr. Jenkins classified as "latent schizophrenia," including another psychotic episode in which he might shoot someone, were relevant to an understanding of the "latent schizophrenia" label Dr. Jenkins was using to diagnose the defendant. Questions about the nature and symptoms, and especially the duration of the illness, were certainly relevant and material to an understanding of Dr. Jenkins's diagnosis and the conclusions he drew from it. I am of the opinion that the cross-examination of Dr. Jenkins was very relevant in testing the reliability of Dr. Jenkins's opinion about the mental state of the defendant at the time he committed the offense; therefore, I must respectfully dissent.
NOTES
[1] Outside the presence of the jury, this line of cross-examination had previously been discussed before the trial judge. The defendant's attorney objected to the district attorney's asking any question that concerned the future conduct of the defendant, on the grounds that such questions would prejudicially influence the jury in its findings and sentencing. At that time, the trial judge did not rule on the objection.
[2] Although these justices did not sit at oral argument, they have listened to the tape of oral argument.