State v. Stuard

176 Ariz. 589 (1993) 863 P.2d 881

STATE of Arizona, Appellee,
v.
James William STUARD, Appellant.

No. CR-90-0355-AP.

Supreme Court of Arizona, En Banc.

November 18, 1993.

Grant Woods, Atty. Gen. by Paul J. McMurdie, Linda L. Knowles, Phoenix, for appellee.

Dean W. Trebesch, Maricopa County Public Defender by James H. Kemper, Phoenix, for appellant.

OPINION

FELDMAN, Chief Justice.

James William Stuard ("Defendant") was convicted of three counts of first degree murder, one count of attempted first degree murder, three counts of first degree burglary, one count of second degree burglary, two counts of attempted sexual assault, and one count of armed robbery. All counts arise from four separate attacks on elderly women. Defendant, age fifty-three at the time of his conviction, was sentenced to death for the murders and received lengthy prison sentences for the remaining convictions. Appeal to this court is automatic. Ariz.R.Crim.P. 26.15, 31.2(b). We have jurisdiction pursuant to Ariz. Const. art. VI, § 5(3), Ariz.R.Crim.P. 31, and A.R.S. § 13-4031.

Although Defendant was convicted of numerous crimes in addition to the capital offenses, his three notices of appeal fail to specify exactly which convictions he appeals. Because Defendant's brief does not specifically challenge any of the non-capital convictions, we confine ourselves solely to those issues raised and to a search for fundamental error. Cf. State v. Schaaf, 169 Ariz. 323, 326, 819 P.2d 909, 912 (1991).

FACTS AND PROCEDURAL HISTORY

Beginning in May 1989, four elderly women were assaulted in their homes. Three lived in the same general neighborhood and were attacked within days of each other. They did not survive. Following the third attack, Defendant was questioned as part of the general investigation, although apparently he was not yet a suspect. Police took Defendant's fingerprints and photographed the boots he was wearing.

The fourth victim was not attacked until nearly three months later. This victim lived in a different neighborhood and, although badly injured, was able to call the police. Defendant was apprehended; he admitted the last attack but denied any involvement in the three murders. Each incident is described in turn.

VICTIM NO. 1, MRS. C

Mrs. C was seventy-seven years old and lived by herself. She weighed 125 pounds and stood five feet seven inches tall. On the evening of May 8, 1989, her son-in-law went to her home after being unable to reach her by telephone. He found her dead body in a spare bedroom. A picture was knocked over on the bed's headboard. In the kitchen, Mrs. C's eyeglasses and a drinking glass lay broken on the floor. The protective grating had been knocked off a fan near the kitchen entrance.

Mrs. C's house coat was open at the top and pulled up and tucked underneath the buttocks and the small of the back, exposing part of her thighs. A bra strap was unfastened and her panties were found lying on top of her abdomen. Dentures, apparently belonging to Mrs. C, were found under the body.

The medical examiner determined that Mrs. C died from strangulation, probably by hand. Her neck was bruised and the bone at the base of the tongue was broken. The choking could have lasted up to five minutes. There were other injuries as well, some of which would have been fatal had she survived the strangulation. Stab wounds were found in her heart, spleen, and stomach. There was also bruising on the chest, the back of both forearms, and the back of both hands. A bruise on the top of the head was probably due to a blow — perhaps by a fist. Mrs. C's sternum and nine of her ribs were also fractured. These injuries could have been caused by kneeling on the sternum area and were consistent with multiple blows up and down the side of the body. Again, a fist could have inflicted these injuries. Despite the condition of Mrs. C's clothing, the medical examiner found no injury consistent with sexual assault.

Evidence linked Defendant to the scene. Hair that did not belong to Mrs. C, but that was consistent with Defendant's, clung to a blanket on which the body was found. Also, Defendant knew Mrs. C because he had done yard work for her. There were also unusual similarities between this murder and the other attacks, where more incriminating evidence existed. Aside from the victims' obvious likeness in age and gender, there was also a likeness among the injuries. Mrs. C's sternum and rib injuries were, according to the medical examiner, "very, very similar" to those inflicted on the second victim. Finally, the close proximity of the murders — both temporal and geographic — pointed to a single perpetrator.

VICTIM NO. 2, MRS. L

Mrs. L was seventy-five years old and also lived alone. She weighed eighty-nine pounds and was five feet three inches tall. On May 9, 1989, her body was found lying face up in a hallway at her home, with her head extending into a spare bedroom.

As with Mrs. C, Mrs. L's clothing was disheveled. Eyeglasses lay askew on her face, and she wore only one shoe. The other shoe was found in a nearby bedroom. Her house dress was partially opened and pulled up to the waist, exposing her legs and undergarments. Her slip was torn and her bra pushed up, exposing one breast. Panties were on the floor near her body, and her girdle was partially rolled down.

The medical examiner estimated that Mrs. L had been dead for about one day. The examiner found no obvious signs of injury and no injury consistent with sexual assault. In fact, the examiner initially believed that Mrs. L may have died from natural causes. Upon further examination, however, the medical examiner concluded that she died from a collapsed heart caused by trauma to the sternum. The upper onethird of her sternum was fractured. Such an injury could be caused by a direct punch, by a kick, or by kneeling on the breast bone and pushing it inward. Death likely occurred within one minute of the injury. In addition to the fatal injury, Mrs. L suffered multiple rib fractures and a contusion to the back of her head.

Evidence also linked Defendant to this crime scene. His palm print was on the door frame near the body. Police also found a shoe print in the dust on the carport driveway with a "wave" or a "V" pattern. Defendant was wearing boots with soles that had a similar pattern when he was initially picked up in early May.^[1] A similar print was found on the kitchen floor. As noted, Defendant had also done yard work for this victim.

VICTIM NO. 3, MRS. W

Mrs. W was eighty-one years old and, like the others, lived by herself. She weighed 149 pounds and was five feet nine inches tall. On May 10, 1989, concerned neighbors called the police because they had not seen Mrs. W for about a day. The police found her dead in her home. Her body was in the living room amid substantial clutter. She was face up with the right leg bent and slightly pulled up. She was not wearing underwear. A curtain draped over her abdomen concealed stab wounds. Blood was sprayed about the room with trails on the wall and curtains. Although Mrs. W was not wearing eyeglasses, a pair was found on the kitchen floor.

The medical examiner determined that Mrs. W was stabbed to death. She suffered multiple stab wounds to the neck, chest, and left forearm. Although only three were fatal, there were more than a dozen such wounds, all likely inflicted before death. Mrs. W also suffered "blunt force" injuries to her face, chest, and right buttock. This type of injury is caused by a blunt object striking the body or the body striking a hard surface. Some of the injuries to her face were consistent with being punched. Her legs, knees, hands, wrists, and forearms were bruised. The medical examiner could not tell whether these injuries occurred before or after the fatal stab wounds. The medical examiner found no trauma consistent with sexual assault.

Evidence also connected Defendant to this scene. Numerous fingerprints identified as Defendant's were discovered in the living room, where the body was found, as well as other places in the house. Defendant's prints were discovered on a food tin, a lamp hood, a fireplace mantle, a soda bottle, a bathroom door, and a freezer door. On the tile next to the fireplace in the living room, police found a shoe print that again was similar to the boots Defendant was known to have worn. Finally, parts of an apple, apparently chewed and spit out, were found on the floor next to the fireplace and on the fireplace itself. Like the other victims, Mrs. W had employed Defendant to do yard work.

VICTIM NO. 4, MRS. V

Mrs. V, eighty years old, discovered Defendant in her home on August 6, 1989. He instructed Mrs. V to lie on the floor, demanded money and sex, and threatened to kill her. Defendant also demanded that Mrs. V prepare him food because he was hungry. He ate some leftover tacos and bananas and drank a soda. He then took some change and some frozen food from Mrs. V's freezer and loaded the food into a box. Although he did not rape Mrs. V, he did assault her. He repeatedly punched her in the chest so hard that she thought each blow would literally "break" her chest. Defendant also repeatedly stabbed Mrs. V in the neck and chest and beat her on the head with a hammer.

Amazingly, Mrs. V survived. Although her wounds were very serious, none was immediately fatal. Mrs. V fortuitously suffered from an ailment that caused her skull to thicken; this condition may have helped protect her from the hammer attack. She retained enough consciousness to call the police.

ARREST, TRIAL, AND CONVICTION

A local citizen heard of Mrs. V's attack on a police scanner. On the way to Mrs. V's house, this citizen spotted Defendant carrying a box of food. The citizen called the police, who arrested Defendant.

Defendant admitted attacking Mrs. V, explaining that he had gone to her house looking for work or food. According to Detective Michael Chambers, the officer who questioned him, Defendant claimed he was hungry and simply "went off" after Mrs. V raised her voice. He admitted stabbing Mrs. V, beating her with a hammer, and stealing her food. In fact, he carefully noted that he hit her with the hammer's flat side, and not its claw or head.

Because of the similarity to the earlier murders, Chambers also questioned Defendant about Mrs. C, Mrs. L, and Mrs. W. Defendant did not immediately recognize any of these victims in their discussion. He did, however, recognize each after Chambers showed him photographs of their homes. Defendant also provided additional information not in the photographs. He admitted working for all of the victims but denied ever being inside any of their homes or killing them. Even when confronted with the fact that his fingerprint was found inside Mrs. W's house, he still denied ever having been inside.

A jury found Defendant guilty of the first degree murders of Mrs. C, Mrs. L, and Mrs. W; the first degree burglary of Mrs. C's, Mrs. W's, and Mrs. V's residences; the second degree burglary of Mrs. L's residence; the attempted sexual assault of Mrs. C and Mrs. L; and the armed robbery and attempted murder of Mrs. V. The jury acquitted Defendant of all other robbery charges and also of the attempted sexual assault of Mrs. W. Following a hearing, the trial judge found two aggravating circumstances and one statutory mitigating circumstance. However, the trial judge did not find the mitigating evidence substantial enough to call for leniency and sentenced Defendant to death for each of the murders. The judge also sentenced Defendant to a total prison term of 140 years for the non-capital convictions.

DISCUSSION

A. TRIAL ISSUES

1. Motion to sever

A grand jury indicted Defendant on fifteen separate felony counts arising out of the four attacks. Before trial, as well as after the state presented its cases, Defendant moved to sever the counts involving the three murders, both from each other and from the attack on Mrs. V. The court denied these motions. Although Defendant originally sought separate trials for each incident, he now claims only that the court should have severed trial of the counts arising out of the three murders from those arising out of Mrs. V's attempted murder.

Defendant argues that the state had an "ice-cold" case against him in the attempted murder count, in which Mrs. V testified and Defendant admitted to the attack. In contrast, he correctly claims that there was much less convincing evidence in the murder cases. According to Defendant, the state's strategy was to win convictions in the weaker murder cases based on the strength of the attempted murder case. Thus, the trial court erred by denying his motion to sever the two sets of cases. The result, Defendant asserts, was a trial so fundamentally unfair as to deprive him of due process required by the Fourteenth Amendment to the United States Constitution.

a. Were the counts properly joined?

Ariz.R.Crim.P. 13.3 and 13.4 govern the joinder and severance of counts. These rules provide, in relevant part:

RULE 13.3 JOINDER

a. Offenses. Provided that each is stated in a separate count, 2 or more offenses may be joined in an indictment, information, or complaint, if they:

(1) Are of the same or similar character; or

(2) Are based on the same conduct or are otherwise connected together in their commission; or

(3) Are alleged to have been part of a common scheme or plan.

* * * * * *

RULE 13.4 SEVERANCE

a. In General. Whenever 2 or more offenses or 2 or more defendants have been joined for trial, and severance of any or all offenses, or of any or all defendants, or both, is necessary to promote a fair determination of the guilt or innocence of any defendant of any offense, the court may on its own initiative, and shall on motion of a party, order such severance.

b. As of Right. The defendant shall be entitled as of right to sever offenses joined only by virtue of Rule 13.3(a)(1).

The trial court held that joinder was appropriate under Rule 13.3(a)(1) and (a)(2). The facts make it quite apparent that the charges were "of the same or similar character." Each involved an attack and beating of an elderly woman living alone; each involved infliction of similar injuries. See State v. Day, 148 Ariz. 490, 715 P.2d 743 (1986). Because we believe that Rule 13.3(a)(1) more clearly supports joinder in this case, we do not address Rule 13.3(a)(2).

b. Was Defendant entitled to severance?

Defendant points out that if Rule 13.3(a)(1) was the basis for joinder, he was entitled to severance as a matter of right under Rule 13.4(b). We agree. We do not believe, however, that the error of failing to sever under Rule 13.4(b) will always require reversal. If the evidence of one crime would have been admissible in a separate trial for the others, it is unlikely that Defendant suffered prejudice by the court's denial of severance. Cf. Day, 148 Ariz. at 494, 715 P.2d at 747 (noting that prejudice in failing to sever is mitigated because the same evidence would have been admissible in the separate trials); State v. Via, 146 Ariz. 108, 115, 704 P.2d 238, 245 (1985) (same), cert. denied, 475 U.S. 1048, 106 S.Ct. 1268, 89 L.Ed.2d 577 (1986).

A treatise on Arizona evidence law notes that when Rule 13.3(a)(1) ("same or similar character") is the basis for joinder, severance is mandated under Rule 13.4(b) even in those cases in which the "similarities are so great as to permit the introduction of one to prove the identity of the perpetrator of the other." Morris K. Udall, et al., Arizona Practice — Law of Evidence § 84, at 184 n. 14 (3d ed. 1991) (hereinafter "Arizona Evidence"). According to the authors, however, "there is little risk of prejudice from joinder if the joined offense would be permitted to be introduced for an evidentiary purpose anyway." Here again, we agree. The authors note that Arizona courts have avoided this problem by interpreting the "identity" exception of Ariz. R.Evid. 404(b) in the same manner as the "common scheme or plan" exception — where there is no severance as a matter of right. Id. We choose not to stretch the rule to reach the result. Aside from the series of crimes themselves, there is little evidence of any scheme or plan and considerable evidence to the contrary. Instead, we simply look for any prejudice occasioned by the court's failure to sever.

Other states have followed a similar course. People v. Miller, 50 Cal.3d 954, 269 Cal. Rptr. 492, 509, 790 P.2d 1289, 1306 (1990) (cross-admissibility ordinarily dispels inference of prejudice that accompanies joinder of counts), cert. denied, 498 U.S. 1041, 111 S.Ct. 713, 112 L.Ed.2d 702 (1991); State v. Hall, 103 Wis.2d 125, 307 N.W.2d 289, 296-97 (1981) (risk of prejudice caused by joinder of counts is not significant if counts would be admissible in the separate trials). In light of this, we look to see if the evidence of the attempted murder would have been admissible in a separate trial for the three murders.

c. Would the evidence of the attempted murder have been admissible in a separate trial for the three murders?

The general rule, of course, is that other "bad acts are not admissible to prove the bad character of the perpetrator." State v. Roscoe, 145 Ariz. 212, 216, 700 P.2d 1312, 1316 (1984), cert. denied, 471 U.S. 1094, 105 S.Ct. 2169, 85 L.Ed.2d 525 (1985); see also Ariz.R.Evid. 404(b). Thus, in a separate trial for the three murders, the attack on Mrs. V would not be admissible solely to show that Defendant's bad character made it more likely that he committed the murders. Such evidence, however, may be admissible for other purposes, including "proof of motive, opportunity, intent, preparation, plan, knowledge, identity, or absence of mistake or accident." Ariz. R.Evid. 404(b). We believe that the evidence of the attempted murder was admissible under this rule's "identity" exception.

The identity exception to Ariz. R.Evid. 404(b) applies if identity is in issue, "and if the behavior of the accused both on the occasion charged and on some other occasion is sufficiently distinctive, then proof that the accused was involved on the other occasion tends to prove his involvement in the crime charged." Arizona Evidence, § 84, at 183-84. Merely showing that the crimes are of the same nature is insufficient to bring conduct within this exception. Instead, "[t]he pattern and characteristics of the crimes must be so unusual and distinctive as to be like a signature." Id. at 185 (quoting Edward W. Cleary et. al., McCormick on Evidence § 190, at 559-60 (3d ed. 1984)); see also Roscoe, 145 Ariz. at 217, 700 P.2d at 1317. In our analysis, therefore, we examine the differences as well as the similarities among the crimes. State v. Jackson, 124 Ariz. 202, 204, 603 P.2d 94, 96 (1979). "While identity in every particular is not required, there must be similarities between the offenses in those important aspects `when normally there could be expected to be found differences.'" Roscoe, 145 Ariz. at 217, 700 P.2d at 1317 (quoting Jackson, 124 Ariz. at 204, 603 P.2d at 96). We apply this analysis to the facts of this case.

As a preliminary matter, we note that the evidence of the attempted murder was sufficient to go to the jury. See State v. Fierro, 166 Ariz. 539, 547, 804 P.2d 72, 80 (1990). Moreover, the features of the attempted murder, when compared to the three completed murders, operate to set these crimes apart from others of the same general variety and point to a single assailant. Cf. Miller, 269 Cal. Rptr. at 509, 790 P.2d at 1306. Although it is true that all murderous attacks will share some characteristics, an examination of the attacks in this case reveals significant similarities and only minor differences.

First, the victims in the two sets of cases shared distinct characteristics. The most obvious one is that they were all elderly women. In addition, Mrs. V and the murder victims had something else in common that was less obvious, but more telling — they had all hired Defendant to do yard work for them. During the interview following his arrest, Defendant admitted knowing and working for each of these women, including Mrs. V. Finally, Mrs. V, like the others, lived in roughly the same area.^[2]

Moreover, Mrs. V's attack shared distinctive features with the murders. Mrs. V, like all three of the murder victims, was attacked in her home. In every case the victims were attacked by hand; the evidence showed that punching was probably common to all of the crimes. In fact, two victims (Mrs. C and Mrs. L) suffered a fractured sternum and fractured ribs that could have been caused by such punching. According to the medical examiner, the sternum is not easily broken; such an injury is commonly found in an automobile accident when the victim slams into a steering wheel. Mrs. V similarly testified that Defendant punched her in the chest "about 100 times" so hard that she thought each blow would literally break her chest. In addition to the punching, three of the four victims, including Mrs. V, suffered multiple stab wounds. Multiple stabbing, combined with severe manual beating, is an uncommon combination. Another similarity between the murders and the attempted murder was the sexual overtones of the crimes. According to the uncontroverted evidence, Defendant manifested a sexual interest in Mrs. V during the attempted murder. Likewise, there was evidence that the perpetrator of the crimes against Mrs. C and Mrs. L also tried to sexually assault them. Finally, Defendant ate and drank during his gruesome attack on Mrs. V, an unusual feature indeed, and his prints were found all over food-related items at Mrs. W's house.^[3]

There were, of course, differences as well. Mrs. V lived in a "different neighborhood" from the murder victims and was attacked about three months later. Also, unlike the murder victims, Mrs. V was not living alone at the time of the attack. Her son was temporarily staying at her house. In addition, the extent and exact type of injury varied among the victims. Finally, the "biggest difference," according to Defendant, was that Mrs. V survived her attack, while the others obviously did not.

These differences "pale to insignificance" compared to the similarities. Day, 148 Ariz. at 494, 715 P.2d at 747. The three-month time gap and the different neighborhoods are not so significant as to demand exclusion of the evidence, at least not when considered in conjunction with Defendant's connection with each of the victims. Furthermore, because the victims were attacked by hand, the difference in magnitude and type of injury is likely due to the varying amounts of resistance each victim offered. Finally, we flatly reject that Mrs. V's survival was due to anything other than chance. Defendant beat her severely with his fists, struck her head with a hammer, stabbed her in the chest and neck, and then disabled the telephone. Evidence indicated that, absent immediate medical attention, Mrs. V would not have lived very long. That she had a thickening of the skull, had another telephone, and managed to call for help may have been a miracle — but it is not a distinguishing factor. See State v. Harding, 141 Ariz. 492, 497, 687 P.2d 1247, 1252 (1984) (fortuitous survival of one victim was insufficient to prevent admission of prior bad act).

We conclude that the similarities among the crimes were sufficiently distinct for the evidence to fall within the identity exception of Ariz.R.Evid. 404(b). Cf. Miller, 269 Cal. Rptr. at 509-10, 790 P.2d at 1306-07 (numerous similarities among murders and attempted murders of homosexual men); Hall, 307 N.W.2d at 297-99 (numerous similarities among robberies during which one murder and one attempted murder occurred).^[4] This, however, does not end our inquiry.

d. Did the court act properly to prevent prejudice?

Even though evidence of other crimes fits within the identity exception, a defendant might still suffer unfair prejudice if the jury uses the identity evidence as character evidence in violation of Ariz. R.Evid. 404(a) and 404(b). Therefore, when evidence of a prior bad act is admitted solely to show identity, the defendant is entitled to an instruction limiting the jury's use of the evidence to the permissible purpose. See Arizona Evidence § 84, at 179; State v. Canedo, 125 Ariz. 197, 199-200, 608 P.2d 774, 776-77 (1980). Moreover, evidence of prior bad acts is inadmissible if it would engender unfair prejudice substantially outweighing its probative value. Ariz.R.Evid. 403. Thus, when such evidence is admitted solely to show identity, the trial court has a "special obligation to ensure that the probative value of the evidence for the purpose offered is sufficiently great in the context of the case to warrant running that risk." Arizona Evidence § 84, at 180; see also State v. Taylor, 169 Ariz. 121, 125, 817 P.2d 488, 492 (1991).

We believe that these requirements were sufficiently met in this case. First, in regard to the precautionary instruction, the judge instructed the jury before deliberations:

The defendant has been charged with separate crimes in the various counts in this indictment. The jury should give separate consideration, and render separate verdicts with respect to each count. The defendant is entitled to have his guilt or innocence as to each of the crimes charged determined from his own conduct and from the evidence which applies to him if he were being tried on that count alone. If the jury finds the defendant guilty beyond a reasonable doubt of any one of the crimes charged in the indictment, a verdict of guilty should be returned as to him on that count. The guilt or innocence on any of the counts charged should not influence the jury's verdict respecting the other counts.

Reporter's Transcript ("R.T.") May 23, 1990, at 59-60. Although this instruction did not directly address the impropriety of using the other act evidence to prove Defendant's bad character and actions in conformity with that character, it did at least caution the jury about the need to consider the evidence separately on each charge.^[5]

Finally, we do not believe that the high probative value of this identity evidence was substantially outweighed by the danger of unfair prejudice. See Ariz.R.Evid. 403. The evidence was separate and distinct for each count. The judge also submitted separate verdict forms tailored to each count, further underscoring their autonomous nature. The record shows, moreover, that the jury followed the judge's instructions and considered the evidence separately on each charge. Indeed, the jury acquitted Defendant of all the robbery and theft charges that accompanied the three murders and also acquitted him of the attempted sexual assault of Mrs. W.

In sum, the evidence of the attack on Mrs. V would have been admissible in a separate trial for the three murders. Moreover, the trial judge's instruction arguably prevented its improper use, its probative value was not substantially outweighed by the danger of unfair prejudice, and the results demonstrate the jury's careful and proper consideration of the evidence. Cf. Hall, 307 N.W.2d at 298. Finding no prejudice, we hold that the court's refusal to sever was neither reversible error nor a deprivation of due process.

2. Detective Chambers' testimony

Defendant next contends the trial court erred when it denied his motion for a mistrial after a police detective revealed that Defendant had previously been imprisoned. At trial, Detective Chambers testified extensively to his post-arrest examination of Defendant. During cross-examination, the following exchange occurred:

Q. [Defendant's attorney]: And during [the] hour and a half [interview], did you discuss all of the things that you've talked about today, or were other things discussed but just weren't important enough to write down?

A. [Chambers]: I'm sure we discussed other things that aren't in the report.

Q. [Defendant's attorney]: What other things would you have discussed?

A. [Chambers]: Well, I recall we discussed, or attempted to discuss his having been in prison. He would confirm that he had.

R.T., May 22, 1990, at 39-40 (emphasis added). Immediately following this colloquy, defense counsel moved for a mistrial. Although admitting the question was "close," the judge denied Defendant's motion, commenting in a side-bar conference that "[y]ou did ask that question and he gave the answer to it."

a. Was the prohibited testimony "invited"?

The state properly concedes that evidence of Defendant's prior criminal record was inadmissible in this case where Defendant did not testify. State v. Bailey, 160 Ariz. 277, 280, 772 P.2d 1130, 1133 (1989). The state argues, however, that defense counsel's questions invited any error.

"The invited error doctrine applies to situations `where evidence adduced or comments made by one party make otherwise irrelevant evidence highly relevant or require some response or rebuttal.'" State v. Woods, 141 Ariz. 446, 455, 687 P.2d 1201, 1210 (1984) (quoting Pool v. Superior Court, 139 Ariz. 98, 103, 677 P.2d 261, 266 (1984)). The doctrine prevents a defendant from introducing forbidden evidence and then seeking reversal based on its erroneous introduction. We have so held on numerous occasions. See, e.g., State v. Cook, 170 Ariz. 40, 52, 821 P.2d 731, 743 (1991), cert. denied, ___ U.S. ___, 113 S.Ct. 137, 121 L.Ed.2d 90 (1992); State v. Stoneman, 115 Ariz. 594, 596, 566 P.2d 1340, 1342 (1977). To hold otherwise would allow defendants to avoid conviction simply by eliciting inadmissible testimony. Cf. State v. Gilreath, 107 Ariz. 318, 320, 487 P.2d 385, 387 (1971), cert. denied, 406 U.S. 921, 92 S.Ct. 1781, 32 L.Ed.2d 121 (1972). While neither side in this case purposefully interjected inadmissible testimony, we believe that responsibility for its introduction lies with defense counsel. Defense counsel knew that Chambers learned of Defendant's criminal record during his post-arrest interview. At a pre-trial voluntariness hearing, the prosecutor questioned Chambers about this issue in the presence of defense counsel:

Q. [the prosecutor]: At the time you were interviewing the defendant, had you been made aware of whether or not he had an arrest history?

* * * * * *

A. [Chambers]: He told me himself of being in prison in California and Arizona, but he refused to discuss details of that with me.

R.T., May 11, 1990, at 49 (emphasis added). Given this pre-trial testimony, defense counsel's question squarely and directly invited Chambers to reiterate his prior testimony.

Defendant makes much of Detective Chambers' extensive experience and argues that any culpability lies with Chambers because he should have known that his response was inadmissible. It is true that an experienced police officer should understand that such testimony is generally prohibited and, in any event, ought to be so admonished before testifying. See, e.g., State v. Brewer, 110 Ariz. 12, 15-16, 514 P.2d 1008, 1011-12 (1973). It is equally true, however, that an able lawyer conducting cross-examination can usually avoid the injection of known inadmissible testimony by using narrow, leading questions. See Ariz.R.Evid. 611(c). Chambers did not volunteer the unwanted testimony. The broad question posed to him specifically called for the response now challenged. See State v. Fish, 109 Ariz. 219, 220, 508 P.2d 49, 50 (1973); State v. Maggard, 104 Ariz. 462, 465, 455 P.2d 259, 262 (1969). In light of Chambers' previous testimony and the question's broad nature, we find any error invited.

b. Was the error fundamental?

When defense counsel unintentionally "invites" error, the error must be fundamental before relief will be granted. State v. Libberton, 141 Ariz. 132, 138, 685 P.2d 1284, 1290 (1984).^[6] We have described fundamental error "as error going to the foundation of the case or that which takes from the defendant a right essential to his defense, and as error of such magnitude that the defendant could not possibly have had a fair trial." Id. (citations omitted).

The trial court possesses broad discretion in deciding whether to grant a mistrial, and failure to do so is error only if it was a clear abuse of discretion. State v. Koch, 138 Ariz. 99, 101, 673 P.2d 297, 299 (1983) (citing cases). In deciding whether a mistrial is required the trial court should consider:

(1) whether the remarks called to the attention of the jurors matters that they would not be justified in considering in determining their verdict, and

(2) the probability that the jurors, under the circumstances of the particular case, were influenced by the remarks.

State v. Hallman, 137 Ariz. 31, 37, 668 P.2d 874, 880 (1983), cited and relied on in Bailey, 160 Ariz. at 279, 772 P.2d at 1132.

Here the jury could not properly consider Chambers' comment; therefore, the first prong of Hallman/Bailey is satisfied. Under the second prong, we must determine whether, in the context of this case, the trial court abused its discretion in concluding that the jury was not so influenced by Chambers' comment that Defendant was denied a fair trial. See Bailey, 160 Ariz. at 280, 772 P.2d at 1133.

The comment was one short remark at the trial's very end. Defense counsel tactfully, and without objecting in front of the jury, asked for a side-bar conference to voice his concerns and preserve his objection. The attention focused on the comment was therefore minimized, and no one made any further mention of Defendant's criminal record to the jury. Nor did the jury learn of the reasons for Defendant's prior imprisonment or its duration. See Bailey, 160 Ariz. at 280, 772 P.2d at 1133. Moreover, the following day the judge instructed the jury:

I'd like to give you one instruction, it is not part of the instructions that I will read to you at the end of the case, its as follows: Detective Chambers' testimony concerning matters not contained in his report is stricken from the record. You are to disregard such evidence and not discuss it in your deliberations.

R.T., May 23, 1990, at 4-5. The judge also later instructed that stricken testimony was not to be considered and that jurors should not be concerned with the reasons for such a decision. Given the fleeting mention of the inadmissible matter, we believe the judge's instructions mitigated any potential prejudice. See State v. Correll, 148 Ariz. 468, 477, 715 P.2d 721, 730 (1986). In light of these factors, we cannot say that any prejudice was so great that the inadvertent admission of this evidence amounted to fundamental error.

3. Inflammatory photographs

Defendant next claims that the trial court erred when it admitted, over his objection, sixteen color photographs of the murder victims. Defendant does not claim that the photos were merely inflammatory; he argues instead that they were irrelevant and inadmissible because he conceded everything except that he was the killer. As support Defendant relies on State v. Chapple, 135 Ariz. 281, 660 P.2d 1208 (1983) (it is error to admit inflammatory photographs unrelated to the only contested issue in the case).

The trial court has discretion to admit photographs and will not be reversed absent a clear abuse of that discretion. Bailey, 160 Ariz. at 280, 772 P.2d at 1133. A two-part test is employed to determine whether photographic evidence is admissible. Id.; see also State v. Moorman, 154 Ariz. 578, 586, 744 P.2d 679, 687 (1987). First, the photographs must be relevant to an issue in the case. Bailey, 160 Ariz. at 280, 772 P.2d at 1133. Photographic evidence is relevant if it helps the jury understand any disputed issue. Id. Second, if the photos would tend to incite passion or inflame the jury, their probative value must be weighed against any unfair prejudice caused by admission. Id. Photographs having no tendency to prove or disprove any contested issue have little use or purpose except to inflame and usually are not admissible. Moorman, 154 Ariz. at 586, 744 P.2d at 687; Chapple, 135 Ariz. at 288, 660 P.2d at 1215.

In this case the photographs were relevant. Although Defendant contested neither the manner of death nor the identity of the victims, we have held that such concessions do not necessarily render photographic evidence inadmissible. See State v. Amaya-Ruiz, 166 Ariz. 152, 171, 800 P.2d 1260, 1279 (1990), cert. denied, ___ U.S. ___, 111 S.Ct. 2044, 114 L.Ed.2d 129 (1991); Chapple, 135 Ariz. at 290, 660 P.2d at 1217. The photographs in this case helped the state's witnesses explain the various wounds and crime scenes. See, e.g., Amaya-Ruiz, 166 Ariz. at 171, 800 P.2d at 1279.

Unlike the photographs in Chapple, these photos were relevant to the issue of the perpetrator's identity, which, according to Defendant, was the "only" real issue in the homicide cases. In each of the four cases there were varying degrees of incriminating evidence. The similarities of the attacks and wounds (as shown in both the crime scene and autopsy photos) made it more likely that Defendant committed each crime. Cf. Roscoe, 145 Ariz. at 223, 700 P.2d at 1323 (finding no abuse of discretion in admitting photo that showed similarity between a prior bad act and the charged offense).

Defendant also challenged the state's contention that Mrs. C, Mrs. L, and Mrs. W were victims of an attempted sexual assault. Defense counsel, in each case, cross-examined the medical examiner about the absence of physical evidence supporting the attempted sexual assault charges and argued in closing that the lack of such evidence entitled Defendant to an acquittal on these charges. Thus, the photos depicting the position of the bodies and the arrangement of the clothing were particularly important. Analogous facts were absent in Chapple. Therefore, we cannot find that the trial judge abused his discretion by concluding that the photographs' probative value outweighed any danger of unfair prejudice created by their admission. Amaya-Ruiz, 166 Ariz. at 170, 800 P.2d at 1278; Bailey, 160 Ariz. at 280-81, 772 at 1133-34. There was no error.

4. Circumstantial evidence instruction

Defendant next argues that the trial court committed fundamental error by giving the following instruction:

A fact may be proved by either direct or circumstantial evidence. Direct evidence is the testimony of one who asserts actual knowledge of a fact, such as an eyewitness; circumstantial evidence is proof of a set of facts and circumstances indicating the guilt or innocence of a defendant. The law makes no distinction between the weight to be given to either direct or circumstantial evidence; it requires only that after weighing all the evidence you be convinced of the guilt of the defendant beyond a reasonable doubt before he can be convicted.

Defendant contends that the jury should have been instructed as follows:

Evidence can be divided into direct and circumstantial evidence. Direct evidence is the testimony of a witness that saw an event. Circumstantial evidence is the proof of a fact from which the existence of another fact may be inferred. You must determine the weight to be given to all the evidence without regard to whether it is direct or circumstantial.

The state offered the latter instruction, but the judge refused it as "covered." Defendant neither offered his own instruction nor objected to the one given.

Defendant argues that the jury was "misinformed about the very process it [was] to engage in" because the given instruction "fail[ed] to inform the jury that circumstantial evidence involves inferring one fact from another." Defendant, however, cites no authority for this proposition.^[7] He also argues that it was even more significant the jury was told that there were only two conclusions that could be drawn from the circumstantial evidence, "guilt or innocence." According to Defendant, innocence is a "metaphysical concept" that is of no concern to the jury, and the only true question is whether the evidence shows Defendant's guilt beyond a reasonable doubt. Defendant again cites no authority.

Normally, failure to object to a jury instruction precludes a later claim of error. State v. King, 158 Ariz. 419, 424, 763 P.2d 239, 244 (1988); State v. Hunter, 142 Ariz. 88, 90, 688 P.2d 980, 982 (1984); Ariz.R.Crim.P. 21.3(c). We will therefore reverse only for fundamental error. King, 158 Ariz. at 424, 763 P.2d at 244; Hunter, 142 Ariz. at 90, 688 P.2d at 982.

Arizona law makes no distinction between circumstantial and direct evidence. State v. Harvill, 106 Ariz. 386, 391, 476 P.2d 841, 846 (1970). Defendant does not object to the giving of an instruction on this matter but rather to the form of the instruction given. Although the instruction was not a model of clarity, we believe that, taken as a whole, it adequately stated the law.^[8] In fact, the instruction told the jury that "[t]he law makes no distinction between the weight to be given to either direct or circumstantial evidence; it requires only that after weighing all the evidence you be convinced of the guilt of the defendant beyond a reasonable doubt before he can be convicted." R.T., May 23, 1990 at 59 (emphasis added). Therefore we find that even if error occurred, it was not fundamental. Cf. State v. Duarte, 165 Ariz. 230, 232, 798 P.2d 368, 370 (1990) (finding no fundamental error in an erroneous instruction where the sentence immediately preceding it set out the state's burden of proof).

We also reject Defendant's argument concerning the use of the word "innocence." Elsewhere the judge clearly laid out the relative burdens of proof. He stated:

The law does not require a defendant to prove his innocence. He is presumed by law to be innocent. This means the state must prove all of its case against the defendant.

R.T., May 23, 1990, at 57 (emphasis added). The judge, however, did not stop there. He went on to explain exactly what it meant to prove the state's case:

The state must prove the defendant guilty beyond a reasonable doubt.

The burden throughout the trial is always on the state to prove not only all of the elements of the crime, but also the identity of the person who committed the crime, beyond a reasonable doubt. The burden never shifts.

R.T., May 23, 1990, at 57-58. The judge, therefore, moved the concept from the metaphysical to the real world. In light of this, we again find no fundamental error.

B. GENERAL CONSTITUTIONAL ISSUES

1. Equal protection claim

Defendant argues that his Fourteenth Amendment equal protection right was violated when the judge denied him a jury trial on the existence of the aggravating factors. This court recently rejected this argument. See State v. Spencer, 176 Ariz. 36, 45, 859 P.2d 146, 155 (1993); State v. Landrigan, 176 Ariz. 1, 6, 859 P.2d 111, 116 (1993). We need not revisit the issue.

2. Eighth Amendment

Defendant also argues that Arizona's death penalty scheme, taken as a whole, violates the Eighth Amendment by failing to genuinely narrow the class of persons eligible for the death penalty. We have also recently rejected this argument. See State v. Spencer, 176 Ariz. 36, 45, 859 P.2d 146, 155 (1993); State v. Landrigan, 176 Ariz. 1, 6, 859 P.2d 111, 116 (1993). We need not revisit the issue.

C. SENTENCING ISSUES

In all capital cases this court independently reviews the record to determine whether the death sentence is appropriate. State v. Hill, 174 Ariz. 313, 326, 848 P.2d 1375, 1388 (1993), cert. denied, ___ U.S. ___, 114 S.Ct. 268, 126 L.Ed.2d 219 (1993); State v. Gretzler, 135 Ariz. 42, 54, 57, 659 P.2d 1, 13, 16 (1983), cert. denied, 461 U.S. 971, 103 S.Ct. 2444, 77 L.Ed.2d 1327 (1983). "Thus, we review the sentencing hearing and the aggravating and mitigating circumstances to ensure that proper procedures were followed and the proper factors determined and weighed." Hill, 174 Ariz. at 326, 848 P.2d at 1388. In this case, the judge considered the trial and sentencing evidence, the presentence report, counsel's sentencing memoranda, and several letters and articles. We now undertake our review of this evidence.

1. Aggravating factors

The trial judge held a separate aggravation-mitigation hearing and returned a special verdict. See A.R.S. §§ 13-703(B), (D). The court found two statutory aggravating factors: (1) Defendant had been previously convicted of another offense (the other murder counts and a previous robbery) in the United States for which a sentence of life imprisonment or death could be imposed under Arizona law, A.R.S. § 13-703(F)(1); and (2) Defendant committed each murder in an especially heinous, cruel, or depraved manner, A.R.S. § 13-703(F)(6). Defendant contests only the second finding. In light of our holding that the trial judge improperly interpreted or misapprehended significant mitigating evidence sufficiently substantial to call for leniency in the weighing process, see A.R.S. § 13-703(E), it is unnecessary to prolong this review by repeating the evidentiary facts set forth ante in "Facts and Procedural History." Suffice it to say that the victims must have suffered terrible pain during the beatings and stabbings. We therefore conclude that the evidence established beyond a reasonable doubt that the § 13-703(F)(6) aggravating circumstance was present.

2. Mitigating factors

The trial judge found that Defendant's capacity to conform his conduct to the requirements of law was significantly impaired, but rejected the claim that Defendant's capacity to appreciate the wrongfulness of his conduct was affected. He concluded that this mitigating evidence was not sufficient to outweigh the aggravating circumstances and that there was nothing else presented that would call for leniency. The trial judge then imposed the death sentence for each murder.

3. Propriety of the death sentences

During the course of this appeal, neither Defendant nor the state addressed the issue of Defendant's psychiatric condition, even though the issue was addressed in the trial court. We note, however, that a sentencing judge has a duty to consider "any aspect of the defendant's character or record and any circumstance of the offense relevant to determining whether a sentence less than death may be appropriate." State v. West, 176 Ariz. 432, 862 P.2d 192 (1993) citing State v. McCall, 139 Ariz. 147, 162, 677 P.2d 920, 935 (1983), cert. denied, 467 U.S. 1220, 104 S.Ct. 2670, 81 L.Ed.2d 375 (1984). This court, in turn, has a duty not only to review the support for the trial court's findings but also to independently review the sentence's propriety. Spencer, 176 Ariz. at 7, 859 P.2d at 152; State v. White, 168 Ariz. 500, 520-21, 815 P.2d 869, 889-90 (1991) (Corcoran, J., specially concurring), cert. denied, ___ U.S. ___, 112 S.Ct. 1199, 117 L.Ed.2d 439 (1992); State v. Fulminante, 161 Ariz. 237, 254, 778 P.2d 602, 619 (1988), aff'd, 499 U.S. 279, 111 S.Ct. 1246, 113 L.Ed.2d 302 (1991). We have held that

the gravity of the death penalty requires that we painstakingly examine the record to determine whether it has been erroneously imposed. Furthermore, because [our statute] sets out the factors which must be found and considered by the sentencing court, we necessarily undertake an independent review of the facts that establish the presence or absence of aggravating and mitigating circumstances. We must determine for ourselves if the latter outweigh the former when we find both to be present.

State v. Richmond, 114 Ariz. 186, 196, 560 P.2d 41, 51 (1976) (citations omitted), cert. denied, 433 U.S. 915, 97 S.Ct. 2988, 53 L.Ed.2d 1101 (1977).

Pursuant to our duty of independent review, we have examined the entire record. We find the issue of Defendant's psychiatric condition to be a pivotal aspect of the sentencing process in this case, and one that is impossible to ignore. We would be remiss if we failed to consider such crucial information solely because the parties did not argue it on appeal.^[9]

Faced with this task we are sometimes called upon to reduce a death sentence to life imprisonment even in cases where the facts are aggravated and the tragedy immense. See, e.g., State v. Mauro, 159 Ariz. 186, 207-08, 766 P.2d 59, 80-81 (1988) (reducing sentence because of defendant's mental illness where defendant brutally killed his seven-year-old son). Three mental health experts evaluated Defendant in the present case: two were retained by Defendant and one by the state. We now consider the assessment of each.

a. Dr. Tatro

Dr. Donald Tatro, a psychologist retained by defense counsel, examined Defendant first. In a twenty-three page report, Dr. Tatro described in detail the results of his examination, and concluded that Defendant suffered from a "serious personality disorder." He stated:

The adult Mr. Stuard, thus, is a man who is very out of touch with his own less-than-positive characteristics. He does not recognize his intellectual shortcomings. He is not aware of his own socially unacceptable hostile and sexual impulses. He does not see himself as having motives other than the ones that fit his high-blown, fault-free estimate of himself. He cannot admit to failings of any kind, even cognitive failings that are obvious to the casual observer.

* * * * * *

When the stresses build, however, and negative impulses begin to intensify, instead of recognizing that it is he who is feeling negative, the surface look of things about him changes in a negative direction. It is as if an inner poison builds up and spills over onto the outer skin of the world around him, and all of the objects in it — excluding him and the people he includes in his image of himself — take on malignant coloration. His mood becomes correspondingly malignant, since he sees himself as the object of all of the negativity around him. He is much more inclined to respond in a hostile, aggressive way to people when this kind of murkiness descends over the surfaces to which he typically reacts.

Tatro Report at 13-14. Dr. Tatro continued:

The defensive habits which give shape to Mr. Stuard's personality are best understood as indications of serious personality disorder, which is maladaptive, usually longstanding, pattern of coping with internal conflicts and environmental stresses. The personality disorder that is in evidence in Mr. Stuard's case appears to be a mixed one, combining histrionic, paranoid, and narcissistic traits.

Id. at 18. The report also revealed a troubled childhood, including the lack of a maternal upbringing, an absent father, and an oppressive grandmother who would, for example, force Defendant and his brother to "get naked" and whip each other with a "razor strap" as punishment.^[10] Dr. Tatro also determined that Defendant's verbal IQ (73) was toward the low end of the borderline range, which is "just one step above the mentally retarded range."

In addition to a personality disorder, Dr. Tatro also inferred that Defendant had sustained organic brain damage as a result of his boxing career. He stated:

Intelligence-test indications that Mr. Stuard may have suffered some impairment of earlier higher levels of cognitive functioning raise the question of possible brain damage resulting from his career as a prizefighter. Brain damage is a common consequence of lengthy involvement in a sport where the object is the continual violent pummeling of the opponent's head. It was not surprising, therefore, when Mr. Stuard's handling of the Bender Visual-Motor Gestalt Test also turned out to be consistent with the likelihood of organic brain damage.

Id. at 10-11. Importantly, Dr. Tatro surmised that the organicity "may have contributed both to the stresses that were building up in [Defendant's] life and to the weakening of defenses that, up till this late point in his life, [Defendant] had successfully employed to keep repressed...." Accordingly, Dr. Tatro concluded that there is a strong possibility that the organic brain damage "may have contributed significantly to [Defendant's] acting-out of violent impulses." (Emphasis added). He recommended that Defendant be tested further to determine whether Defendant was suffering from a condition known as "organic personality syndrome."

Dr. Tatro's testimony at sentencing tracked his report in most respects. However, after reviewing the results from the subsequent examinations (discussed below), he revised his initial diagnosis and agreed with the state's expert that Defendant was in fact suffering from "dementia," a related but more serious ailment. Dr. Tatro nevertheless adhered to his initial belief that Defendant was brain damaged — a finding that the state's expert did not dispute and ultimately confirmed.

b. Dr. Blackwood

Defense counsel retained Dr. H. Daniel Blackwood, a clinical neuropsychologist, to follow-up on Dr. Tatro's diagnosis. Dr. Blackwood confirmed Dr. Tatro's suspicions that Defendant had organic brain dysfunction:

These results serve to confirm Dr. Tatro's concern about organic brain dysfunction as a complicating factor in Mr. Stuard's presentation. The current test results do not suggest a specific etiology for Mr. Stuard's brain dysfunction. Developmental factors and his career as a boxer as described by Mr. Stuard and Dr. Tatro could certainly account for the current results.

Based upon the information available to me, I have no reason to think that Mr. Stuard's mental condition at the time of his offenses was any different than it is at this time. He clearly does suffer from mental deficiency and generalized brain dysfunction.

Blackwood Report at 4-5 (emphasis added). Like Dr. Tatro, Dr. Blackwood's IQ testing placed Defendant in the "borderline range," his full scale score (74) being in the bottom four percent of the population.

c. Dr. Scialli

The state retained Dr. John V. Scialli, a psychiatrist, who also examined Defendant. Like the others, Dr. Scialli's testing also indicated that Defendant had "serious brain damage." He concluded that Defendant suffered from "dementia." At the sentencing hearing, Dr. Scialli described this dementia as:

a neuropsychiatric condition. The cause of which is brain damage, either from a known cause or from a cause which is highly suspected and most likely to be a cause of brain damage. Dementia is a term that encompasses a number of different symptoms and refers to a general acquired defect in different brain functions ... which may include reasoning and judgment, of memory, impulse control, expression of personality.

R.T., Nov. 26, 1990, at 15.^[11] In response to questioning regarding how Defendant's mental illness may have affected him during one of the attacks, Dr. Scialli testified that

when confronted by a victim, his impulse control is so tenuous, so hair triggered, impaired by his dementia, that he would have, I would suspect, flown into a rage at the time and not handled a situation that someone with more reasoning ability might have handled with considerably less force.

Id. at 17 (emphasis added). When asked to explain this "rage," Dr. Scialli testified:

The rage as associated with dementia is typically those that are provoked by things which might not provoke rage in others. Usually a rage, which is relatively uncontrolled, or poorly controlled and not directed towards a particular — not necessarily toward a particular goal, there may be — there usually is a sudden acceleration of the rage, not a slow burn, as some people might have, but sudden explosive rage with a degree of violence.

Id. (emphasis added).

Dr. Scialli refuted any implication that organic factors caused Defendant's non-violent criminal behavior. Instead, he believed that Defendant's mental illness impaired his ability to carry out his criminal goals efficiently. Essentially, the impairment caused Defendant to murder when he originally ventured out to commit a much less serious crime. The murders, therefore, were the unfortunate result of his mental impairment. The dementia caused Defendant to "[react] suddenly and overwhelmingly when he confronted and was confronted by his victims."^[12] Like the other doctors, Dr. Scialli believed Defendant's boxing career could have caused his brain damage.

d. Discussion

With only minor variations in theme and certainty, all three experts^[13] agreed that Defendant was mentally impaired at the time of the murders and that this impairment contributed to the homicides. Other evidence indicated that Defendant had been mentally ill for some time. In addition to the testimony of the mental health experts, defense counsel presented evidence that Defendant was denied a license to box in the early 1960's because of brain damage suffered from previous boxing injuries.^[14]

The trial judge, in his special verdict, concluded:

[T]he Court finds that while the defendant suffers from some mental deficiency and generalized brain dysfunction which compromises his ability to respond effectively to the events in his environment, it does not significantly impair his capacity to appreciate the wrongfulness of his conduct. Neither of the two psychologists nor the one psychiatrist that examined the defendant found that he suffered from any mental illness or disease or that he was psychotic or suffered from a compulsive personality disorder.

* * * * * *

The defendant's capacity to appreciate the wrongfulness of his conduct was not affected, although his capacity to conform his conduct to the requirements of law was significantly impaired, but not so impaired as to constitute a defense to the prosecution.

Special verdict at 8-9 (citations omitted) (emphasis added).

We agree with the trial judge's findings that Defendant's capacity to conform his conduct to the requirements of law was significantly impaired. However, A.R.S. § 13-703(G)(1) is written in the disjunctive and requires the presence of only one of the two elements — significant impairment of capacity to appreciate or capacity to conform. See State v. Rossi, 154 Ariz. 245, 251, 741 P.2d 1223, 1229 (1987); Richmond, 114 Ariz. at 197, 560 P.2d at 52. Therefore, we conclude that the statutory mitigating factor of A.R.S. § 13-703(G)(1) was established. See Rossi, 154 Ariz. at 251, 741 P.2d at 1229.

The trial judge did not explicitly state in his special verdict that he found any mitigating factors to exist. Neither did he specify that all but the (G)(1) factor did not exist. We conclude from the language used by the trial judge that he most probably found § 13-703(G)(1) despite his failure to say so. We believe, however, that the trial judge did not give sufficient weight to the psychiatric testimony in the weighing process.

The trial judge's conclusion that Defendant did not suffer "from any mental illness or disease" is plainly incorrect. The undisputed and unanimous evidence showed that Defendant had significant organic brain damage and suffered from dementia — a recognized and significant mental illness. See American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders (3d ed. rev. 1987) ("DSM-III-R").^[15] The illness is characterized, in part, by

impairment in short- and long-term memory, associated with impairment in abstract thinking, impaired judgment, other disturbances of higher cortical function, or personality change. The disturbance is severe enough to interfere significantly with work or usual social activities or relationships with others.

* * * * * *

Impaired judgment and impulse control are also commonly observed. Coarse language, inappropriate jokes, neglect of personal appearance and hygiene, and a general disregard for the conventional rules of social conduct are evidence of bad judgment and poor impulse control. A previously cautious businesswoman may embark on a reckless business venture. An elderly man who never married may make sexual advances to strangers. A retiree may shoplift without considering the consequences. Marked impairment of judgment and impulse control is particularly characteristic of certain Dementias that affect primarily the frontal lobes.^[16]

Id. at 103-04 (emphasis added).

The evidence described (see ante §§ 3(a) and (c)) indicates that the conditions and mental illness diagnosed by the experts were significant causative factors of the crimes. We conclude that this mitigating evidence has not been given the weight it merits. Even if the trial judge made a finding pursuant to A.R.S. § 13-703(G)(1), this does not affect our duty to make an independent review of the propriety of the death sentence. See, e.g., State v. Jimenez, 165 Ariz. 444, 459-60, 799 P.2d 785, 800-01 (1990); Mauro, 159 Ariz. at 207-08, 766 P.2d at 80-81; State v. Brookover, 124 Ariz. 38, 42, 601 P.2d 1322, 1326 (1979). The dissent notes the trial judge's careful study of the record. See post at 614, 863 P.2d at 906. While this is true, it is well settled that Arizona law also requires this court to do justice in death cases, including if necessary, substituting its judgment for that of the trial judge, "something anathema in any other context." State v. Salazar, 173 Ariz. 399, 418, 844 P.2d 566, 585 (Martone, J. specially concurring) (1992), cert. denied, ___ U.S. ___, 113 S.Ct. 3017, 125 L.Ed.2d 707 (1993).

There is no question that the sentencing hearing and the accompanying reports produced substantial mitigating evidence applicable to all three first degree murders. Our sentencing statute does not require the mitigating factors to "outnumber" the aggravating factors to avoid a death sentence. See A.R.S. § 13-703(E). "One mitigating circumstance, for example may be `sufficiently substantial' to outweigh two aggravating factors." Brookover, 124 Ariz. at 42, 601 P.2d at 1326.

After carefully considering all of the evidence, we conclude that Defendant's mental impairment, which included severe organic brain damage, resulting dementia, an IQ bordering on retarded, and, according to Dr. Tatro, a serious personality disorder, "was not only a substantial mitigating factor in this case, but a major contributing cause of his conduct that was `sufficiently substantial' to outweigh the aggravating factors present." Jimenez, 165 Ariz. at 459, 799 P.2d at 800. The lack of motive or any other discernable reason or explanation for these killings reinforces our conclusions. Leniency is therefore required. See, e.g., State v. Herrera, 174 Ariz. 387, 400, 850 P.2d 100, 113 (1993); Jimenez, 165 Ariz. at 459-60, 799 P.2d at 800-01; State v. Rockwell, 161 Ariz. 5, 15-16, 775 P.2d 1069, 1079-80 (1989); Mauro, 159 Ariz. at 207-08, 766 P.2d at 80-81; Brookover, 124 Ariz. at 42, 601 P.2d at 1326; State v. Doss, 116 Ariz. 156, 162, 568 P.2d 1054, 1060 (1977).

Because in this case the mitigating evidence applied to all three homicides, we reduce the three sentences for the first degree murders from death to life in prison. However, because of the mental illness which affects Defendant he obviously is and will continue to be very dangerous to others. Therefore, each of the sentences is to be served consecutively to each other and to be followed by the prison terms imposed for the other convictions, so that Defendant will never be released.

Our holding in no way discounts the severity or tragedy of these crimes. Instead, because we are bound by law to consider significant mitigating evidence, we must hold the death penalty inappropriate here where Defendant's organically-caused mental illness was such a significant causative factor. Cf. State v. Bible, 175 Ariz. 549, 858 P.2d 1152 (1993).

CONCLUSION

We have independently reviewed the record for fundamental error on all counts of which Defendant was convicted, and have found none. We affirm all of Defendant's convictions. We reduce Defendant's three death sentences to three life sentences, each without the possibility of parole for twenty-five years, pursuant to A.R.S. §§ 13-703(A) and (E). These life sentences are to be served consecutively, followed by the prison sentences imposed in the noncapital cases.

MOELLER, V.C.J., ZLAKET, J., and LANKFORD, Judge, concur.

CORCORAN, J., did not participate in this decision; pursuant to Ariz. Const. art. VI, § 3, the Honorable JEFFERSON L. LANKFORD of the Court of Appeals, Division One, was designated to sit in his stead. MARTONE, Justice, dissenting.

The record does not support the court's decision to vacate the sentence of death. I do not agree with the court's characterization of the findings of the trial judge or with the court's understanding of the opinions of the three expert witnesses. More importantly, I do not agree that the slim evidence of defendant's impairment outweighed the significant aggravating factors in this case.

The court first describes scant evidence as crucial information and then reasons that this slight evidence outweighs the two significant aggravating factors. But neither the defendant nor his experts considered this slight evidence to be crucial information. The defendant never raised the question of impairment in his brief or at oral argument. Our power of independent review does not extend to the sua sponte creation of psychological theories not supported by the evidence.

1. The Special Verdict

The trial judge found "that while the defendant suffers from some mental deficiency and generalized brain dysfunction which compromises his ability to respond effectively to the events in his environment, it does not significantly impair his capacity to appreciate the wrongfulness of his conduct." Special verdict at 8. He also found that "[t]he defendant's capacity to appreciate the wrongfulness of his conduct was not affected, although his capacity to conform his conduct to the requirements of law was significantly impaired, but not so impaired as to constitute a defense to the prosecution." Id. at 9. In other words, the judge found that the statutory G(1) mitigating factor existed, but that it was not sufficiently substantial to outweigh the aggravating circumstances. Id. at 10.

2. The Experts

A review of the evaluations by the three experts shows that the defendant has some impairment of memory. Drs. Scialli and Tatro recommended neurological testing and a CT scan or MRI to determine whether the defendant had brain damage. These tests were never performed. Thus, it remains unconfirmed whether defendant has organic brain damage. None of the doctors found psychosis. Their tests collectively revealed that the defendant had a low IQ and memory impairment. Moreover, there was no consensus that the deficiency existed at the time of the crimes, and even less certainty as to the role, if any, the deficiency played in the crimes. However, the doctors did discuss the nature of the defendant's impairment and concluded that he did not lose contact with reality and was able to understand the wrongfulness of his conduct.

Dr. Tatro concluded that the defendant had "mixed personality disorder: narcissistic, histrionic, and paranoid features." Tatro report at 23. He also suggested that there might be an organic personality syndrome. He said:

This is not to suggest that Mr. Stuard gets so cut off from reality that he does not know what he is doing or that what he is doing is wrong in the eyes of society, in general, or the law, in particular. Again, there is no indication that Mr. Stuard is psychotic now or that he has been in the past.

Id. at 20.

In addition, Dr. Tatro did not think his tests were sufficient to diagnose organic brain damage. He merely inferred this, id. at 11, and thus recommended "further examination be done to either confirm or rule out the existence of this form of mental defect." Id. at 22.

In discussing whether the unverified impairment played a role in the defendant's commission of the crimes, Dr. Tatro said "[i]t is difficult to assess what role the organicity that is evident in Mr. Stuard's test responses may have played in the commission of the crimes he committed." Id. at 21.

The report of Dr. Scialli concluded that:

It is my opinion that at the time of the crimes Mr. Stuard was able to appreciate the wrongfulness of his conduct. However ... his ability to conform his conduct to the requirements of the law was significantly impaired but not so impaired as to constitute a defense to prosecution.

Scialli report at 1.

Dr. Scialli did believe that defendant had dementia, presumably organic brain dysfunction. He described the condition as memory impairment. Importantly, he stated that he had no way of knowing whether the dementia existed at the time of the commission of the three murders. Id. at 3. In order to reach this conclusion, he would need a brain imaging scan and a neurological consultation. Id. at 4.

Dr. Scialli disagreed with Dr. Tatro's diagnosis of organic personality syndrome. He also disagreed with Dr. Tatro's conclusion that the defendant had some loss of reality contact. He said "[n]either Dr. Tatro's description of his conversation with Mr. Stuard nor my own experience support this." Id. at 9. Indeed, Dr. Scialli said:

In summary, regarding Dr. Tatro's report, he seems to imply that Mr. Stuard could not appreciate the wrongfulness of his conduct, and felt justified in committing the crimes due to a "paranoid stance." I disagree with this.

Id. at 10.

Instead, Dr. Scialli said "Mr. Stuard did appreciate the wrongfulness of his conduct. To a certain extent, he also exercised his choice not to comply with the law." Id.

Dr. Blackwood concluded that the defendant's mental deficiency and generalized brain dysfunction would not significantly impair his capacity to appreciate the wrongfulness of his conduct, but could affect his ability to conform his conduct to the requirements of the law. Blackwood report at 5.

3. Impairment

The court questions the weight the trial judge accorded the psychiatric testimony. Ante, at 38. But the trial judge's evaluation of the evidence and the doctors' reports was accurate.

Dr. Tatro, a clinical psychologist, and not a medical doctor, did not pretend to be in the business of diagnosing organic brain injuries. Other than his interview of the defendant, all he did was administer psychological tests. He did not diagnose the defendant as having brain damage. He simply stated that intelligence tests indicated that the defendant "may have suffered some impairment of earlier higher levels of cognitive functioning that raise [sic] the question of possible brain damage." Tatro report at 10. Dr. Tatro's final diagnosis was "mixed personality disorder." Id. at 23. But we have already held that a "personality disorder alone is insufficient to constitute a mitigating circumstance." State v. Brewer, 170 Ariz. 486, 505, 826 P.2d 783, 802 (1992).

The court characterizes the trial court's conclusion that the defendant did not suffer from a mental illness or disease as "incorrect." Ante, at 609, 863 P.2d at 901. But leading authorities state that dementia is not a mental illness or disease:

Both dementia and delirium are included under the general diagnostic category of organic brain syndromes. But the term is not particularly useful, for it is nonspecific and conveys little information about the conditions it is meant to describe.

* * * * * *

Dementia and delirium are not diseases.

Robert J. Waldinger, Psychiatry for Medical Students 200 (1984) (emphasis in original).

Nor does the DSM-III-R characterize dementia as an illness. See American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders 103-107 (3d ed. rev. 1987).

Even if Stuard had dementia, as the court notes, ante, at 607 n. 11, 863 P.2d at 899 n. 11, it was mild to moderate. Moreover, Dr. Scialli said:

I have no way at present of knowing if this Dementia existed at the time of the commission of the three murders.

Scialli report at 3. The court suggests Dr. Scialli found that "the murders ... were the unfortunate result of his mental impairment." Ante, at 608, 863 P.2d at 900. But this is the court's conclusion, not Dr. Scialli's. He equivocated on the question of causation. At one point, he acknowledged that he did not know exactly what caused the defendant to act during the course of the homicides. R.T. Nov. 26, 1990 at 25. But at another point he said "[r]ather due to his dementia, [the defendant] reacted suddenly and overwhelmingly when he confronted and was confronted by his victims." Scialli Report at 10. In light of equivocal testimony, the trial judge was in the best position to choose one version over another. This court, instead, relies only on that part of the Scialli testimony that supports its conclusion.

We have said that evidence of the existence of a disorder is insufficient. "It does not prove that, at the time of the crimes, the disorder controlled defendant's conduct or impaired his mental capacity to such a degree that leniency is required." Brewer, 170 Ariz. at 505, 826 P.2d at 802. We also recognized that the mitigating strength of the impairment depends upon the degree or nature of the impairment. "Such conditions differ in degree from a slow, dull brain damaged defendant whose judgement and rationality are marginal.... Mental impairments have a far greater mitigating effect because they may evidence an inability of the defendant to control his conduct." Id. at 505, 826 P.2d at 802. Thus, assuming that defendant indeed suffers from dementia, a disorder primarily inhibiting memory, what is its effect?

In Brewer we evaluated whether, despite the defendant's impairment — borderline personality disorder — he could control his actions. We noted that under the precipitating stress of the homicide, Brewer's impairment compromised his ability to reason and he lashed out with anger. Id. at 506, 826 P.2d at 803. However, we found that he possessed the ability to restrain himself, understood the wrongfulness of his conduct, and did not lose touch with reality. We concluded that his disorder did not warrant leniency. Id.

The effect of dementia on Stuard resembles the effect of borderline personality disorder on Brewer. Even if Stuard, like Brewer, became enraged when confronted by his victims, he still displayed some ability to control his actions. For example, after the police secured the neighborhood where the three murders occurred, the rampage stopped. Stuard waited three months to strike again and did so in an entirely different neighborhood. The doctors agree he appreciated the wrongfulness of his conduct and that he did not lose touch with reality.

4. Weighing

But even if defendant were impaired, and even if there were evidence that he was impaired at the time of the crimes, and even if there were evidence of causation, we still must balance that against aggravating factors. The trial judge heard the witnesses — we did not. The trial judge presided over the aggravation/mitigation hearing — we did not. He thoughtfully concluded that whatever the nature of the defendant's condition, it did not require leniency in this case. That is, it was insufficient to outweigh the significant aggravating factors.

I agree with the trial judge. He found two aggravating factors. The defendant did not contest the validity of the § 13-703(F)(1) factor. And, the court is so persuaded by the second factor — that his murders were especially heinous, cruel or depraved — that it says "it is unnecessary to prolong this review.... Suffice it to say that each victim must have suffered terrible pain during the beatings and stabbings they endured." Ante, at 605, 863 P.2d at 897. But this understates the horror and ghastliness of the multiple murders Stuard committed. The frail, elderly victims endured torturous deaths. They were choked, stabbed, and beaten to death by a former boxer. He then arranged each corpse into a perverse and morbid display.

The trial judge carefully weighed the enormity of the defendant's atrocities against his mental condition and concluded that the scale tipped overwhelmingly on the side of aggravating circumstances. Even the defendant does not suggest otherwise on appeal. The defendant brutally murdered three elderly, defenseless women. The manner in which he did it shows cruelty and depravity in the extreme.

We have an obligation to ensure that the death sentence is not imposed in an arbitrary and freakish way. Furman v. Georgia, 408 U.S. 238, 313, 92 S.Ct. 2726, 2764, 33 L.Ed.2d 346 (1972); Proffit v. Florida, 428 U.S. 242, 252-53, 96 S.Ct. 2960, 2967, 49 L.Ed.2d 913 (1976). We are aware of no case in which dementia, a condition primarily affecting memory, has been held to be sufficient to outweigh substantial aggravating factors.^[11] It is arbitrary to impose death on others and then relieve the defendant of the death sentence on inconclusive and insubstantial psychological evidence.

The trial judge's comprehensive 14 page special verdict reflects a very careful and impartial evaluation of the evidence. While justice sometimes requires us to substitute our judgment for that of the trial judge in death cases, this case is not one of them.

NOTES

[1] By the time Defendant was finally arrested, the boots themselves were not available for testing; it was therefore impossible to ascertain whether Defendant's boots definitely made the prints.

[2] Nothing in the record indicates the actual distance between the various crime scenes. We believe that Mrs. V's home, although farther south, would be considered to be in the same general area, even if, as Defendant contends, Mrs. V lived in a "different neighborhood."

[3] A soda can with crushed sides was also found on the counter at Mrs. C's house. Mrs. C's son-in-law testified that Mrs. C was not in the habit of crushing her soda cans, thus supporting the idea that her attacker also drank at some point during this murder.

[4] Defendant was acquitted of the robbery charges in the murder cases as well as the attempted sexual assault of Mrs. W. We need not address the propriety of relying on the acquitted conduct in our review of this issue. Even without it, sufficient similarities exist.

[5] We note that the state argued in its closing:

I just ask you to consider, as you are in there, what kind of a person would take a hammer to the side of an 80-year-old lady and continue to pound and pound ... and go to a knife and stab that person? Is that the same person who is capable of having inflicted the kind of violence on [Mrs. W], the same kind of person who would have strangled [Mrs. C], and not having enough then stab her?

* * * * * *

I think the conclusion that the state has offered to you since the beginning of this trial, and offered to you today, is that the defendant is clearly guilty of all of the assorted crimes that we have charged him with....

R.T., May 23, 1990, at 54-55 (emphasis added). In the absence of a continuing instruction on the use of the evidence, such argument might well be interpreted as a reference to Defendant's bad character as shown by the prior crime and, if so interpreted by the jury, goes much too far toward use as character evidence. See Ariz. R.Evid. 404(a), (b). In light of the judge's instruction, and not faced with objection either at trial or on appeal, however, we find no error.

[6] But cf. Cook, 170 Ariz. at 52, 821 P.2d at 743 (noting that remarks which would normally be fundamental error, if invited, may not be grounds for reversal) (citing State v. Arredondo, 111 Ariz. 141, 144, 526 P.2d 163, 166 (1974)).

[7] But Cf. United States v. Clark, 475 F.2d 240, 249-50 (2d Cir.1973) (finding an erroneous circumstantial evidence instruction, together with other erroneous instructions, required reversal despite the absence of objection).

[8] This instruction mirrors Maricopa County Recommended Jury Instructions For Criminal Cases ("MARJI") no. 207. The instruction offered by the state is identical to Recommended Arizona Jury Instructions (Criminal) ("RAJI") no. 24. Defendant argues that the RAJI instruction made the MARJI instruction "obsolete." We do not address what effect, if any, RAJI no. 24 had on MARJI no. 207.

[9] The dissent suggests this court is sua sponte creating "psychological theories not supported by the evidence." Post at 611, 863 P.2d at 903. This is incorrect. Although the issue was not argued on appeal, the record is replete with evidence given by qualified experts concerning Defendant's psychological condition.

[10] The report is replete with evidence that Defendant was not purposefully offering evidence favorable to a finding of mitigation. In fact, it appears that Defendant tried to conceal his troubled background to avoid giving what he characterized as a "degrading" picture. When asked questions regarding his family, Defendant immediately "balked." Dr. Tatro noted that "[h]is manner, throughout, was very guarded ... and he seemed set to avoid any suggestion of problems between him and his loved ones." Tatro Report at 5.

[11] The dissent correctly notes Dr. Scialli's disagreement with Dr. Tatro's diagnosis of organic personality syndrome. However, Dr. Scialli did find, in his capacity as a board certified Psychiatrist, that Defendant suffered from Dementia, a far more serious impairment. Dr. Scialli testified that Defendant's dementia was "somewhere between mild and moderate, mild in that he is able to care for his personal hygiene ... moderate in terms of the extent of brain functions which have been affected." R.T., Nov. 26, 1990 at 21. Moreover, contrary to the dissent's implication that Defendant was unimpaired, post at 613, 863 P.2d at 905, Dr. Scialli testified that defendant was "severely impaired in terms of brain damage," and also described the damage as "significant." Id. at 33-34. Dr. Scialli's report also states that Stuard's "performance on [the] cognitive examination is extremely poor, and does indicate serious brain damage." Scialli Report at 9 (emphasis added).

[12] The dissent cites State v. Brewer, 170 Ariz. 486, 506, 826 P.2d 783, 803 (1992) cert. denied ___ U.S. ___, 113 S.Ct. 206, 121 L.Ed.2d 147 (1992) for the proposition that evidence of causation is required before mental impairment can be considered a significant mitigating factor. Post at 613, 863 P.2d at 905. We agree. Here, Dr. Scialli, the state's own expert, stated "I do not believe he pre-meditated murder. Rather, due to his Dementia, he reacted suddenly and overwhelmingly when he confronted and was confronted by his victims." Scialli Report at 10 (emphasis added). Moreover, Dr. Scialli suspected that Stuard's short-term memory impairment caused him not to "`learn from his mistakes' and proceeded within days to use the same modus operandi, encounter the same situation and over-react in the same way." Scialli Report at 11. We part company with the dissent, however, insofar as it implies that Defendant could fully control his actions. Post at 613, 863 P.2d at 905. This contradicts the weight of the evidence, as well as the trial court's explicit finding. See special verdict at 9.

[13] Dr. Scialli concurred with Dr. Blackwood's diagnosis and "generally" agreed with Dr. Tatro's assessment. R.T., Nov. 26, 1990 at 20; Scialli Report at 9.

[14] Contrary to the dissent's implications, post at 611, 612, and 613, 863 P.2d at 903, 904 and 905, we believe this testimony is evidence that Stuard's impairment predated the murders, and therefore existed when they were committed. Moreover, Dr. Scialli stated, "It is my opinion that [Stuard's brain dysfunction] has existed for some time, including the time period around his arrest. I base this on his presentation on the videotaped press conference the day of his arrest. His confusion and reluctance to answer some questions is similar to that which he demonstrated when I interviewed him." Scialli Report at 3. See also Dr. Blackwood's report, ante § 3(b). While there is no infallible proof of impairment at the time of the murders, all available evidence indicates that he was impaired, and none indicates otherwise.

[15] The dissent argues that "leading authorities" do not characterize dementia as an "illness" or "disease." Post at 612, 863 P.2d at 904. While even eminent scholars occasionally refer to some forms of dementia as a "disease," see, e.g., Harold I. Kaplan and Benjamin J. Sadock, Comprehensive Textbook of Psychiatry (5th ed. 1989) at 224, we agree that dementia is more properly referred to as a "mental disorder" than a "disease." Id. at 604. We do not attempt to draw a technical distinction between the terms "disease" or "illness" and "disorder"; instead we use the term "illness" in this opinion in its nontechnical sense. Nor do we believe that this technical distinction is what the trial judge was referring to when he stated that Defendant did not suffer from a mental illness or disease.

[16] The dissent focuses on only one of dementia's features: memory loss. While memory loss is a central feature of dementia, it is not its only feature, and any implication to the contrary is incorrect. See Dr. Scialli's testimony ante § 3(c); see also DSM-III-R at 103. Although memory loss is relevant to this case, this does not mean we should overlook other features, such as impaired judgment and lack of impulse control, which are those most relevant to this case.

[11] Because the court reverses on an issue not raised by the defendant, the state has had no opportunity to brief the issue. If our independent review suggests an issue not raised by the parties we should solicit supplemental briefs before deciding it.