48 P.3d 1180
Ariz.
2002

Lead Opinion

OPINION WITH AMENDED DISSENT

FELDMAN, Justice.

¶ 1 On July 27, 1999, a jury found Doris Ann Carlson (Defendant) guilty of first-degree murder, conspiracy to commit firstrdegree murder, and first-degree burglary. She was sentenced to death on the murder charge, life imprisonment without the possibility of parole for twenty-five years on the conspiracy count, and an aggravated term of twenty-one years for the burglary. All sentences were concurrent. Because the trial judge sentenced Defendant to death for the murder, direct appeal to this court is automatic. A.R.S. § 13-703.01. We have jurisdiction pursuant to Arizona Constitution article VI, § 5(3), A.R.S. § 13-4031, and Rule 31.2(b), Arizona Rules of'Criminal Procedure.

FACTS AND PROCEDURAL HISTORY

¶ 2 In 1996, Defendant and her husband, codefendant David Carlson (David), were living in a house in Peoria, Arizona, that they shared with David’s mother, the victim in this case, Mary Lynne Carlson (Lynne). Defendant and David were financially dependent on Lynne. Lynne received about $850 each month from a trust fund valued at several hundred thousand dollars. In addition, she had two annuities, with a combined value of approximately $140,000. Lynne received roughly $800 per month from the first annuity and was allowed to draw on the principal from the second. David, as Lynne’s only child, was the beneficiary of the trust and both annuities.

¶ 3 When Defendant and David moved from Illinois to Arizona several years earlier, Lynne withdrew $70,000 from her second annuity and bought the Peoria house to accommodate all of them. Defendant and David depended on Lynne’s trust and annuities to pay their living expenses. Lynne had multiple sclerosis, was confined to a wheelchair, and had trouble controlling her bodily functions. Defendant was very impatient with Lynne, claimed she was only pretending to have multiple sclerosis, and yelled and cursed at her. Several times a week, Defendant would suggest that Lynne should be killed so that she and David could get Lynne’s money.

¶ 4 Because Lynne needed more care than David and Defendant could give her at home, she moved into a residential care facility in July 1996. The trust then stopped paying the utility bills and had them redirected to the home address. Lynne’s trust fund and annuity checks also stopped coming to the house, leaving Defendant and David broke.

¶ 5 In late September or early October 1996, Defendant approached their twenty-year-old boarder, John Daniel McReaken (Dan), and asked him if he knew anybody who wanted to make $20,000 by killing Lynne. Dan accepted Defendant’s offer. Another boarder, seventeen-year-old Scott Smith (Scott), offered to help Dan, and Dan agreed to give Scott half of the $20,000.

*575¶ 6 Defendant gave Dan money to buy gloves to use when killing Lynne. Dan and Scott already had the weapons, each having his own butterfly knife. Several days later, Defendant drove Dan and Scott to Lynne’s residential care facility because she wanted them to locate Lynne’s apartment and familiarize themselves with the area around it together with the different ways into and out of the facility.

¶ 7 On October 23, 1996, Defendant and David went to see Lynne and asked her to sign annuity documents in order to get money to help pay the mortgage so they would not lose the house. Lynne refused to sign without first consulting her financial advisor, which made Defendant angry. The next evening, October 24, Defendant told Dan that Lynne needed to be killed really soon, and Dan relayed the message to Scott. When Scott got home from work that evening, he and Dan dressed in black and got their gloves and knives. Defendant gave them a key to Lynne’s apartment and offered to drive them there.

¶8 Sometime after 1:00 a.m. on October 25, Defendant drove Dan and Scott to a supermarket near Lynne’s care facility and told them she would wait for them there. Once in Lynne’s apartment, Scott stayed in the living room, where he disconnected the television and moved the items from its top to make it appear there had been a burglary. Dan, meanwhile, went into the bedroom, and after hesitating, closed his eyes and stabbed Lynne eight to ten times. Dan later told Scott that he had stabbed Lynne in her throat and upper body and that she should die. When they returned to the car, Defendant asked whether they had done it, and Scott replied that they had. Defendant then drove them back to the house.

¶ 9 About 5:00 that same morning, a nursing assistant went to Lynne’s apartment to make her regular check. As she was unlocking the door, Lynne called out the assistant’s name and yelled at her for help, telling her she had fought “them” off as hard as she could. Lynne underwent several operations, but she never recovered from the knife attack and died on April 21,1997.

10 Defendant, David, Dan, and Scott were arrested on November 21, 1996, less than a month after the attack. Following Lynne’s death, they were charged with her murder. Based on the foregoing evidence, a jury found Defendant guilty of first-degree murder. As required by statute, the trial judge conducted a special sentencing hearing. A.R.S. § lS-UOSCB).1 Under the Arizona system, this hearing is conducted without a jury — the judge makes the factual findings that determine whether a defendant is to be sentenced to life imprisonment or death. Id. In Defendant’s aggravation/mitigation hearing, the judge determined that the state had proven three aggravating factors beyond a reasonable doubt: Defendant procured Lynne’s murder by promise of payment of something of pecuniary value, namely $20,000, A.R.S. § 13 — 703(F)(4); Lynne’s murder was committed in expectation of Defendant’s pecuniary gain, A.R.S. § 13-703(F)(5); and Lynne’s murder was committed in an especially heinous, cruel, or depraved manner, A.R.S. § 13-703(F)(6). The judge found one statutory mitigating circumstance, duress, A.R.S. § 13-703(G)(2); and two non-statutory mitigating circumstances — no prior criminal history and brain damage. The judge then determined that the mitigating evidence, when weighed against the aggravating evidence, was insufficient to call for leniency and sentenced Defendant to death.

DISCUSSION

A. Trial issue — jury selection

¶ 11 Defendant called no witnesses at trial. The only trial issue raised on appeal is whether the trial judge should have granted Defendant’s motion to strike all members of the jury because of their knowledge of — and discussion about — adverse publicity concerning Defendant’s trial counsel.

*576¶ 12 Defendant contends that the judge abused his discretion when he denied her motion to strike the entire jury panel after it became clear the panelists had been infected by “lurid” publicity concerning defense counsel’s public sexual conduct with a defendant she represented in a previous murder case.

1. July selection — background

¶ 13 Defendant was represented by Carmen Fischer (Fischer), who had been the subject of extensive publicity concerning her romantic and possible sexual relationship with a prior client in the high-profile “bounty hunter” murder case. There was widespread publicity in the media, up to and including the day before jury selection in the present case. Among other things, it was alleged both in newspapers and on television that Fischer had sexual contact with her client in a lawyer’s visiting room at the Maricopa County Jail. On television, Fischer could be seen kissing her client. This publicity, however, contained no reference to Defendant or her pending case.

¶ 14 At the beginning of jury selection, the judge asked if the state and Defendant were ready, and the lawyers, including Fischer, identified themselves to the court. At this point, before the panelists were sworn and before the lawyers were formally introduced to them, several of the panelists recognized Fischer, although none mentioned it at the time. Immediately following this, the panelists were sworn and there was a recess for lunch. During this recess, there was at least one conversation about Fischer between several of the venire members.

¶ 15 After the recess, in the course of voir dire, Fischer asked whether any of the panelists had seen anything in the newspapers or on television about anyone involved in the trial. Eleven of the panelists, including those involved in the lunchtime conversation, raised their hands. There was then another recess, and these eleven plus two others who later identified themselves were questioned in chambers about the extent of their exposure to the publicity surrounding Fischer and how it might affect them impartiality in this case. Seven of. the thirteen panelists indicated that what they had seen or heard of the allegations did not lead them to form a negative opinion of Defendant’s attorney. Three potential jurors indicated that reports of the relationship between Defendant’s attorney and one of her clients led them to form a negative opinion concerning the attorney, but they would be able to put aside their feelings and serve impartially. Three others indicated that their negative opinion of Defendant’s attorney either would not, or might not, allow them to serve impartially. These three were discharged for cause by the judge.

¶ 16 After the first nine prospective jurors had been questioned in chambers, Fischer reminded the judge that none of the panelists had been ordered to refrain from discussing this matter with other panel members. One man revealed in chambers that before voir dire commenced, he had learned about the controversy over lunch from some of the other venire persons who were talking about the television broadcast of Fischer kissing her client in the other murder case. The judge then ordered the man not to discuss this topic with anyone. Fischer then moved to strike all of the panel members who had been exposed to the media coverage. The judge refused to discharge any prospective juror other than the three he had already discharged. In the course of subsequent questioning in the courtroom, another panelist said that while the remainder of the panel was waiting in the jury room during in-chambers voir dire, he had overheard other prospective jurors talking about Fischer’s bad judgment. This reminded him that he had read of Fischer’s sexual relationship with a prisoner and that he thought she had used very poor judgment. Fischer again moved to strike all panel members due to their continuing discussion of this topic, but the judge denied the motion.

2. Constitutional basis

¶ 17 Defendant alleges that the Arizona Constitution affords greater protection than the federal constitution to the right to jury trial. She submits that if article II, §§23 and 24 are read together, they promise an “inviolate” right to a trial by an “impartial” jury. Because “inviolate” is not found in the Sixth Amendment to the United States Con*577stitution, the Arizona right is broader than the federal guarantee. She further contends that errors affecting juror impartiality cannot be shrugged off as harmless or waived and that courts should protect juror impartiality at every opportunity.

¶ 18 However, our courts have held that Arizona’s right to an impartial jury is no broader than the Sixth Amendment. State v. Wiley, 144 Ariz. 525, 536, 698 P.2d 1244, 1255 (1985) (“We ... will interpret defendant’s right to a trial by an impartial jury, Ariz. Const. Art. § 24, as co-extensive with the sixth amendment to the United States Constitution.”), overruled on other grounds by State v. Superior Court (Gardner), 157 Ariz. 541, 544, 760 P.2d 541, 544 (1988). Furthermore, when determining whether pretrial publicity resulted in a violation of a defendant’s right to an impartial jury, we have consistently cited federal cases. See, e.g., State v. Jones, 197 Ariz. 290, 307 ¶ 44, 4 P.3d 345, 362 ¶ 44 (2000) (citing Nebraska Press Ass’n v. Stuart, 427 U.S. 539, 554, 96 S.Ct. 2791, 2800, 49 L.Ed.2d 683 (1976) (courts rarely presume prejudice due to outrageous pretrial publicity)); State v. Stokley, 182 Ariz. 505, 514, 898 P.2d 454, 463 (1995). The facts of this case give us no reason to read the Arizona Constitution more broadly than the federal in determining issues of pretrial publicity.

3. Prejudice

¶ 19 Defendant argues that she was denied a fair trial because the jury panel was prejudiced by explosive media coverage of trial counsel’s alleged sexual conduct, buttressing her claim with copies of two newspaper articles about Fischer’s alleged escapades. The trial judge’s ruling on this issue is reviewable for abuse of discretion. State v. Bible, 175 Ariz. 549, 566, 858 P.2d 1152, 1169 (1993). Arizona has adopted a two-step inquiry to determine the effect of pretrial publicity: (1) did the publicity create a presumption of prejudice, or (2) has the defendant shown actual prejudice? State v. Murray, 184 Ariz. 9, 26, 906 P.2d 542, 559 (1995).

a. Presumed prejudice

¶20 For prejudice to be presumed, the defendant must show “that the publicity was so unfair, so prejudicial, and so pervasive that we cannot give any credibility to the jurors’ answers during voir dire affirming their ability to decide the case fairly.” Bible, 175 Ariz. at 565, 858 P.2d at 1168. This burden is extremely heavy and rarely met. Id. at 564, 858 P.2d at 1167 (citing Nebraska Press Ass’n, 427 U.S. at 554, 96 S.Ct. at 2800).

¶ 21 To establish a presumption of prejudice based on pretrial publicity, a defendant must establish that the publicity manipulated the jurors or otherwise distracted them from their duty to decide the case based solely on the evidence presented, making the trial little more than a “mockery of justice or a mere formality.” Stokley, 182 Ariz. at 513, 898 P.2d at 462 (quoting Bible, 175 Ariz. at 563, 858 P.2d at 1166; State v. Atwood, 171 Ariz. 576, 613, 832 P.2d 593, 648 (1992), disapproved on other grounds by State v. Nordstrom, 200 Ariz. 229, 241 ¶ 25, 25 P.3d 717, 729 ¶ 25 (2001)). In making this inquiry, the reviewing court examines the entire record, without regard to the panelists’ avowals of impartiality. Stokley, 182 Ariz. at 513, 898 P.2d at 462; Bible, 175 Ariz. at 565, 858 P.2d at 1168. This is because we must find that the publicity was so extreme and influential that no credibility can be given to the jurors’ claims of impartiality. Bible, 175 Ariz. at 565, 858 P.2d at 1168.

¶22 While some of the panelists appeared to be interested in Fischer’s so-called notorious relationship and the attendant publicity, there is no indication that after voir dire the jurors were distracted from their duty to decide the case based solely on the evidence presented. Although thirteen of the thirty-four prospective jurors on the panel had some knowledge of Fischer’s alleged behavior, the record shows only that four of the eventual twelve jurors had any knowledge of the allegations against her. This tends to refute Defendant’s contention that the pretrial publicity was so widespread that it infected the entire panel.

*578¶ 23 Of course, a trial panel containing any prejudiced juror would violate Defendant’s rights, but there is nothing in the record to suggest that because of pretrial publicity, the trial was little more than a mockery of justice so that we must disregard the jurors’ averments of impartiality. Rather, the jurors heard uncontested evidence that Defendant, who was in desperate need of money, offered Dan $20,000 to kill Lynne, provided money for Dan and Scott to buy gloves, transported Dan and Scott to Lynne’s apartment, and gave them a key to Lynne’s door. Thus, the verdicts are strongly supported by overwhelming and undisputed evidence presented at trial. There is nothing in the record to demonstrate that any of the twelve jurors failed in their duty to serve impartially.

¶ 24 Reviewing the record as a whole, we do not conclude that the trial was “utterly corrupted” by publicity such that prejudice must be presumed. Id. While admittedly the publicity was uncomfortably close to the commencement of Defendant’s trial, it must be remembered that it was about Defendant’s counsel, not Defendant, and did not concern Defendant’s case. Nothing indicates that the publicity prevented the jurors from considering only the evidence in the case before them. This case, for example, is markedly different from one of the leading cases on the subject, in which the jurors were

subjected to newspaper, radio and television coverage of the trial____ They were allowed to go their separate ways outside of the courtroom, without adequate directions not to read or listen to anything concerning the case____Moreover, the jurors were thrust into the role of celebrities by the judge’s failure to insulate them from reporters and photographers. The numerous pictures of the jurors, with their addresses, which appeared in the newspapers before and during the trial itself exposed them to expressions of opinion from both cranks and friends.

Sheppard v. Maxwell, 384 U.S. 333, 353, 86 S.Ct. 1507, 1517, 16 L.Ed.2d 600 (1966) (citation omitted); see also Rideau v. Louisiana, 373 U.S. 723, 726-27, 83 S.Ct. 1417, 1419-20, 10 L.Ed.2d 663 (1963) (inadmissible televised confession seen by many potential jurors). These cases demonstrate the kind of publicity that makes a mockery of justice and gives rise to a presumption of prejudice. Clearly, the record in this case fails to disclose outrageous circumstances that would justify such a presumption of prejudice.

b. Actual prejudice

¶ 25 Even if a defendant fails to prove that prejudice should be presumed, she can still obtain relief if she can prove actual prejudice-that the jurors had preconceived notions concerning her guilt that they could not disregard. Jones, 197 Ariz. at 307 ¶ 44, 4 P.3d at 362 ¶ 44.

¶ 26 Prior knowledge alone is not sufficient to prove actual prejudice in light of the fact that the four jurors with such knowledge all said, without qualification, that they were able to keep an open mind with regard to the case at hand and to be fair and impartial toward Defendant. See State v. Eastlack, 180 Ariz. 243, 253, 883 P.2d 999, 1009 (1994) (no finding of actual prejudice, even though ten jurors had prior knowledge of case, when all jurors expressed the ability to determine guilt from evidence presented at trial); Bible, 175 Ariz. at 566, 858 P.2d at 1169 (“[T]he relevant inquiry is the effect of publicity on a juror’s objectivity, not the mere fact of publicity.”); Atwood, 171 Ariz. at 632, 832 P.2d at 649 (no prejudice “where none of the jurors gave even the slightest indication during voir dire that prior knowledge of the case would impede their ability to serve as objective jurors.”).

¶ 27 To prove actual prejudice, Defendant “must show that the jurors have formed preconceived notions concerning [her] guilt and that they cannot lay those notions aside.” State v. Chaney, 141 Ariz. 295, 302, 686 P.2d 1265, 1272 (1984). The appropriate inquiry focuses on the effect of pretrial publicity — mere knowledge of the case is insufficient to disqualify a juror. Id. As noted, moreovei’, the pretrial publicity was only about Defendant’s counsel. The record does not show that any of the twelve jurors had any preconceived notion of Defendant’s guilt. There is nothing in the answers to voir dire to indicate that any of the jurors were actually biased against Defendant. *579While four2 of the twelve jurors who actually deliberated were aware of the allegations about Fischer, there is no indication that any of those four had even seen either of the newspaper articles of which Defendant complains.3 Furthermore, only one of the four jurors had formed a negative opinion of Fischer due to the publicity, and that juror assured the judge that he had no doubt he could decide the case on its merits.

¶ 28 The three panelists who admitted they could not set aside their negative image of Fischer and therefore could not be fair were struck. Because prospective jurors were not initially instructed to refrain from discussing Fischer’s sexual relationship, and at least one prospective juror learned of it through these discussions, it might well have been appropriate to have questioned and cautioned them further. However, this issue is not raised on appeal.4 Nor does the record show whether any panelist who may have heard lunch-time or jury room gossip about Fischer’s escapades actually served on the jury. Moreover, all twelve jurors indicated that they could decide the issues solely on the basis of the evidence presented at trial.

¶ 29 On balance, we conclude only that the judge did not abuse his discretion in denying the motion to strike the jury panel.

4. The dissent

¶ 30 Judge Voss’ dissent makes a strong argument. Our disagreement is narrow. The dissent argues that the trial judge should have granted the motion to strike the panel. We disagree for the reasons stated in the previous section. The dissent also argues that the judge should have granted Defendant’s request for further voir dire. Dissent at ¶ 79. Had there been a motion or even such a request, we would agree. But what happened is this: the prosecutor said he was “kind of thinking that maybe we ought to check” the prospective jurors to “see if they were going to answer yes, so we don’t have” to return to chambers. To these ruminations, Fischer’s cocounsel said, “I agree,” because he had “a feeling that some of the people from the original 36 may raise their hand.” Thus, he thought the “safe way to go” was to ask the question of the entire panel. The prosecutor replied that he guessed it did not “make any difference” if they were going to go back into the courtroom. Reporter’s Transcript, July 17, 1999, at 80-81.

¶ 31 It is impossible to find either a motion or even a request for additional voir dire in these Proustian dream sequences. While a motion would have been correct, we have no desire to be hypertechnical, and a request would have sufficed. Whatever the label, anything would have sufficed if it had raised a question on which the judge could have and should have ruled. We would then be able to say the judge was correct or erred in his ruling. But what was there to rule on here? At best, there was the prosecutor’s comment that maybe they ought to have more voir dire. There was no more voir dire on this issue,5 but the prosecutor is not the one complaining. As the dissent notes, additional voir dire would have been appropriate had Defendant moved or asked or requested the judge to undertake or allow it. If the judge had then declined, we would agree with the dissent. But Defendant did not ask and did not request a ruling, so we must disagree.

¶ 32 On appeal, Defendant is even less definitive. She complains the judge should have struck the entire jury panel or granted a mistrial because of the biased jury panel. Even setting aside the fact that the judge was never asked to grant a mistrial — no mo*580tion or request was made — it is impossible to say the judge abused his discretion by failing to grant the motion to strike the entire panel or in failing to mistry the case. There is no way to know whether any of the panelists in addition to those subjected to in-chambers voir dire. heard any further conversation about Fischer or, if they did, whether they sat on the jury. If they sat, we have no way of knowing what they had heard, so we would have to speculate on whether the juror or jurors were unable to set aside whatever information they might have heard and try the case fairly. Finally, when defense counsel was asked if she had made any record she intended to make, she twice stated that she had nothing further.

¶ 33 To reverse a case on the basis of such an inadequate record, we would have to speculate as to what would have been learned had counsel followed or come close to following the correct procedure and required the judge to rale. We are unwilling to speculate as a basis for reversal. Of course, any issue of ineffective assistance of counsel will abide post-conviction proceedings.

B. Sentencing issues

¶ 34 In every capital case, we must review the facts establishing the presence or absence of aggravating and mitigating circumstances to determine if the death penalty is appropriate. A.R.S. § 13-703.01(A); see also State v. Laird, 186 Ariz. 203, 208, 920 P.2d 769, 774 (1996).

1. Whether the aggravating factors were proved beyond a reasonable doubt

¶ 35 The trial judge found that the state had proven three aggravating factors beyond a reasonable doubt: Defendant procured Lynne’s murder by promise of payment of something of pecuniary value, namely $20,000, A.R.S. § 13-703(F)(4); Lynne’s murder was committed in expectation of pecuniary gain, A.R.S. § 13-703(F)(5); and Lynne’s murder was committed in an especially heinous, cruel, or depraved manner, A.R.S. § 13-703(F)(6).

a. Pecuniary gain factors

¶ 36 This case presents two related aggravating factors — procuring the commission of the offense by promise of payment — (F)(4); and committing the offense in the expectation of pecuniary gain — (F)(5).6

¶ 37 To prove (F)(4), the state must prove that Defendant got the actual killers to commit the murder by promising to pay them. This aggravator thus applies to a hired killer. See, e.g., State v. Bracy, 145 Ariz. 520, 537, 703 P.2d 464, 481 (1985). The state proved beyond a reasonable doubt that Defendant hired Dan and Scott to murder Lynne. To prove (F)(5), the state must prove beyond a reasonable doubt that pecuniary gain was “a motive, cause or impetus for the murder and not merely the result of the murder.” State v. Kayer, 194 Ariz. 423, 433 ¶ 32, 984 P.2d 31, 41 ¶ 32 (1999) (quoting State v. Spears, 184 Ariz. 277, 292, 908 P.2d 1062, 1077 (1996)). The finding may be based on “tangible evidence or strong circumstantial inference.” State v. Hyde, 186 Ariz. 252, 280, 921 P.2d 655, 683 (1996) (citing State v. Gillies, 135 Ariz. 500, 512, 662 P.2d 1007, 1019 (1983)). Defendant wanted Lynne killed precisely so she could benefit from Lynne’s trust fund and annuities. The evidence in this case thus establishes both the (F)(4) and (F)(5) factors.

¶ 38 Rarely have both (F)(4) and (F)(5) been found in the same case, but we did find both factors in State v. (Michael) Apelt, 176 Ariz. 349, 861 P.2d 634 (1993). It could be argued that there is only one plan or transaction that has given rise to both (F)(4) and (F)(5). The one transaction is very simple: Defendant wanted her mother-in-law murdered so she would get the proceeds of the trust fund and annuities; she would then share a small portion of this money with Dan and Scott, the people she *581hired to commit the murder. Because the two factors are so closely related, there is arguably an issue of whether the judge improperly double counted this one transaction. We have dealt with double counting before, but not with regard to the two types of pecuniary gain. In Apelt, the husband killed his wife to obtain life insurance proceeds and got his brother to help with her murder by promising him a share of those proceeds. However, we did not consider the single transaction problem in that case.

¶ 39 Even if there is but one transaction leading to the murder, a judge can properly use a single fact to support the application of more than one aggravating factor. State v. Bly, 127 Ariz. 370, 373, 621 P.2d 279, 282 (1980). It is not the separate counting per se that is a problem but the weight allocated. In State v. Scott, we stated:

The use of one fact to establish two aggravating circumstances is proper, provided the court, in balancing the aggravating and mitigating factors does not weigh the [fact] twice. Because it is but one fact, it cannot be weighed twice, even though it satisfied two separate aggravating factors.

177 Ariz. 131, 144, 865 P.2d 792, 805 (1993). In this case, use of the related factors to establish two aggravating circumstances is proper, provided the judge, in balancing the aggravating and mitigating factors, does not give full weight to both factors. In the present case, both pecuniary aggravators were properly found. They are separate and independent and thus can be counted twice. However, because they are so closely related, each factor should not be independently assigned full weight.

b. Especially heinous, cruel, or depraved

¶ 40 Under A.R.S. § 13-703(F)(6), commission of the offense “in an especially heinous, cruel or depraved manner” is an aggravating circumstance. To satisfy constitutional concerns, we narrowly construe these terms to apply only to “killingfs] wherein additional circumstances of the nature enumerated above set the crime apart from the usual or the norm.” State v. Knapp, 114 Ariz. 531, 543, 562 P.2d 704, 716 (1977).

¶ 41 Because the statute was written in the disjunctive (heinous, cruel or depraved), a sentencing judge need find only one of the factors to establish an (F)(6) aggravating factor. See State v. Gretzler, 135 Ariz. 42, 51, 659 P.2d 1, 10 (1983). Cruelty relates to the physical and mental suffering of the victim during the murder. State v. Clark, 126 Ariz. 428, 436, 616 P.2d 888, 896 (1980). Heinousness and depravity focus on the mental state of the defendant. Id.

¶ 42 The trial judge found that the crime was committed in an especially cruel, heinous, and depraved manner. He set forth the support for this finding in his special verdict:

The victim was a 53-year-old woman with multiple sclerosis and was, for all intents and purposes, bedridden. The victim awoke being stabbed and attempted to defend herself, suffering defensive wounds. The victim received at least eight stab wounds. She lingered over three hours alone, wounded and unable to call for help until a practical nurse came in and found her in the condition above described. The victim lingered for some six months before her death and underwent numerous surgical procedures____The victim was mutilated; the crime was senseless; the victim was helpless____

Special Verdict at 7. Defendant challenges these findings.

¶43 Undoubtedly Lynne suffered — both physically as well as mentally. In State v. Trostle, we said that “[c]ruelty exists if the victim consciously experienced physical or mental pain prior to death, and Defendant knew or should have known that suffering would occur.” 191 Ariz. 4, 18, 951 P.2d 869, 883 (1997) (citing State v. (Rudi) Apelt, 176 Ariz. 369, 376, 861 P.2d 654, 661 (1993); State v. Kiles, 175 Ariz. 358, 371, 857 P.2d 1212, 1225 (1993)). In the case before us, the trial judge noted that while “Defendant might not have foreseen that her co-conspirators would not complete the slaying of the victim in a timely manner, she is nonetheless by law responsible for the ensuing pain and suffering to the victim over a prolonged period of *582time.” Special Verdict at 7. Pointing out that she did not stab the victim, Defendant asserts she should not be held liable under agency principles for the unforeseeable ineptitude of her codefendants, which she did not foresee.

¶ 44 Foreseeability in connection with the cruelty factor has been based on an objective rather than subjective standard. We have held that the physical pain or mental anguish suffered by a victim before death must only be reasonably foreseeable, regardless of whether the defendant actually foresaw it. State v. Djerf 191 Ariz. 583, 595 ¶ 45, 959 P.2d 1274, 1286 ¶ 45 (1998); State v. Adamson, 136 Ariz. 250, 266, 665 P.2d 972, 988 (1983). The state argues that when Defendant hired a twenty-year-old unemployed drug addict and a seventeen-year-old fast food restaurant employee, both armed with only butterfly knives, it was reasonably foreseeable that they would bungle their assignment. The state also contends that given Lynne’s condition, it was reasonably foreseeable that if she were not instantaneously killed, she would be unable to call for help, suffering until someone came to check on her, and that any injuries she sustained could lead to further complications and prolonged suffering. In the sense that foreseeability is used in tort law, this is undoubtedly correct. But we do not believe the tort concept is relevant to determine eligibility for capital punishment. The death penalty is reserved for only those individuals whose crimes exceed the norm of first-degree murders and so set them apart as having acted in an especially cruel, heinous, and depraved manner. State v. Milke, 177 Ariz. 118, 126, 865 P.2d 779, 787 (1993). Thus, concepts developed in the law of torts or law of agency are inapposite.

¶ 45 As we have repeatedly held, the death penalty should not be imposed in every capital murder case but, rather, it should be reserved for cases in which either the manner of the commission of the offense or the background of the defendant places the crime “above the norm of first-degree murders.” State v. Hoskins, 199 Ariz. 127, 163 ¶ 169, 14 P.3d 997, 1033 ¶ 169 (2000) (quoting State v. Blazak, 131 Ariz. 598, 604, 643 P.2d 694, 700 (1982)); State v. Zaragoza, 135 Ariz. 63, 68-69, 659 P.2d 22, 27-28 (1983) (“either the circumstances of the killing are so shocking ... or the background of the murderer sets him apart from the usual first degree murderer.”); see also State v. Smith, 146 Ariz. 491, 505, 707 P.2d 289, 303 (1985). Thus, we have held that to pass constitutional muster, sentencing schemes must narrow the class of persons to those for whom the sentence is justified. Jones, 197 Ariz. at 309 ¶ 52, 4 P.3d at 364 ¶ 52; see also Zant v. Stephens, 462 U.S. 862, 877, 103 S.Ct. 2733, 2742-43, 77 L.Ed.2d 235 (1983). As a result, the “specified statutory aggravators in Arizona’s death penalty scheme are designed to narrow, in a constitutional manner, the class of first degree murderers who are death-eligible.” State v. Soto-Fong, 187 Ariz. 186, 202, 928 P.2d 610, 626 (1996).

¶ 46 Our court of appeals put it well in saying that “reasonable foreseeability of death is an insufficiently stringent measure of culpable state of mind to justify capital punishment as a consequence of accomplice liability.” State v. Marchesano, 162 Ariz. 308, 315, 783 P.2d 247, 254 (App.1989), disapproved on other grounds by State v. Phillips, 202 Ariz. 427, 436 n. 4, 46 P.3d 1048, 1057 n. 4 (2002).7 We agree. As the court understood when it discussed the culpable state of mind, accomplice guilt requires only that a defendant have “the intent to promote or facilitate the commission of an offense.” A.R.S. § 13-301. Mere foreseeability as a benchmark for death in capital cases would not permit the aggravators to serve their constitutional purpose of narrowing the class of first-degree murderers who can be sentenced to death.

*583¶ 47 If, rather than a tort concept of foreseeability, we apply the criminal law concept of mens rea to this case, we see that Defendant was not present during commission of the crime, did not supply the murder weapon, and was not involved in planning the details or method of murder. While she is certainly an accomplice to the murder and equally guilty along with the actual killer, there is nothing in the record to indicate she intended that Lynne should suffer during the attack itself or that any other act qualifying for the aggravator of cruelty would occur. In capital cases involving accomplices, a better test than mere foreseeability of suffering is a finding that the defendant intended that the murder be committed in such a manner as to cause the victim to suffer or, absent intent, knew it would be so.

¶ 48 This reasoning is consistent with prior decisions. We have held that the (F)(6) aggravator of cruelty existed because a defendant knew or should have foreseen that the victim would suffer. State v. Dickens, 187 Ariz. 1, 24-25, 926 P.2d 468, 491-92 (1996). As in the instant case, the defendant in Dickens planned the murder and provided transportation for the actual murderer. However, in Dickens, the defendant provided a gun to a person he knew to be violent, selected the two robbery victims, issued instructions to leave no witnesses, knowing that meant one victim would be alive to watch the execution-style slaying of the other, and was actually present at the scene during commission of the crime. The defendant in Dickens had to know that the killing would be cruel and thus had the requisite intent and culpable state of mind discussed above. In the case before us, however, while Defendant asked Dan to kill Lynne, she did not plan how the murder would be committed and could not have known that Dan would bungle it by closing his eyes while he repeatedly stabbed Lynne.

¶ 49 This is the first case in which we have been called on to define the boundaries of foreseeability when the defendant was neither the actual killer nor a witness to the murder. We believe the tort theory of culpability advanced by the state is too broad for practical application. There is no vicarious liability for cruelty in capital cases absent a plan intended or reasonably certain to cause suffering. The plan must be such that suffering before death must be inherently and reasonably' certain to occur, not just an untoward event. Id. Defendant is not responsible for the bungling of her hired killers. Picking these two inexperienced murderers did not make it reasonably certain that things would go dreadfully wrong.

¶50 The culpable state of mind requirement we define today is further supported by two earlier decisions. In Adamson, we held that the means the defendant chose and used to kill, an explosive device, made the victim’s suffering reasonably foreseeable as a direct consequence of that chosen method. 136 Ariz. at 266, 665 P.2d at 988. In the present case, the record does not indicate that Defendant knew the method Dan and Scott had planned was reasonably certain to cause Lynne to suffer. Lynne’s suffering was not a part of Defendant’s plan. In State v. Walton, we held that the victim’s survival after a shot to the head and his subsequent floundering in the desert for a week were neither reasonably foreseeable nor intended. It must be the intent preceding the mortal blow, not “what surprisingly transpired” afterward, that “guides our analysis.” 159 Ariz. 571, 587, 769 P.2d 1017, 1033 (1989). As in Walton, Lynne’s suffering was not intended or foreseeable to the extent necessary to charge Defendant with a culpable state of mind springing from the acts of her hired killers. Thus, we conclude cruelty was not properly found in this ease. It is therefore necessary to turn to the heinous and depraved findings of the trial judge to affirm the (F)(6) aggravator.

¶ 51 The heinous and depraved portion of the (F)(6) aggravator focuses on the defendant’s state of mind at the time of the crime. Gretzler, 135 Ariz. at 51, 659 P.2d at 10. However, the inquiry concentrates on the defendant’s mental state as evidenced through her actions. State v. Rienhardt, 190 Ariz. 579, 590, 951 P.2d 454, 465 (1997). The factors used to establish a heinous and depraved state of mind are (1) relishing the killing, (2) commission of gratuitous violence, (3) mutilation of the victim, (4) senselessness of the killing, and (5) helplessness of the *584victim. Gretzler, 135 Ariz. at 52-53, 659 P.2d at 11-12. Because “[a]ll first degree murders are to some extent heinous ... or depraved[,] ... to warrant the imposition of the death penalty, a murder must be more heinous ... or depraved than usual.” Smith, 146 Ariz. at 503, 707 P.2d at 301.

¶52 The trial judge found the Gretzler factors of mutilation, senselessness, and helplessness present.8 As the state concedes, the findings of mutilation and senselessness are not supported by the facts. Mutilation requires a finding of a separate purpose to mutilate. State v. Medina, 193 Ariz. 504, 514 ¶ 38, 975 P.2d 94, 104 ¶ 38 (1999). This record does not establish such an intent. Similarly, it does not show senselessness — the killing was central to the criminal objective of inheriting Lynne’s money so Defendant and David could resolve their dire financial situation. A murder is senseless only if it is unrelated to the defendant’s goal. State v. West, 176 Ariz. 432, 448, 862 P.2d 192, 208 (1993), overruled on other grounds by State v. Rodriguez, 192 Ariz. 58, 961 P.2d 1006 (1998).

¶ 53 Due to her condition, however, Lynne was clearly helpless. Defendant argues that a finding of helplessness alone is not sufficient to establish the (F)(6) factor. Even helplessness combined with senselessness would be unlikely to support a finding of heinousness and depravity without an additional factor. State v. Schackart, 190 Ariz. 238, 250, 947 P.2d 315, 327 (1997) (“Senselessness and helplessness, without more, are ordinarily insufficient to prove heinousness or depravity.”). This is because the Gretzler factors of senselessness and helplessness are “less probative of the defendant’s state of mind than are relishing, gratuitous violence, and mutilation.” Hyde, 186 Ariz. at 281, 921 P.2d at 684. The additional factor here was the familial relationship Defendant had with the victim.

¶ 54 We have held it permissible to use the parent-child relationship in partial support of the heinousness and depravity finding. Milke, 177 Ariz. at 126, 865 P.2d at 787 (“We hold that the use of the parent/child relationship ... is permissible and is within the Gretzler-Knapp parameters.”) (citations omitted) (emphasis added). But both Milke and Knapp dealt with a parent killing his or her young child. Milke, 177 Ariz. at 126, 865 P.2d at 787; Knapp, 114 Ariz. at 543, 562 P.2d at 716. So, also, did State v. Stanley, 167 Ariz. 519, 529, 809 P.2d 944, 954 (1991); see also State v. Wallace, 151 Ariz. 362, 369, 728 P.2d 232, 239 (1986) (older children killed by mother’s domestic partner). All these cases, except Milke, also contained the primary components of heinous and depraved conduct — relishing or gratuitous violence— something the current case does not present.

¶55 In this case, dealing with a woman and her mother-in-law, we believe it unwise to expand the concept of relationship as an aggravating factor. It is unfortunately true that a substantial number of first-degree murders occur between domestic partners and family members.9 We must again bear in mind that because all first-degree murders are heinous and depraved, these aggravating factors are applied only to those killings that are especially heinous or depraved and “can be described without reservation as ‘hatefully or shockingly evil’ and ‘marked by debasement, corruption, perversion or deterioration.’ ” Milke, 177 Ariz. at 126, 865 P.2d at 787 (quoting Gretzler, 135 Ariz. at 51, 659 P.2d at 10). In Milke, we were careful to limit our language: “The parent/child relationship is a circumstance that separates [infanticide] from the ‘norm’ of first-degree murders. The use of that relationship in partial support of a finding of heinousness and depravity ... is constitutionally permissible.” Id. (emphasis added). While there might be other relationships in which appli*585cation of Milke’s principle could be appropriate, we must caution against ad hoc expansions of the Gretzler factors. The United States Supreme Court previously held that the Gretzler factors provide a “constitutionally sufficient” channeling of the court’s discretionary authority. See Walton v. Arizona, 497 U.S. 639, 652-56, 110 S.Ct. 3047, 3056-58, 111 L.Ed.2d 511 (1990), overruled, on other grounds by Ring v. Arizona, 2002 WL 1357257 (U.S. June 24, 2002). Continual ease-by-case expansion of these factors would lead to serious constitutional problems in view of the constitutional mandate to avoid arbitrary imposition of the death penalty. The legislature, on the other hand, may enact and define reasonable and narrowing aggravating circumstances that apply, across the board, to all cases.

¶ 56 Thus, we conclude the crime was neither heinous nor depraved in the constitutional sense and cruelty was not chargeable to Defendant. We therefore vacate the (F)(6) finding. The only aggravating circumstances present are (F)(4) and (F)(5).

2. Whether mitigating evidence was given sufficient weight

a. Statutory mitigation

¶ 57 Defendant must prove the mitigating factors listed in A.R.S. § 13-703 by a preponderance of the evidence. See Laird, 186 Ariz. at 207-08, 920 P.2d at 773-74. The trial judge found one statutory mitigating circumstance, that of duress, A.R.S. § 13-703(G)(2). The judge found that Defendant was under unusual and substantial duress although not such as to constitute a defense to prosecution. A.R.S. § 13-703(G)(2). He also found that Defendant’s money, house, and child custody worries created stress that rose to the level of a mitigating circumstance.

¶ 58 Prior to sentencing, Defendant maintained that A.R.S. § 13-703(G)(1) was applicable (“Defendant’s capacity to appreciate the wrongfulness of his conduct or to conform his conduct to the requirements of law was significantly impaired, but not so impaired as to constitute a defense to prosecution.”). The trial judge found that Defendant’s capacity to appreciate the wrongfulness of her conduct or to conform her conduct to the requirements of the law was not significantly impaired.

¶ 59 To establish this mitigator, a defendant must prove the existence of an identifiable mental disease or psychological defect that results in significant impairment. Hoskins, 199 Ariz. at 148 ¶ 93, 14 P.3d at 1018 ¶ 93. Moreover, proof of a causal connection between the impairment and the criminal act is essential. State v. Martinez, 196 Ariz. 451, 464 ¶ 57, 999 P.2d 795, 808 ¶ 57 (2000). Although the trial judge found Defendant suffered from brain damage, Defendant did not prove her brain damage resulted in significant impairment that led her to commit the crime. While the brain damage resulted in impulsivity and poor judgment, this does not equate to inability to conform one’s conduct to the law. Hoskins, 199 Ariz. at 148 ¶ 93, 14 P.3d at 1018 ¶ 93. The facts support the judge’s finding: Defendant’s expert conceded that Defendant did not suffer from severe brain damage, that five to ten percent of the general population suffers from her type of minor brain damage, that a lot of these people function well in society, and that his testing did not rule out the possibility of Defendant being the mastermind in the plan to murder Lynne. This was actually the testimony most favorable to Defendant because the state’s expert testified that no medical record indicated Defendant suffered from any type of brain damage, that neither testing by Defendant’s expert nor letters written by Defendant suggested a diagnosis of brain damage, and that there was a real possibility Defendant was malingering as to brain damage issues. Thus, even assuming Defendant was able to prove brain damage by a preponderance of the evidence, she has failed to prove that she could not appreciate the wrongfulness of her conduct or that her ability to conform her conduct to the requirements of law was significantly impaired.

¶ 60 Finally, the trial judge found Defendant was legally accountable for the conduct of another and her participation was major. See A.R.S. § 13-703(G)(3). Defendant planned the murder, hired the killers, gave them money to buy gloves, provided *586them with a key to Lynne’s apartment, drove them to a place near Lynne’s apartment, and awaited their return. The trial judge deemed this statutory mitigator did not exist. We agree.

b. Non-statutory mitigation

¶ 61 When a defendant is being sentenced for first-degree murder, the sentencing judge must consider, in addition to the mitigating circumstances enumerated in A.R.S. § 13-703(G), any aspect of a defendant’s character or record and any circumstance of the offense relevant to determining whether a sentence less than death might be appropriate. State v. McCall, 139 Ariz. 147, 162, 677 P.2d 920, 935 (1983). The defendant must prove the existence of non-statutory mitigating circumstances by a preponderance of the evidence. State v. McMurtrey, 143 Ariz. 71, 73, 691 P.2d 1099, 1101 (1984). We independently review the judge’s findings concerning non-statutory mitigation factors. In this case, the judge found two non-statutory mitigating circumstances, absence of prior criminal history and brain damage.

¶ 62 A lack of prior felony convictions may constitute a non-statutory mitigating factor. State v. Doerr, 193 Ariz. 56, 70, 969 P.2d 1168, 1182 (1998). In its presentence memorandum, the state conceded that Defendant had no prior criminal convictions. The judge therefore correctly found this mitigator to exist.

¶ 63 The judge then found that Defendant had established that she suffered from some brain damage as a result of difficulties at birth. While this impairment did not rise to the level of a statutory mitigator, it could be considered in mitigation to the extent it may have affected Defendant’s conduct. State v. Clabourne, 194 Ariz. 379, 387 ¶ 31, 983 P.2d 748, 756 ¶ 31 (1999); see also McMurtrey, 136 Ariz. at 102, 664 P.2d at 646. But brain damage seems not to have affected Defendant’s ability to manage the entire household and its financial affairs. Defendant argued that brain damage prevented her from coming up with the scheme to murder Lynne. However, the judge found that Defendant was the person who provided the motive and method. Defendant also argued that the brain damage could cause her to act impulsively and irresponsibly. The trial judge agreed to some extent, finding brain damage as non-statutory mitigation, but he gave it little weight. We agree with this conclusion.

¶ 64 Defendant next argues that the disparity between her sentence and those imposed on her codefendants should amount to a mitigating circumstance. She was the only one to be sentenced to death, but the judge reasoned that the disparity was justified under the circumstances and had no mitigating effect. The state argued that there were explanations for David, Dan, and Scott’s sentences. David was charged with the same crimes as Defendant but was only convicted of conspiracy to commit first-degree murder; he was sentenced in accordance with the law applicable to this offense: life imprisonment without the possibility of parole for twenty-five years. Dan was convicted of the same charges as Defendant, but the judge received letters requesting leniency from members of the jury that convicted him. In addition, without promise of any benefit, Dan testified against Defendant, thus possibly waiving his appeal rights and exposing himself to retaliation by fellow prisoners. Dan was sentenced to natural life in prison. Scott, who was only the lookout, entered a guilty plea to second-degree murder in return for testifying on behalf of the state.

¶ 65 Disparity between the sentences received by a defendant and an accomplice may be a mitigating factor. Stokley, 182 Ariz. at 523, 898 P.2d at 472. However, disparity in sentences is mitigating only when not adequately explained. State v. Schurz, 176 Ariz. 46, 57, 859 P.2d 156, 167 (1993). Simply because an accomplice has received leniency does not, in itself, prevent imposition of the death penalty. State v. Marlow, 163 Ariz. 65, 72, 786 P.2d 395, 402 (1989). In Marlow, the court said it appreciated the difficult tactical choices that must sometimes be made by the prosecution in obtaining a conviction. Id. In Scott’s case, he was offered a plea in return for testifying on behalf of the state. Difference in sentences resulting from appropriate plea bargaining is not necessarily *587mitigating. State v. Gillies, 142 Ariz. 564, 571, 691 P.2d 655, 662 (1984). Furthermore, Scott’s participation as lookout was relatively minimal compared to that of Defendant, who masterminded the plot.

¶ 66 It was Dan, however, who actually stabbed Lynne, and it was David who stood to directly benefit from Lynne’s death and had the closest familial relationship to her. Neither Dan nor David received a death sentence. We dealt with a similar disparity problem in State v. Mann, in which the disparity was not considered mitigating because the person treated lightly was only an accomplice and Mann was the actual killer. 188 Ariz. 220, 230, 934 P.2d 784, 794 (1997). In the present case, the actual killer got the life sentence, and Defendant, who was not the actual killer, was sentenced to death — the opposite of what justified the disparate treatment in Mann. See also State v. Clabourne, 142 Ariz. 335 at 348, 690 P.2d 54 at 68 (1984) (holding there was no mitigation where defendant, but not accomplices, did actual killing, which involved strangling and stabbing). The disparity in the present case must be given some weight.

¶ 67 In her presentence memorandum, Defendant also maintained that she was motivated to murder Lynne because she wanted to use the money that her husband would receive to obtain custody of her three minor children. Assuming, without deciding, this could have some mitigating weight, Defendant did not provide evidence that she planned and organized Lynne’s murder to gain custody of her children. We agree with the trial judge that no mitigating weight should be given to this argument.

3. Independent reweighing

¶ 68 When the trial judge errs in findings on aggravation or mitigation, remand for resentencing is generally inappropriate unless the judge wrongly excluded evidence or the record does not adequately reflect all of the relevant facts. A.R.S. § 13-703.01(C). Neither situation is present here. Therefore, we must independently determine the weight to be accorded each circumstance in determining whether to impose the death penalty. State v. Lavers, 168 Ariz. 376, 391, 814 P.2d 333, 348 (1991). We must then independently reweigh the aggravating and mitigating circumstances in deciding whether the proven mitigation is sufficiently substantial to warrant leniency. A.R.S. § 13-703.01(B); Murray, 184 Ariz. at 36, 906 P.2d at 569. In carrying out this duty, we have stated that the “question before us is not whether the trial court properly imposed the death penalty, but whether, based upon the record before us, we believe that the death penalty should be imposed.” State v. Watson, 129 Ariz. 60, 63, 628 P.2d 943, 946 (1981).

¶ 69 Our task in evaluating and weighing the mitigation is difficult at best. There is no scale upon which to measure what is or is not sufficiently substantial. In deciding whether death is an appropriate sentence, however, we must make an individualized sentencing determination as required by both the Arizona and United States Constitutions. In carrying out this responsibility, “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.” State v. Stuard, 176 Ariz. 589, 605, 863 P.2d 881, 897 (1993). We are instructed as follows by A.R.S. § 13-703.01(B):

If the supreme court determines that an error was made regarding a finding of aggravation or mitigation, the supreme court shall independently determine if the mitigation the supreme court finds is sufficiently substantial to warrant leniency in light of the existing aggravation. If the supreme court finds that the mitigation is not sufficiently substantial to warrant leniency, the supreme court shall affirm the death sentence. If the supreme court finds that the mitigation is sufficiently substantial to warrant leniency, the supreme court shall impose a life sentence pursuant to § 13-703, subsection A.

¶ 70 The trial judge found that three aggravating circumstances had been proven, as well as one statutory and two non-statutory mitigating circumstances. We have concluded that the (F)(6) aggravating factor is not supported. We have approved both pecuniary gain factors, (F)(4) and (F)(5), which are counted separately but in this case cannot be *588considered as two full and separate factors. In mitigation, the judge properly found that Defendant proved the finding of duress, (G)(2). As for non-statutory mitigators, we agree with the judge that Defendant proved her brain damage and her lack of criminal convictions. In addition, we conclude that to some extent, the disparity of sentences received by Defendant’s accomplices also must be given mitigating weight. Taken as a whole, this evidence raises a substantial question whether death is appropriate. When “there is a doubt whether the death sentence should be imposed, we will resolve that doubt in favor of a life sentence.” State v. Valencia, 132 Ariz. 248, 250, 645 P.2d 239, 241 (1982). Due to the nature of Defendant’s crime, however, we believe she should be imprisoned for the rest of her natural life and never be released.

DISPOSITION

¶ 71 Defendant’s convictions and non-capital sentences are affirmed. The sentence of death imposed on count three is reduced to imprisonment for Defendant’s natural life without the possibility of parole. See A.R.S. § 13-703.01(B). This disposition moots several issues concerning the constitutionality of the Arizona death penalty system.

CONCURRING: CHARLES E. JONES, Chief Justice, RUTH V. McGREGOR, Vice Chief Justice, and THOMAS A. ZLAKET, Justice (retired).

Notes

. The 2001 amendment to A.R.S. § 13-703 inserted a new subsection B, thereby redesignating existing subsections B to H as C to I, accordingly. At the sentencing hearing, because the aggravating and mitigating circumstances were known as A.R.S. § 13-703(F) and (G) respectively, these are the designations we will use here.

. The record does not indicate whether any of the eight other jurors ever became aware of Fischer’s alleged conduct.

. These articles from The Arizona Republic (July 5 and July 14, 1999) described the intimacy between Fischer and her client and the conflict-of-interest charges the State Bar of Arizona filed against Fischer.

. Defendant’s statement of the issue is as follows: "The court should have struck the entire jury panel, or at least those jurors who had discussed lurid pretrial publicity of the court-appointed trial counsel’s public sexual conduct with a client in a different death penalty case. The prejudice has been proved or ought to be presumed."

. Two replacement panelists were asked only whether they had any knowledge about any of the participants in this case.

. A.R.S. § 13-703(F)(4) provides where "[t]he defendant procured the commission of the offense by payment, or promise of payment, of anything of pecuniary value,” the judge shall consider this an aggravating circumstance. A.R.S. § 13-703(F)(5) provides where "[t]he defendant committed the offense as consideration for the receipt, or in expectation of the receipt, of anything of pecuniary gain,” the judge shall consider this an aggravating circumstance.

. Marchesa.no cited Tison v. Arizona, in which the United States Supreme Court discussed accomplice liability for murder. That court held that a defendant may be sentenced to death for a murder done by an accomplice only if the defendant had substantial involvement in its commission and showed a reckless indifference to human life. 481 U.S. 137, 157, 107 S.Ct 1676, 1688, 95 L.Ed.2d 127 (1987). Tison, however, dealt with eligibility for the death penalty rather than application of aggravating factors, as in the present case.

. There is no evidence of the two remaining Gretzler factors: relishing of the murder by Defendant and infliction of gratuitous violence on Lynne.

. For example, family members and sexual partners (including cohabitants) make up nearly 17% of first-degree murderers indicted in four Arizona counties (Maricopa, Pima, Coconino, and Mohave). Summary of First-degree Murder Cases in Four Arizona Counties, 1995-1999: Data Set II Research Report to Arizona Capital Case Commission, submitted by the Research Subcommittee of the Arizona Capital Case Commission, January 2002.



Dissenting Opinion

VOSS, Judge1,

dissenting.

¶ 72 I respectfully dissent. Because I believe the trial court abused its discretion in failing to strike the entire jury panel or, at a minimum, in failing to conduct further voir dire to determine whether the panel had been infected by negative and lurid gossip concerning defense attorney Carmen Fischer’s conduct, I would reverse and remand for a new trial.

¶ 73 As the Majority acknowledges, television and newspaper accounts — some the very day before jury selection in Defendant’s trial — reported that Fischer was caught having sexual contact with her “Bounty Hunter” client in a visiting room of the Maricopa County Jail. As a result of these stories, several prospective jurors recognized Fischer from the moment she entered the courtroom for jury selection. In fact, eleven of these, nearly one-third of the thirty-four panel members, raised their hands when Fischer asked if any of them “in reading the paper or watching the news” had seen any stories about any of the attorneys.

¶ 74 When these eleven prospective jury members were individually questioned in chambers as to what they had, heard, read, or seen about Fischer, three of them (May, Thomas, and Hunt) were so offended by Fischer’s behavior, that they felt they could not be fair and impartial to her client in this case. To make matters worse, after their individual voir dire in chambers concerning Fischer, nine out of the eleven panelists, including May, Thomas, and Hunt, were sent back into the jury room without any admonition to refrain from discussing Fischer’s behavior with the other prospective jurors.

¶ 75 And discuss it they did. Within a short time, two additional prospective jurors informed the court that they had heard gossip about Fischer from other members of the panel, both during lunch and in the jury room. Prospective juror Williams was brought into chambers directly after the first eleven had been questioned and sent back to the jury room. Prospective juror Divakaruni told the court he had something to discuss privately after all of the parties had returned to the courtroom and was then questioned in chambers.

¶76 During lunch, venireman Williams heard from other prospective jurors that Fischer “had been seen on television kissing one of [her] clients ... a murder suspect or something like that.” After Williams described his lunchtime discussion with other prospective jurors, the prosecutor suggested that the next few potential jurors be called into chambers to determine whether they had been exposed to stories about Fischer. *589Defense attorney Noland agreed, but in addition requested that the entire panel be asked again what they had seen and heard about Fischer in light of the fact that the jurors had been talking to each other about it since the initial question was asked. As the record reveals, both the prosecutor and the trial judge appeared to acquiesce in Noland’s request:

MR. LYNCH: Your Honor, I was kind of thinking that maybe we ought to check the next three or four or five and see if they are going to answer yes, so we don’t have to, if they are, then we don’t have to come back in here, all of us.
MR. NOLAND: I agree. I think even after we have had this time to mull this over, if we were to ask the general question again, I have a feeling that some of the people from the original 36 may raise their hand.
I think they may have either through talking to other people in there or it just may jog their memory. I think the safe way to go is to ask the question of the entire panel, someone that has not yet raised their hand, including the people that may not be called as potential jurors.
MR. LYNCH: Well, if we are going to go back in there, I guess it doesn’t make any difference.
THE COURT: Let’s go back in there.

¶77 Before the parties returned to the courtroom, Fischer moved to strike all prospective jurors who had seen the negative news coverage of her. The court granted the motion only as to prospective jurors May, Thomas, and Hunt, and again stated, “[ljet’s go back in there.” The judge never specifically ruled on Noland’s request to conduct additional voir dire of the entire panel as to what they had seen or heard about Fischer.

¶78 Venireman Divakaruni related that overhearing three other jurors in the jury room discussing Fischer’s “bad judgment” in carrying on “some kind of personal relationship with a prisoner” reminded him that he had read about Fischer in the newspaper. After Divakaruni informed the court and the parties of the gossip concerning Fischer that had taken place in the jury room, Fischer moved to strike the entire panel, stating, “I would move to strike the panel because of this conversation here while we were back in here, and to — and I guess there is some confusion about whether these other jurors we talked to went back and talked to other people also.” Again, the judge never specifically ruled on Fischer’s motion to strike the entire panel. Instead, he asked Fischer if “subject to the record previously made,” she passed the panel for cause. Fischer stated that she did.

¶ 79 I believe that the trial court’s failure to strike the entire panel, or at least conduct further voir dire as requested by defense counsel in this case, violated Defendant’s right to trial by a fair and impartial jury. Both the United States and Arizona Constitutions guarantee a criminal defendant the right to a trial by an impartial jury. U.S. Const. Amends. VI, XIV; Ariz. Const. Art. 2, §§ 23-24. Indeed, the right to an impartial jury is fundamental and deeply embedded in American jurisprudence. In Murphy v. Florida, 421 U.S. 794, 799, 95 S.Ct. 2031, 44 L.Ed.2d 589 (1975), the Supreme Court stated that, “[t]he constitutional standard of fairness requires that a defendant have ‘a panel of impartial, ‘indifferent’ jurors.’ ” (quoting Irvin v. Dowd, 366 U.S. 717, 722, 81 S.Ct. 1639, 6 L.Ed.2d 751 (1961)).

¶80 The guarantee of an impartial jury necessarily rests upon an adequate voir dire to identify unqualified jurors. Morgan v. Illinois, 504 U.S. 719, 729, 112 S.Ct. 2222, 119 L.Ed.2d 492 (1992); Rosales-Lopez v. United States, 451 U.S. 182, 188, 101 S.Ct. 1629, 68 L.Ed.2d 22 (1981). While admittedly the adequacy of voir dire is not easily the subject of appellate review, this court should not hesitate, particularly in a capital case, to find that certain questions simply have to be asked by the trial court to vindicate the accused’s right to trial by an impartial jury. See Morgan, 504 U.S. at 730, 112 S.Ct. 2222.

¶ 81 Rule 18.5(d) of the Arizona Rules of Criminal Procedure requires the trial court to “conduct a thorough oral examination of the prospective jurors.” Moreover, [ujpon the request of any party, the court shall permit that party a reasonable time to conduct a further oral examination of the pro*590speetive jurors.” Id. The purpose of such examination by both the trial court and the parties is to provide the parties with the information needed to exercise both peremptory challenges and challenges for cause. Ariz. R.Crim. P. 18.5(e); State v. Chaney, 141 Ariz. 295, 304, 686 P.2d 1265, 1274 (1984). Indeed, the trial court does not have the discretion to deny a defense counsel’s request to conduct additional voir dire to discover information relevant to such challenges under Rule 18.5. State v. Anderson, 197 Ariz. 314, 320-21, ¶ 14, 4 P.3d 369, 375-76 (2000).

¶ 82 The required “thorough oral examination” of the prospective jurors did not occur in this case. The Majority acknowledges this fact, but dismisses it as an issue that was not raised on appeal. Ironically, the Majority at the same time finds it “impossible” to conclude that the trial judge abused his discretion in failing to strike the entire jury panel because “there is no way to know” how, or even if, sitting jurors, in addition to those subjected to in-chambers voir dire, were affected by any further gossip about Fischer’s conduct.

¶ 83 But, that is precisely the point. The issue of whether the trial court should have stricken some or all of the jury panel, and hence the issue of whether we can uphold the trial court’s decision on appeal, necessarily includes the question of whether the trial judge should have conducted additional voir dire. For it is the inadequate voir dire in this case that effectively prevented the parties, the trial court, and now this court on appeal, from determining whether other potential jurors should have been excused for cause, or whether, as defendant argues on appeal, the whole panel was tainted.

¶ 84 I believe that the issue of additional voir dire was clearly preserved in the trial court. While defense attorney Noland’s request for additional voir dire was not a model one, it certainly was enough to preserve this issue. Although the Majority professes “no desire to be hypertechnical,” it releases No-land’s statements to “ruminations” and “Proustian dream sequences.” I cannot agree.

¶ 85 Noland’s words must be taken in context and in light of the circumstances existing when he uttered them. The record clearly indicates Noland was seeking additional voir dire; he did so directly after it became known that a panel member in addition to those who initially raised their hands had heard about Fischer from other prospective jurors; he advised the court what action he desired the court to take and made known the precise area of inquiry he wished to pursue; and he did all of this before any challenges for cause were made or ruled upon. This request was real.

¶ 86 The question of whether additional voir dire should have been conducted is also of necessity incorporated in the issues raised on appeal. In her Opening Brief, Defendant framed the issue as follows: “The court should have struck the entire jury panel, or at least those jurors who had discussed lurid pretrial publicity of the court-appointed trial counsel’s public sexual conduct with a client in a different death penalty case.” The Brief goes on to argue that Defendant’s right to a fair and impartial jury panel was violated. The claim that a defendant was tried by a biased and tainted jury necessarily implicates the issue of whether the trial court should have conducted further voir dire to determine whether the panel had in fact been infected. See Mach v. Stewart, 137 F.3d 630, 632-33 (9th Cir. 1998).

¶ 87 According to the Majority, Defendant has failed to demonstrate any prejudice from the trial court’s denial of her motion to strike the jury panel because there is nothing in the record to indicate that any of the twelve jurors who decided the case could not be fair and impartial to defendant as a result of negative publicity and gossip concerning Fischer. Such prejudice, the Majority reasons, could not exist where four of the twelve sitting jurors who had knowledge of Fischer’s exploits stated that they could nevertheless be fair and impartial to Defendant, and where the record does not show whether the other eight sitting jurors were even aware of Fischer’s behavior. The Majority also suggests that this case is somehow different because the publicity and gossip concerned Defendant’s counsel, rather than Defendant. *591Finally, the fact that the jurors’ verdicts were strongly supported by overwhelming evidence against Defendant leads the Majority to conclude that the jurors were not distracted from their duties to serve impartially.

¶ 88 Once again, the Majority’s Opinion evades the central question here — why is there no evidence in the record concerning the other eight jurors’ knowledge of Fischer’s conduct? Why is there “no way to know” from the record whether these other jurors were so offended by negative gossip concerning Fischer that it may have affected their ability to render a fair and impartial verdict in this case? I do not believe that a specific biased juror needs to be identified in order for prejudice to have resulted here; indeed, the very reason that we cannot point to a specific error in the record is that the requested additional voir dire question was never re-asked of the original panel members as requested by attorney Noland.

¶ 89 Eleven of the twelve jurors who rendered verdicts against Defendant in this case were drawn from the original thirty-four panel members. It must also be remembered that those panel members were asked if any of them would be unable or unwilling to render a verdict solely based on the evidence presented at trial before any of the discussion about Fischer — before eleven of the thirty-four revealed knowledge of the negative publicity and gossip, before three of those eleven admitted they were so disgusted by Fischer they would hold it against Defendant, and before two additional panel members informed the parties and the court that Fischer’s conduct was continuing to be discussed among the prospective jurors. This inquiry to prospective jurors also preceded Defendant’s request to conduct further voir dire of those panel members who had not been subjected to in-chambers questioning— a request that was never acted upon by the trial court. Thus, the jurors’ so-called assurances of impartiality implied at the outset of voir dire by their silence in response to the foregoing question is certainly not dispositive.

¶ 90 Some errors necessarily render a trial fundamentally unfair and cannot be harmless. This is one of them. Indeed, the harmless error analysis itself “presupposes a trial, at which the defendant, represented by counsel, may present evidence and argument before an impartial judge and jury.” Rose v. Clark, 478 U.S. 570, 577-78, 106 S.Ct. 3101, 92 L.Ed.2d 460 (1986). We cannot require a defendant to demonstrate prejudice resulting from an error by the trial court “when, as a practical matter the nature of the error renders it impossible to prove the extent of any prejudice.” Perkins v. Komarnyckyj, 172 Ariz. 115, 119, 834 P.2d 1260, 1264 (1992). Nor can overwhelming evidence of guilt be quantitatively balanced against such an error to render it harmless. As Justice Feldman recently pointed out in Anderson, 197 Ariz. at 323, ¶ 22, 4 P.3d at 378, “[ejrrors involving the composition of the court or jury affect the legitimacy of the entire proceeding, leaving nothing to measure or weigh and requiring reversal.”

¶ 91 There are further reasons why this error cannot and should not be considered harmless. The right to a trial by jury means not only a fair and impartial jury, but one lawfully constituted. State v. Zimmer, 106 Ariz. 166, 168, 472 P.2d 35, 37 (1970). “Arizona courts have long held a litigant who is denied the full use of the allotted peremptory challenges is denied a substantial right, which requires reversal, even absent an independent showing of prejudice.” State v. Huerta, 175 Ariz. 262, 263, 855 P.2d 776, 777 (1993). The trial court’s denial of the additional voir dire requested here made it impossible for defense counsel to intelligently utilize challenges for cause and peremptory strikes. This was the functional equivalent of depriving Defendant of those substantial rights, which should also lead to automatic reversal, without a showing of prejudice, under the reasoning of Huerta. The trial court also impaired Defendant’s right to the effective assistance of counsel by preventing defense counsel from developing the necessary information to exercise these challenges to the composition of the jury. See Zarabia v. Bradshaw, 185 Ariz. 1, 3, 912 P.2d 5, 7 (1996).

¶ 92 As to the Majority’s conclusion that the jurors’ opinions should matter less because they concern defense counsel, rather than Defendant, I am compelled to para*592phrase Justice Feldman’s own recent remarks on an analogous issue: “Arizona’s system implicitly and explicitly acknowledges that jurors’ views [of counsel] could affect their ability to impartially evaluate the defendant’s guilt. Otherwise, why do we voir dire at all on ... questions dealing with [counsel]? The issue is irrelevant unless we acknowledge that jurors’ views [on counsel] affect the verdict of guilt or innocence.” Anderson, 197 Ariz. at 320, ¶ 12, 4 P.3d at 375. And a juror’s view of counsel can affect his or her ability to be fair and impartial to the defendant.2

¶ 93 Given the nature of the gossip in this case, and the fact that two panelists told the court that the issue continued to be discussed among prospective jurors, it was imperative for the trial court to determine through additional voir dire, as requested by defense counsel, the extent to which the remainder of the jury panel had been prejudicially infected. Without an adequate voir dire, Defendant’s federal and state constitutional guarantees to an impartial jury were not honored.

. Due to a vacancy on the court, pursuant to Article 6, Section 3 of the Arizona Constitution, the Honorable Edward C. Voss, Chief Judge of the Court of Appeals, Division One, was designated to sit on this case.

. See Cox v. Norris, 133 F.3d 565, 571-72 (8th Cir.1997) (trial court properly excused potential juror for cause who was acquainted with and had animosity towards prosecutor in the case); Hughes v. United States, 689 A.2d 1206, 1210-11 (D.C.1997) (failure of trial court to strike juror who was close friend of former prosecutor or, at a minimum, to conduct additional voir dire to determine whether actual bias existed, constituted structural error, requiring reversal); People v. Barret, 145 A.D.2d 842, 535 N.Y.S.2d 829, 830-31 (N.Y.App.Div.1988) (potential for prejudice was "readily apparent" as a result of jurors' exposure to publicity and discussion of defense counsel's own indictment on drug charges, despite juror assurances of impartiality during voir dire).