Gattis v. State

637 A.2d 808 (1994)

Robert Allen GATTIS, Defendant Below, Appellant,
v.
STATE of Delaware, Plaintiff Below, Appellee.

Supreme Court of Delaware.

Submitted: December 16, 1993. Decided: February 28, 1994. Rehearing Denied March 24, 1994.

Jerome M. Capone, Wilmington, for appellant.

Richard E. Fairbanks, Jr. (argued), Chief of Appeals Div., Timothy J. Donovan, Jr., and Loren C. Meyers, Deputy Attys. Gen., Dept. of Justice, Wilmington, for appellee.

Before VEASEY, C.J., and HORSEY, MOORE, WALSH and HOLLAND, JJ., constituting the Court en Banc.

WALSH, Justice:

This is an appeal from the Superior Court following the imposition of a death sentence. The appellant, Robert Allen Gattis ("Gattis"), was convicted of Murder First Degree, Burglary First Degree and Possession of a Deadly Weapon by a Person Prohibited. The charges arose out of the shooting death of his girlfriend, Shirley Y. Slay ("Slay"). In the penalty phase of the trial, the jury, acting pursuant to 11 Del.C. § 4209(c)(3), determined by a vote of ten to two that the aggravating circumstances outweighed the mitigating circumstances. The Superior Court agreed with that determination and, accordingly, imposed the death sentence. This appeal followed.

Gattis asserts various claims of error relating to the admissibility of evidence in both the guilt and penalty phase of his trial. He also argues that the imposition of the death sentence was not proportionate to the offense committed. His principal contention, however, is that his trial jury was not randomly selected. Although we conclude that the jury selection process followed in this case lacked total randomness, it did not prejudice defendant's entitlement to a fairly constituted jury. We also find his other claims of error to be without merit and affirm his conviction. The imposition of the death sentence was not disproportionate and is also affirmed.

I

The State's evidence presented at trial portrayed the following sequence of events. Gattis and Slay had been in a boyfriend-girlfriend relationship for almost six years prior to her death on May 9, 1990. The relationship had been, in part, a stormy one. Gattis was extremely jealous and possessive of the victim. On one occasion, in 1987, Gattis confronted the victim in a bar talking to another man. Gattis threatened the victim with a loaded handgun, fired a shot into the floor and pistol-whipped the man. Gattis was prosecuted for his conduct and pleaded guilty to Assault Second Degree and Reckless Endangering First Degree. He received five years probation.

At one point in their relationship, Slay lived in an apartment in Wilmington to which Gattis had a key and where he would occasionally spend the night. In April, 1990, Slay moved to an apartment outside Wilmington, partly to avoid contact with Gattis. On the day of her death, Slay advised her supervisor at work that she intended to terminate her relationship with Gattis that evening.

On May 9, Slay went to a softball game after she left work. Gattis was also at the game and apparently persuaded a reluctant Slay to leave with him. The two went to Slay's apartment where they argued over Gattis' inability to make telephone contact with Slay the previous evening. Slay claimed her telephone was not operating but Gattis arranged for a friend to telephone the apartment and, when the phone rang, he became enraged. He accused Slay of seeing another man, beat her, and left. Slay called the police. While the police were present in Slay's apartment, Gattis called several times. One officer spoke with Gattis, warning him to have no contact with Slay.

Gattis returned to Slay's apartment later that evening after borrowing a friend's car, apparently to avoid detection by the police. Outside the apartment, Gattis encountered a friend and neighbor of Slay, Linda Watson, who advised Gattis to stop fighting. The neighbor then telephoned Slay to advise her that Gattis was back. Watson then heard the sound of Slay's door being knocked in. As Watson ran out of her apartment, she heard a gunshot and saw Gattis running down the stairs. Slay died from a gunshot wound between her eyes. Autopsy results indicated that she was shot by a .38 caliber handgun from a distance of 4 to 18 inches. Gattis surrendered to police the following day after admitting himself to a mental hospital.

Gattis testified in his own defense. He claimed that the shooting of Slay was an accident which occurred when the gun discharged while he struggled with her to enter the apartment. The jury rejected this explanation and found Gattis guilty of Murder First Degree as well as the burglary and weapons offense. After hearing evidence and argument in the penalty phase, the jury decided that the aggravating factors outweighed the mitigating factors by a vote of ten to two. The trial judge subsequently received additional argument from counsel and separately determined that the death sentence should be imposed.

II

Gattis' principal claim of error on appeal is that his trial jury was not selected through a random process. In order to address this argument, it is necessary to examine the method used by the Superior Court to assemble the jurors for Gattis' trial and to select from that group the jurors to hear the evidence.

For Gattis' trial, about 400 people were summoned from the qualified jury wheel^[1] to report to the courthouse. There is no indication that the process used to assemble this group was invalid and Gattis does not challenge it on appeal. However, on the first morning of trial, as each juror arrived at the courthouse to report to the jury manager, he or she was assigned a sequential number until a group of 150 people had been assembled. The trial judge then called the prospective jurors, by their sequential numbers, in groups of 20, for individual voir dire. The effect of this was that the assembled panel of prospective jurors were interviewed in the order that each arrived at the courthouse, or on a "first-come, first-served" basis.

The first 35 prospective jurors called for voir dire were white. Gattis is black, as was the victim. Gattis' counsel became suspicious and inquired of the jury manager concerning the method of voir dire. He was advised of the sequential number assignment. Defense counsel then moved for a stay of the proceedings or a dismissal, charging that the jury selection had not complied with procedures mandated by the Delaware Jury Selection and Service Act ("Delaware Jury Act" or "Act"), 10 Del.C. Ch. 45.

At an evidentiary hearing, the Superior Court jury manager testified that the method used in Gattis' trial differed from that used in regular cases and in capital cases prior to May. 1991. Historically, and in contrast to the procedure used here, after the full panel of prospective jurors was assembled, the clerk would draw the names of the jurors from a box and assign numbers for individual voir dire based on the order of the draw. In May, 1991, the process was modified to reduce the amount of time the clerks devoted to jury selection. Under the "first-come, first-served" system, court clerks assigned numbers to jurors as they reported at 9:30 a.m., even though court was not in session and neither counsel nor the defendant were present.

The trial judge, after an evidentiary hearing, denied Gattis' motion to dismiss the panel. The court noted that the racial composition of the 99 jurors who were subjected to individual voir dire reflected 10 percent black representation — the same percentage reflected in the entire assembly of 400 jurors. The court observed that while the first 35 jurors individually questioned were white, the remaining 64 jurors questioned included 10 blacks. Since the final racial percentage of the voir dire reflected that of the entire jury assembly, the racial composition of the early jurors was deemed "irrelevant."

Gattis' attack upon the selection and impanelment of his jury is three pronged. First, he argues that it was improper to conduct the early assignment of numbers to jurors outside of the presence of himself and his counsel. Next, he contends that the trial court erred in providing both parties, in advance, with the names of the jurors in the order in which they would be questioned. Finally, he asserts that the selection process, based on a first-come, first-served arrangement, deprived him of a jury selected at random.

A.

As previously noted, the assignment of numbers to jurors in this case began as jurors appeared, beginning about one-half hour before court formally convened. Gattis claims that he had a right to be present when these assignments were made and his absence constitutes reversible error. Superior Court Criminal Rule 43(a) provides that a defendant shall be present during all phases of the trial "including the impaneling of the jury." Gattis concedes that he raised no objection to his absence in the trial court but argues that the matter is one of plain error.

Under the plain error standard of review this Court will consider errors not raised below only where the claim implicates material defects which are fundamental in character "and which clearly deprive an accused of a substantial right, or which clearly show manifest injustice." Wainwright v. State, Del.Supr., 504 A.2d 1096, 1100 (1986); see also Wright v. State, Del.Supr., 633 A.2d 329, 336 (1993). While entitlement to trial by jury is, in a generic sense, a fundamental right, Claudio v. State, Del.Supr., 585 A.2d 1278 (1991), not every aspect of the jury selection process requires the participation of the defendant. The process of distributing numbers to prospective jurors is purely ministerial and the defendant's presence during that procedure would not have altered the process in any respect, or served to protect any substantial right. United States v. Bordallo, 9th Cir., 857 F.2d 519, 522-23 (1988), cert. denied, 493 U.S. 818, 110 S.Ct. 71, 107 L.Ed.2d 38 (1989). Moreover, Gattis was afforded the opportunity to raise objections to any aspect of the jury selection process under 10 Del.C. § 4512,^[2] and, indeed, did so with respect to the lack of randomness in the individual voir dire process. Prior to filing his motion, Gattis' counsel had spoken to the jury manager and was fully aware not only of the first-come, first-served procedure, but also of the assignment of numbers to early arriving jurors. In view of the ministerial nature of that aspect of the procedure, we conclude that any deficiency did not affect the "fairness and integrity of the trial process." Wainwright, 504 A.2d at 1100. Accordingly, any claim of error in this connection is deemed to have been waived by the failure to object and is not reviewable under a plain error standard.

B.

Gattis next contends that the trial judge abused his discretion in allowing the attorneys for both the State and the defendant to secure, in advance, a list of the names of jurors to be individually questioned. This objection was raised at trial, but the trial judge was of the view that since the defendant, as well as the State, had equal access to the list, there was no advantage to either side.

The practice of permitting counsel to know in advance the order of individual voir dire was criticized and disapproved by this Court in Robertson v. State, Del.Supr., 630 A.2d 1084 (1993). The practice heightens the risk of the exercise of peremptory challenges in a discriminatory manner and should not be permitted. The Robertson holding, however, operates prospectively and does not serve to invalidate, per se, the use of the jury list in cases tried before its announcement. The issue is whether pre-knowledge of the order of jurors to be questioned, on a mutual basis, created prejudice which affected the random process of jury selection. In the absence of a showing of prejudice, the matter is within the trial judge's discretion. Burke v. State, Del. Supr., 484 A.2d 490, 498 (1984).

As we noted in Robertson, knowledge of the precise order for individual voir dire creates "the potential for compromising the random selection of jurors." 630 A.2d at 1092. In Robertson, however, it was claimed that the prosecutor used the advance knowledge of individual juror appearance to exercise race discriminatory challenges. No such claim has been asserted here and Gattis is unable to point to the exclusion of a cognizable group from the panel ultimately selected to try him. There is thus no claim that a scheme of juror manipulation existed or that the use of a juror list promoted that scheme. While the potential for prejudice and abuse was obviously present here, in the absence of a showing of some palpable discrimination in the selection of the jury, Gattis' Robertson claim does not rise to the level of reversible error.

C.

Gattis' principal claim, which was the basis for the challenge to the impanelling of his jury at his trial, is that the individual voir dire of jurors based on the order of their appearance at the courthouse deprived him of a jury selected at random. The State concedes that the "first-come, first-served" method employed for Gattis' trial did not attain a level of statistical randomness, but argues that Gattis is unable to demonstrate that the practice employed constituted a substantial deviation from the Delaware Jury Act or resulted in a jury which did not represent a fair cross section of the community. In the absence of such a showing of prejudice, it is argued, Gattis' claim must be rejected.

Our analysis of the lack of randomness claim requires that we examine the statutory and constitutional underpinnings of this aspect of the right to trial by jury. In Delaware, the right to trial by jury implies an entitlement to a jury system which contains all the fundamental features which attended that right at common law. Claudio v. State, Del.Supr., 585 A.2d 1278 (1991). This right is not co-extensive with the jury trial guarantees afforded by the United States Constitution, whose framers declined to include the full panoply of the common law jury system. Id. at 1295. Over time, the General Assembly has by statute provided for the implementation of the right to trial by jury through various procedures and court rules. These procedures, however, may not be applied or construed in such a manner as to alter the fundamental features of the jury system as it existed "heretofore," i.e. at common law. Id. at 1298 (citing Ruffin v. State, Del.Supr., 50 Del. 83, 123 A.2d 461 (1956)); see also Hopkins v. Justices of Peace Court No. 1, Del.Super., 342 A.2d 243 (1975).

The Delaware Jury Act is patterned after the federal Jury Selection and Service Act ("Federal Act"), 28 U.S.C. §§ 1861 et seq. Its announced policy, like that of its federal counterpart, is to insure that jurors "be selected at random from a fair cross section of the population" and to provide the opportunity for "all qualified persons ... to be considered for jury service." 10 Del.C. § 4501. Under the Act, the Superior Court is enjoined to adopt a jury selection plan to implement the "policy and provisions" of the Act. 10 Del.C. § 4507(a).^[3] The Act provides for randomness in jury selection in general terms: "Prospective jurors shall be selected randomly from the [qualified jury wheel for assignment to grand, petit and special jury panels] from time to time as needed." 10 Del.C. § 4508(a). Both the Act and the jury plan adopted by the Superior Court (the "Plan") insure randomness in the composition of the panel by precise designation of the jury pool, 10 Del.C. § 4507(a)(2); the make-up of the master list derived from the pool, 10 Del.C. § 4503(6); the designation of persons disqualified from jury service, 10 Del.C. § 4509; and the basis for excusal of otherwise qualified jurors, 10 Del.C. § 4511. Significantly, neither the Act nor the Plan impose specific procedures beyond the point of jury assembly in the courthouse to insure that randomness will occur in the process by which the names of individual jurors are called to enter the jury box or for individual voir dire.

The absence of a random selection method extending to the individual selection of jurors is in sharp contrast to the specific statutory procedure in effect prior to the 1975 adoption of the Act. 60 Del.Laws C. 225. Prior to its repeal, 10 Del.C. § 4518 set forth the precise manner in which individual jurors would be selected, following their placement on assembled panels by the Jury Commissioners:

§ 4518. Impaneling petit jury.

(a) The Prothonotary shall, under the direction of the Court, write the name of each petit juror, returned and impaneled as provided in this subchapter, upon a distinct ballot, all the ballots being of the same color and size. He shall roll up the ballots, all in the same manner, as near as may be, and put them together in a box to be provided for the purpose. When a cause is called for trial, an indifferent person shall, in open court, draw out 12 of the ballots, successively, and if any of the jurors, whose names are so drawn, shall not appear, or shall be challenged and set aside, then a further number shall be drawn, until 12 jurors appear, and, upon challenge, are allowed. The 12 persons first drawn, appearing and allowed, their names being marked in the panel, and they being sworn or affirmed, shall be the jury to try the cause. The names of the persons so sworn or affirmed shall be kept apart in another box to be provided for the purpose, until the verdict of such jury has been rendered, or until the jury is discharged by the Court. Then the same names shall be rolled up and returned to the first mentioned box, to be kept with the other names remaining then undrawn; and so repeatedly, as long as any cause remains for trial. (emphasis added).

Although we are advised that the practice of drawing names from the "box" has been the usual practice in the Superior Court since the repeal of § 4518, that traditional aspect of random jury selection is no longer statutorily required, nor does the Plan require it.^[4] Its absence as a standard constitutes a gap in the jury selection process. More importantly, the lack of a random drawing requirement in this case permitted the Superior Court to exercise its discretion in permitting an alternative method based on jurors' reporting times. To the extent the individual voir dire occurred in this case according to a jury manager's plan for expediting selection, the process was not entirely random. The question posed is, however, whether the use of a first-come, first-served procedure created such a lack of randomness as to prejudice the defendant's right to be tried by a fair cross section of the community.

The State argues that the first-come, first-served method used at Gattis' trial does not offend the Plan, nor the underlying statutory scheme, since neither purports to define the standards for insuring randomness beyond the assembly of the panel. Moreover, the State argues, even if a deviation exists, there must be a substantial failure to comply with the Act to invalidate a resulting jury verdict. These arguments have essential merit but their reach is limited.

While it is true that the sequence for individual voir dire is no longer statutorily fixed, or prescribed by the Plan, the requirement of randomness in every phase of jury selection is a continuing one. The spirit of the Act, contained in the statement of policy, mandates selection of jurors "at random from a cross section of the population" of the county. 10 Del.C. § 4501. The policy of the Act would be ill-served if randomness ended once the jury was assembled in the courtroom. Even in the absence of a statute or jury plan, the impaneling of a jury without assurance of randomness in every aspect of the selection process would be contrary to an essential safeguard attending that right at common law and thus offend the constitutional command that "trial by jury shall be as heretofore." Del. Const. art. I, § 4; Claudio, 585 A.2d at 1297.

If it be assumed that the Act requires, in spirit at least, randomness in all phases of juror selection, nonetheless a departure from the requirements of the Act will be deemed error only if the departure constitutes "a substantial failure to comply" with the statute. 10 Del.C. § 4512(b). In Celotex Corp. v. Wilson, Del.Supr., 607 A.2d 1223 (1992), this Court analyzed the Act in the context of a claim that jurors were excused without a showing of undue hardship allegedly resulting in a non-representative jury panel. In Celotex we held that "technical deviations, or even a number of them, that do not frustrate the random selection and fair cross section requirements" of the Act will not be a deemed substantial failure. Id. at 1228.

Even apart from the policy requirements of the Act, and to the extent that the Act's literal standard of randomness does not extend to the individual voir dire stage, the lack of randomness must bear some relationship to the guarantee of a fair cross section of the community in the jury composition. In short, a party asserting a lack of randomness, in the absence of a substantial deviation from the statutory plan, must demonstrate a resulting exclusion of a constitutionally cognizable group.

Courts which have addressed the question of resulting prejudice overwhelmingly have concluded that such a showing is necessary. When the original jury pool is valid, the selection process from that pool will be considered "random" in the "absence of any arbitrary attempt to exclude a class of persons from the jury." McClendon v. United States, 8th Cir., 587 F.2d 384, 387 (1978), cert. denied, 440 U.S. 983, 99 S.Ct. 1793, 60 L.Ed.2d 244 (1979) (footnote omitted); United States v. Davis, 10th Cir., 518 F.2d 81, cert. denied, 423 U.S. 997, 96 S.Ct. 425, 46 L.Ed.2d 371 (1975); United States v. Butts, M.D.Fla., 514 F.Supp. 1225, 1236 (1981) ("absent a showing that some cognizable group was excluded from the jury selection process, no substantial violation will lie."); United States v. Haley, N.D.Ga., 521 F.Supp. 290, 295 (1981) (principles of randomness violated by impermissible discrimination against individuals or groups); United States v. Gregory, 11th Cir., 730 F.2d 692, 699 (1984), cert. denied, 469 U.S. 1208, 105 S.Ct. 1170-71, 84 L.Ed.2d 321 (1985) (discriminatory selection of jurors required to violate Federal Act).

However, some federal cases have held that a party need not show actual prejudice to establish a "substantial failure" to comply. United States v. Nelson, 9th Cir., 718 F.2d 315, 318 (1983); United States v. Okiyama, 9th Cir., 521 F.2d 601, 604 (1975). In United States v. Kennedy, 5th Cir., 548 F.2d 608, cert. denied, 434 U.S. 865, 98 S.Ct. 199, 54 L.Ed.2d 140 (1977), the court ruled that a statutory violation which directly affects the random nature of the jury selection is sufficient without demonstrating that the violation excluded some cognizable group. The court rejected the challenge on appeal, however, because the defendant had failed to perfect his objection to the jury at the trial level with the appropriate sworn statement required by the Federal Act. Significantly, the court also rejected the defendant's claim that the use of volunteer jurors violated his due process rights. The court noted that such a claim requires a showing the selection process "tended to exclude or underrepresent some discernable class of persons...." 548 F.2d at 614.

The federal cases have upheld many methods of nonrandom jury selection which did not substantially fail to comply with the Federal Act. In United States v. Eyster, 11th Cir., 948 F.2d 1196 (1991), the court upheld a plan which arranged the jury pool in alphabetical order. Accord United States v. Haley, N.D.Ga., 521 F.Supp. 290 (1981). In United States v. Smith, 5th Cir., 588 F.2d 111 (1979), the court upheld a process where the clerk placed the names of persons who served more than two years previously and those who had been called, but not served, back into the qualified jury wheel. In United States v. Gometz, 7th Cir., 730 F.2d 475 (en banc), cert. denied, 469 U.S. 845, 105 S.Ct. 155, 83 L.Ed.2d 92 (1984), the court upheld a system where the qualified jury wheel was made up entirely of only those persons who responded to jury questionnaires even though they represented only 30 percent of the total number of persons contacted.

Probably the most analogous cases in the federal system are those dealing with volunteer jurors. Two circuits have found substantial violations of the Federal Act where volunteers were used to compose juries, reasoning that allowing a subjective element of opting in or out of jury selection offends the statute. Kennedy, 548 F.2d at 610-12; United States v. Branscome, 4th Cir., 682 F.2d 484, 485 (1982) (grand jury). See also United States v. Layton, N.D.Cal., 519 F.Supp. 946, 955 (1981) ("we have no doubt that widespread use of volunteer jurors would constitute a substantial failure to comply with the Act."). Not all courts accept a per se rule that the use of volunteers is error. In United States v. Ramirez, 1st Cir., 884 F.2d 1524, 1530 (1989), the court held that the use of volunteers offended the random selection requirement of the Federal Act but that the violation was not substantial. Further, some courts have interpreted Kennedy to condemn only the prospect of giving jurors complete discretion on whether to serve on a jury. United States v. Barnette, 11th Cir., 800 F.2d 1558, 1567-68 (1986) cert. denied 480 U.S. 935, 107 S.Ct. 1578, 94 L.Ed.2d 769 (1987); United States v. Bearden, 5th Cir., 659 F.2d 590, 603 (1981), cert. denied, 456 U.S. 936, 102 S.Ct. 1993, 72 L.Ed.2d 456 (1982). Although the federal precedent is somewhat ambiguous, the weight of the authority favors the requirement demonstrating the exclusion of a cognizable group.

State court decisions pertinent to this issue also yield mixed results and are of limited assistance. In State v. Long, 204 N.J.Super. 469, 499 A.2d 264 (1985), a county wide jury selection process based on a master pool arrangement similar to that in effect in Delaware was invalidated prospectively. The court ruled that deviation from the jury plan rendered the process non-random, and thus invalid, without a showing that any cognizable class had been excluded from the pool or that there had been any resulting prejudice. As noted, Gattis does not here attack the randomness of the pool selection. In State v. Wagner, 180 N.J.Super. 564, 435 A.2d 1190 (1981), seating jurors in the order of their appearance was deemed reversible error in the absence of a showing of prejudice because it was contrary to the statute which required drawing at random from the box. Similarly, in State v. Hoffman, La.Supr., 345 So.2d 1 (1977), calling jurors for voir dire in the order their names were listed on the panel sheet was invalidated because it was contrary to the statutorily defined procedure. However, in People v. Wright, Cal.Supr., 52 Cal.3d 367, 276 Cal.Rptr. 731, 748-49, 802 P.2d 221, 238-39 (1990), individual voir dire based on arrival times of jurors, while considered "unorthodox," was not deemed a "material departure" from statutory procedure since jurors were not aware that their order of appearance would dictate selection.

There is no question that the process used in this case for selecting jurors for individual voir dire did not conform to the common law practice or that required under the now-repealed § 4518, which was basically a codification of the common law practice of selecting jurors for voir dire by lot. See note 4 supra. It is equally evident that, while the policy of the Delaware Jury Act promotes randomness of the entire jury selection process, neither the Act nor the Plan prescribes a specific method for insuring randomness in jury selection beyond the point of panel assembly. In the absence of a specific statutory requirement, to the extent that the process used in this case offended the policy of randomness, we hold that such a deviation is subject to a harmless error analysis, i.e., a determination of actual prejudice resulting from an egregious infringement of defendant's right in the Delaware Constitution to trial by jury "as heretofore." Claudio, 585 A.2d at 1304.

The error of permitting individual voir dire based on order of juror appearance in this case did not undermine the essential purpose underlying randomness in the jury selection process — the prohibition of discriminatory exclusion from jury service. As the trial judge found after comparing the racial profile of the jurors individually voir dired against the composition of the full panel, the minority percentage was exactly the same. Moreover, no person has been shown to have been rejected for individual voir dire because of the application of any subjective criteria or plan.

Whatever deviation from the policy of randomness occurred, it did not operate to "distort[] the representativeness of the jury." Kennedy, 548 F.2d at 614. Gattis is unable to identify, beyond mere speculation, what characteristics, relevant to jury service, would be reflected in the early-appearing jurors. Our independent reflection on the matter also fails to yield a cognizable group. In the absence of such a showing, while the method utilized here was in violation of a practice subsumed under Delaware Constitution's guarantee of trial by jury "as heretofore," we are satisfied that such error was harmless beyond a reasonable doubt. Van Arsdall v. State, Del.Supr., 524 A.2d 3, 11 (1987) (quoting Chapman v. California, 386 U.S. 18, 24, 87 S.Ct. 824, 828, 17 L.Ed.2d 705 (1967));^[5]accord Claudio, 585 A.2d at 1304.

Although we deem the error harmless when viewed against the lack of demonstrable prejudice, the practice of first-come, first-served jury selection lends itself to the possibility of abuse. Efficiency in jury management is a desirable goal but certain fundamental features in the jury selection do not permit methods or devices which save time at the expense of randomness.

The common law requirement of selection of individual jurors in open court is a time-honored attribute of trial by jury "as heretofore" in Delaware. It provides not only a theoretical means to achieve randomness, but a public demonstration that randomness is occurring. We are informed that the "first-come, first-served" arrangement which occurred here is no longer the practice in the Superior Court. To insure that so important a matter is not left to practice or innovation, however, the Superior Court is directed to amend the Plan as expeditiously as practicable to provide for the specific implementation of the common law practice exemplified by the former § 4518 or an equivalent procedure consistent with the American Bar Association Standards Relating to Juror Use and Management.^[6]

III

A.

Prior to trial, Gattis sought a ruling from the court restricting the scope of the evidence which the State proposed to present concerning the previous relationship between Gattis and the victim. Specifically, Gattis sought to limit statements made by the victim to third parties regarding her fear of the defendant and her intention to terminate their relationship. The trial court ruled that the State could present evidence concerning the "general characteristics of the relationship" by those who observed it, but not specific acts of violence or other evidence of his jealousy or possessive nature. Gattis contends that the prosecutor, in his opening statement, exceeded the limits of the pretrial ruling. That objection was asserted at trial and the trial judge admonished the prosecutor to stay within the limits of the ruling but permitted discussion of the defendant's character in the context of his relationship with the victim. Gattis complains that the trial court permitted the prosecutor to present argument beyond the general characteristics of the relationship.

In a prosecution for homicide arising out of a marital or romantic relationship, evidence of previous discord between the victim and the defendant is clearly material to issues of motive and intent. Hutchins v. State, Del.Supr., 153 A.2d 204, 207 (1959). Here, the trial judge was careful to restrict the State, in its opening statement, from describing specific acts of violence by the defendant against the victim while permitting discussion of the general characteristics of the relationship. The court correctly ruled that evidence of prior bad acts directed to the victim could be presented only in rebuttal if the defendant testified concerning accident or lack of intent. Getz v. State, Del.Supr., 538 A.2d 726 (1988).

The setting of limits on the opening statements of counsel, and securing compliance with such restrictions falls within the supervisory authority of the trial judge, as a matter of discretion. Holmes v. State, Del. Supr., 422 A.2d 338, 340 (1980). The trial judge here exercised that control carefully in an effort to guard against the injection into the prosecutor's opening statement of evidence which could not be presented in the State's case-in-chief. We find no abuse of that discretion and Gattis has not articulated in what respect his trial strategy was altered by reason of the trial court's rulings. Accordingly, we find no merit in this contention.

Similarly, Gattis contends that a State's witness, Ruth Ann Noel, was permitted to give testimony beyond the limits of the trial court's "bad acts" ruling by testifying that the defendant had "obviously followed" the victim to her apartment on the day she moved into the apartment. There was no objection made to this testimony when given and it does not rise to the level of plain error. Wainwright, 504 A.2d at 1100. In any event, the act of following the victim to her apartment, while relevant as background, does not constitute misconduct, or a bad act, within the Getz proscription.

After a full review of the record, we are satisfied that the State was not permitted to exceed the bounds set by the trial judge in his ruling concerning the introduction of evidence of the prior relationship between the defendant and the victim.

IV

Gattis contends that the trial court erred in its refusal to grant a mistrial after two police officers were permitted to give allegedly prejudicial testimony. The first instance occurred when a police officer testified that he knew Gattis "from my days with the Wilmington P.D." Gattis immediately objected and argued that such testimony would permit the jury to infer a prior criminal history. The trial court upheld the objection and instructed the jury to "totally disregard" the officer's remark. Although we agree that the testimony was objectionable, it was of marginal significance. There was no direct reference to prior criminal conduct and, as the trial judge noted, the officer's previous contact with the defendant could have been in a context other than arrest. An error resulting from inadvertent presentation of irrelevant testimony is usually cured by the trial court's immediate instruction. Diaz v. State, Del.Supr., 508 A.2d 861, 866 (1986); Sawyer v. State, Del.Supr., 634 A.2d 377, 380 (1993). We find no error in the trial court's treatment of this matter.

A related claim of error is directed to an objection made during the testimony of another police officer who had taken Gattis into custody. During direct examination, the officer was asked to describe the defendant's demeanor when taken into custody. The officer replied: "Well, he was visibly upset.... He was emotional. He was cooperative. He did everything I asked him to do. He made no statements." Defense counsel objected that the officer's testimony represented an impermissible comment on the defendant's right to remain silent. The trial court agreed but denied a motion for a mistrial. The court then gave a curative instruction.^[7]

Again, we review a claim of error based on the refusal of a trial judge to grant a mistrial under an abuse of discretion standard. Sawyer, 634 A.2d at 379. While the State may not comment on, or permit any witness to refer to, a defendant's silence at the time of arrest, Bowe v. State, Del.Supr., 514 A.2d 408, 411 (1986), not every reference to a defendant's exercise of the Fifth Amendment right to remain silent warrants reversal. Shantz v. State, Del.Supr., 344 A.2d 245, 247 (1975) (where testimony concerning a defendant's silence was not intentionally elicited, a cautionary instruction, in the discretion of the trial judge, will suffice). Here the objectionable testimony was not directly elicited by the State. Indeed, the officer's recounting of what the defendant said was not responsive to the question which sought a description of his demeanor. The trial judge acted quickly and effectively to blunt whatever prejudice was conveyed by the testimony. Under the circumstances, the refusal to grant a mistrial was not an abuse of discretion.

Gattis' final claim of error in the guilt phase of the trial is directed to a trial court's restriction on courtroom demonstration. At trial, Gattis testified that the gun accidentally discharged while he and the victim struggled at the door to her apartment. At side bar, Gattis' counsel requested permission of the court for the defendant to leave the witness stand and go to the robing room door to demonstrate his position at the apartment door. The court ruled that such a demonstration might present a security problem but would be permitted in the presence of guards. Defense counsel then indicated he would not pursue the request if the prosecution, on cross examination, did not ask the defendant to demonstrate his position at the door. Gattis now claims that the prosecutor disregarded that "understanding" during cross examination by asking the defendant to demonstrate while in the jury box.

Our reading of the record does not support the defendant's contention that the trial judge restricted his use of demonstration evidence. The trial judge granted permission provided security measures were taken, but the defendant did not pursue the matter. To the extent that the prosecutor may have breached an "understanding" concerning the method of demonstration, defendant made no objection to the trial court concerning the departure. Because of the marginal significance of the issue and the fact that it was not brought to the attention of the trial judge, we decline to review it under a plain error standard. Supreme Court Rule 8; Wainwright, 504 A.2d at 1100.

V

Gattis' next claim of error relates to the sentencing phase of the trial. The State, pursuing its right to present victim impact evidence, offered the testimony of the victim's mother concerning the effect of her daughter's death. The prosecutor asked: "How difficult was it for you to hear about your deceased daughter using cocaine, hearing that the defendant said she stayed out at night, and him testifying about how it was an accident." The mother replied that, other than the burial of her child, "the trial had to be the hardest thing I had to deal with in my life" particularly to "listen to the way he sits up there and says it was an accident." Defense counsel objected on the ground that victim impact testimony may not include opinions by family members about the crime or the defendant. While the trial court did not agree that the mother's response was totally inadmissible, the court did instruct the jury that "[h]ow a family member views the trial of the case is not particularly relevant in this direction." Gattis now claims this instruction was legally insufficient, although he raised no objection to it at trial.

In Payne v. Tennessee, 501 U.S. 808, 111 S.Ct. 2597, 115 L.Ed.2d 720 (1991), the United States Supreme Court ruled victim impact statements admissible as a "method of informing the sentencing authority about the specific harm caused by the crime in question." 501 U.S. at ___, 111 S.Ct. at 2608. Following Payne, this Court ruled that victim impact evidence was deemed "relevant and admissible" in the penalty proceedings mandated by 11 Del.C. § 4209(c). Petition of the State of Delaware, Del.Supr., 597 A.2d 1, 3 (1991).

The rationale for permitting the State to present victim impact evidence is to permit the jury, and the sentencing judge, to bear in mind the harm caused by the victim's death as an aggravating circumstance at the same time it considers the mitigating evidence introduced by the defendant. Petition of the State of Delaware, 597 A.2d at 3 (citing Payne, 501 U.S. at ___, 111 S.Ct. at 2608-09). Although the focus of victim impact testimony should be on the crime itself and not on the trial aftermath, it is difficult to separate the events in the minds of the victim's relatives. It was thus understandable that the victim's mother voiced her resentment at the defendant's attempt to besmirch her daughter's character as part of his defense. The trial judge's cautionary instruction, while somewhat equivocal in its language, served to remind the jury that the impact of the crime, not the trial, is deserving of consideration.

Moreover, to the extent any victim impact evidence was received, it was directed merely to the establishment of a non-statutory aggravating circumstance. In view of the fact that the jury, in its guilty verdict, determined that the State had proved two statutory aggravating factors, the victim impact evidence was of limited significance in the total mix of aggravating circumstances or in weighing all aggravating factors against the mitigating factors. Similarly, the trial judge, who had the ultimate sentencing responsibility in striking the aggravating circumstances/mitigating circumstances balance, confined the victim impact evidence within the narrow bounds of "the pain and trauma which [the victim's] death has caused." Thus, it is clear that the trial judge, the ultimate determiner of the sentence, considered the victim impact evidence properly. We find no basis for concluding that the victim impact evidence in this case was unfairly considered by the jury or the court. Accordingly, we find no error in this regard.

VI

Gattis' offense occurred prior to the 1991 adoption of the new Delaware death penalty statute, 11 Del.C. § 4209. He asserts, therefore, that the application of that law to his prosecution violates the ex post facto clauses of the State and federal constitutions. This Court rejected an identical claim in State v. Cohen, Del.Supr., 604 A.2d 846, 852-55 (1992). We have consistently adhered to our ruling in Cohen in subsequent decisions, Dawson v. State, Del.Supr., 637 A.2d 57, 60-61 (1994); Wright v. State, 633 A.2d 329, 343 (1993); Red Dog v. State, 616 A.2d 298, 305-06 (1992), and do so here.

VII

Finally, we consider the question of whether the imposition of the death sentence on Gattis was proper under pertinent statutory standards. This review is mandatory under 11 Del.C. § 4209(g)(2),^[8] but is also warranted by Gattis' claim that the imposition of the death penalty was disproportionate to the penalty imposed in similar cases.

In Wright, 633 A.2d at 339, this Court outlined the format which governs our mandatory review:

This Court has traditionally commenced its mandatory statutory review by initially addressing subparagraph (b) of Section 4209(g)(2). Pennell v. State, [604 A.2d 1368, 1375 (1992)]. That subsection requires this Court to examine the evidence in the record to determine whether it supports the findings of the Superior Court judge which relate to the establishment of statutory aggravating circumstances. 11 Del.C. § 4209(e). Thereafter, two additional inquiries are required by subparagraph (a) of Section 4209(g)(2): first, whether the Superior Court judge's imposition of the death penalty was either arbitrary or capricious; and second, whether the death penalty imposed was disproportionate to the penalty imposed in similar cases arising under this statute. Pennell v. State, 604 A.2d at 1375; Riley v. State, Del. Supr., 496 A.2d 997, 1026 (1985) [cert. denied, 478 U.S. 1022, 106 S.Ct. 3339, 92 L.Ed.2d 743 (1986)]. "Each question requires a consideration of the totality of evidence in aggravation and mitigation which bears upon the particular circumstances or details of the offense and the character and propensities of the offender." 11 Del.C. § 4209(g)(2); Pennell v. State, 604 A.2d at 1375.

(quoting Red Dog, 616 A.2d at 306-07). See generally Gregg v. Georgia, 428 U.S. 153, 96 S.Ct. 2909, 49 L.Ed.2d 859 (1976).

In its verdict following the penalty hearing, the jury unanimously concluded that the State had established beyond a reasonable doubt the existence of two statutory aggravating circumstances: (1) that the murder had been committed during the commission of a burglary, 11 Del.C. § 4209(e)(1)(j); and (2) that the defendant had previously been convicted of a violent felony, 11 Del.C. § 4209(e)(1)(i). Gattis does not contest the finding of these two statutory aggravating circumstances. Thereafter, by a vote of ten to two, the jury, after weighing all relevant evidence in aggravation and mitigation bearing upon the offense and the character and propensities of the defendant, found by a preponderance of the evidence that the aggravating circumstances outweighed the mitigating circumstances. 11 Del.C. § 4209(c)(3).

Under the Delaware death penalty statute, the trial judge makes the final determination of whether the death penalty should be imposed, after consideration of the jury's findings. While the trial judge must consider the recommendation of the jury, he or she functions independently in deciding whether to impose the death penalty or life imprisonment based upon the existence of aggravating circumstances and a weighing of such circumstances against any mitigating circumstances. Wright, 633 A.2d at 335.

The trial judge, in a lengthy written decision, determined the existence of the same statutory aggravating factors found by the jury. Additionally, he articulated the following non-statutory aggravating circumstances: (1) that the killing was a "cold-blooded and deliberate" murder in execution style "carried out because of the defendant's misplaced and ill-conceived notions of infidelity" on the part of the victim, (2) the defendant's propensity towards violence and threats of violence based on his criminal history, (3) the impact of the victim's death upon members of her family, a factor to which the trial judge did not give "undue weight," (4) the defendant's lack of respect toward authority as reflected in his military service and his "Other than Honorable Discharge" for unsuitability, and (5) the defendant's non-compliant conduct while under court supervision as the result of previous criminal convictions.

In evaluating the evidence in mitigation, the trial judge considered the defendant's personality disorder characterized by an explosive temper and his obsessive and possessive relationship with the victim. The court noted that in his childhood the defendant had experienced abuse and been exposed to domestic violence. Several witnesses testified to the defendant's positive personal traits of helpfulness and caring both before and during his imprisonment. The court also noted the defendant's feeling of remorse which he expressed to the victim's family during the penalty hearing.

In concluding that the totality of the circumstances warranted a finding that the aggravating circumstances outweighed the mitigating circumstances, the trial court observed:

This was not a typical case of a domestic homicide. The defendant had no proprietary interest in the victim's place of residence. No child custody battles loomed on the horizon. Except on a sporadic, on-again-off-again basis, the defendant and the victim had not lived together for any extended period of time. Significantly, the defendant was intimate with another woman at the time of the murder.

* * * * * *

Here, two statutory aggravating circumstances were proven beyond a reasonable doubt. Moreover, the murder was committed in a callous manner without moral or legal justification. It was planned beforehand with adequate time for the defendant to contemplate the consequences of carrying out his intent. Further, the victim, tired of her relationship with the defendant, had endeavored to start a new life on her own. (footnotes omitted).

We have reviewed the record in this case and conclude that the Superior Court complied fully with the statutory standards for the imposition of the death penalty. 11 Del.C. § 4209(d). The record reflects that the Superior Court's decision to impose the death sentence was "the product of a deliberate, rational and logical deductive process." Red Dog, 616 A.2d at 310. We thus conclude that the sentence of death was not imposed on Gattis by the Superior Court arbitrarily or capriciously. 11 Del.C. § 4209(g)(2)a.

Notwithstanding our determination that the imposition of the death sentence in this case was not arbitrary or capricious, we must nonetheless consider whether the imposition of the death sentence was disproportionate to the penalty imposed in other cases arising under Delaware's death penalty statute. Id. In conducting that inquiry, we have considered sentences imposed both before and after the 1991 revision to the Delaware death penalty statute since all such cases are pertinent to the proportionality analysis. Wright, 633 A.2d at 342 n. 21. The "universe" of such cases appear in the Appendix to this opinion.

As this Court has noted in prior cases involving proportionality review, "[a] definitive comparison of the `universe' of cases is almost impossible." Red Dog, 616 A.2d at 311 (citing Pennell, 604 A.2d at 1376; Riley, 496 A.2d at 1027; Flamer v. State, Del.Supr., 490 A.2d 104, 144 (1983), cert denied, 464 U.S. 865, 104 S.Ct. 198, 78 L.Ed.2d 173 (1983), and cert. denied, 474 U.S. 865, 106 S.Ct. 185, 88 L.Ed.2d 154 (1985)). We have, however, compared Gattis' sentence with the penalties imposed in all First Degree Murder cases which have included a death penalty hearing. In this comparison we have considered all objective factors such as the gravity of the offense, the circumstances of the crime and the severity of the penalty. See Red Dog, 616 A.2d at 311 (citing Solem v. Helm, 463 U.S. 277, 290-92, 103 S.Ct. 3001, 3010, 77 L.Ed.2d 637 (1983)).

Like others sentenced to death in Delaware, Gattis was found guilty of an unprovoked, cold-blooded murder of a defenseless person. Dawson v. State, Del.Supr., 637 A.2d 57, 68 (1994); Sullivan v. State, Del. Supr., 636 A.2d 931, 950-951 (1994); Wright, 633 A.2d at 343; Red Dog, 616 A.2d at 311, Pennell, 604 A.2d at 1377; DeShields v. State, Del.Supr., 534 A.2d 630, 649 (1987), cert. denied, 486 U.S. 1017, 108 S.Ct. 1754, 100 L.Ed.2d 217 (1988); Riley, 496 A.2d at 1027. As was true in Deputy v. State, Del. Supr., 500 A.2d 581 (1985), cert. denied, 480 U.S. 940, 107 S.Ct. 1589, 94 L.Ed.2d 778 (1987), Flamer, Red Dog, and Dawson, Gattis invaded the victim's home to commit murder.

Gattis argues that the penalty imposed on him is disproportionate to other death sentence cases in that the murder in this case was a "crime of passion" committed by an irrationally jealous lover. Such offenses, it is argued, have resulted in sentences of life imprisonment, not death. We decline to view this murder as simply another domestic violence event deserving of a lesser standard of accountability. The deliberate, cold-blooded killing, in execution style, of a helpless spouse or lover is deserving of society's harshest condemnation. Moreover, the defendant's return to the victim's apartment after being specifically warned by the police not to harass the victim further, belies his claim that he killed under the passion of the moment. Considering all the circumstances of the case, we conclude that Gattis' claim that his sentence is disproportionate in comparison to the "universe" of death penalty cases is without merit.

VIII

This Court has considered all claims of error asserted by Gattis. While the jury selection process did not comply with traditional methods for insuring randomness, we find any error in that connection harmless beyond a reasonable doubt. We find no merit in all other claims of error. We conclude that the death sentence imposed upon Gattis was neither arbitrary nor capricious and is not comparatively disproportionate to the sentences imposed in other First Degree Murder cases that have proceeded to a penalty hearing pursuant to the Delaware capital punishment statute. 11 Del.C. § 4209(g)(2)a. We also conclude that the evidence supports the Superior Court's finding of two statutory aggravating circumstances. 11 Del.C. § 4209(g)(2)b. Accordingly, the underlying convictions and the judgment imposing the death sentence are AFFIRMED.

This matter is remanded to the Superior Court for further proceedings consistent with this opinion, including the setting of a new date of execution. This Court's order of October 30, 1992, staying the execution of Gattis' sentence, shall terminate upon the issuance of this Court's mandate.

APPENDIX

FIRST DEGREE MURDER CASES THAT WENT TO PENALTY HEARINGS

1985 to date

Appendix A to Wright v. State, Del.Supr., 633 A.2d 329, 344 (1993) is incorporated herein by reference, subject to the following changes and additions.

Case Name: David F. Dawson Case No.: IK86-0024; IK87-01-0841, 0843, 0845 County: New Castle (venue changed) Sentence: Death Case Name: Robert A. Gattis* Case No.: IN90-05-1017 thru 1019, 1106, 1107 County: New Castle Sentence: Death — present proceeding Case Name: Jose Rodriguez Case No.: IN93-020-1121 County: New Castle Sentence: Life Imprisonment Case Name: Willie G. Sullivan Case No.: IK-01-0192 thru 0196; IK92-02-0001; IK92-03-0022 County: Kent Sentence: Death Case Name: Jermaine M. Wright Case No.: IN91-04-1047 thru 1953 County: New Castle Sentence: Death

The "universe" of cases is comprised of those First Degree Murder cases which have gone to a penalty hearing and in which the sentence has become final, either without or following a review by this Court. An asterisk marks those capital murder cases which are not included in the defined "universe" of cases.

NOTES

[1] Under the Superior Court jury plan, the master jury wheel, or pool, contains names randomly selected from voter registration lists supplemented with names from the list of licensed drivers maintained by the Delaware Division of Motor Vehicles.

[2] 10 Del.C. § 4512 provides:

(a) Within 7 days after the moving party discovers, or by the exercise of diligence could have discovered, the grounds therefor, and in any event before the jury is sworn to try the case, a party may move to stay the proceedings, and in a criminal case to dismiss the indictment, or for other appropriate relief, on the ground of substantial failure to comply with this chapter in selecting the grand, petit or special jury.

[3] The Superior Court has adopted a jury plan (the "Plan") which essentially mirrors the selection, disqualification and excusal provisions of the Act.

[4] The practice of drawing jurors names at random from the "box" is of ancient origin. See 3 W. Blackstone, COMMENTARIES *358. Woolley characterizes the impanelling of the jury as bearing important relation "to the validity of the judgment resulting from [the trial]." 1 Woolley on Delaware Practice § 639. He describes the procedure as follows:

SEC. 640. Drawing the Jury. The Sheriff, whose duty it is to summon the jury, in the manner provided by the statute, must, within one hour after the opening of the Superior Court, on the first day of every term, return to the court a distinct panel of the persons summoned by him to attend as petit jurors. The prothonotary must then take this panel, and under the direction of the judges presiding, write the name of each petit juror, so returned and impanelled, upon a distinct ballot, all the ballots being of the same color and size. He must then roll up the ballots, all in the same manner, as near as may be, and put them together in a box provided for that purpose.

When a cause is called for trial, "an indifferent person," (always, as a matter of practice, the prothonotary), in open court, draws out twelve of the said ballots successively; and if any of the jurors whose names are so drawn do not appear, or are challenged and set aside, then a further number is drawn until twelve jurors appear, and upon challenge, are allowed.

The twelve persons first drawn, appearing and allowed, their names being marked in the panel, and being sworn or affirmed, constitute the jury to try the cause. (emphasis added).

[5] In the absence of any showing that the jury selection method resulted in the systematic exclusion of a cognizable group, there is no violation of the Sixth Amendment to the United States Constitution. See Duren v. Missouri, 439 U.S. 357, 364, 99 S.Ct. 664, 668, 58 L.Ed.2d 579 (1979).

[6] Standard 3: RANDOM SELECTION PROCEDURES

(a) RANDOM SELECTION PROCEDURES SHOULD BE USED THROUGHOUT THE JUROR SELECTION PROCESS. ANY METHOD MAY BE USED, MANUAL OR AUTOMATED, THAT PROVIDES EACH ELIGIBLE AND AVAILABLE PERSON WITH AN EQUAL PROBABILITY OF SELECTION.

(b) RANDOM SELECTION PROCEDURES SHOULD BE EMPLOYED IN

(i) SELECTING PERSONS TO BE SUMMONED FOR JURY SERVICE;

(ii) ASSIGNING PROSPECTIVE JURORS TO PANELS; AND

(iii) CALLING PROSPECTIVE JURORS FOR VOIR DIRE.

(c) DEPARTURES FROM THE PRINCIPLE OF RANDOM SELECTION ARE APPROPRIATE

(i) TO EXCLUDE PERSONS INELIGIBLE FOR SERVICE IN ACCORDANCE WITH STANDARD 4;

(ii) TO EXCUSE OR DEFER PROSPECTIVE JURORS IN ACCORDANCE WITH STANDARD 6;

(iii) TO REMOVE PROSPECTIVE JURORS FOR CAUSE OR IF CHALLENGED PEREMPTORILY IN ACCORDANCE WITH STANDARDS 8 AND 9; AND

(iv) TO PROVIDE ALL PROSPECTIVE JURORS WITH AN OPPORTUNITY TO BE CALLED FOR JURY SERVICE AND TO BE ASSIGNED TO A PANEL IN ACCORDANCE WITH STANDARD 13.

[7] COURT: Ladies and gentlemen of the jury, as we talked about in the process of getting this jury together, you remember my individual questions, and I'll kind of repeat some of them now or some of the statements I made in connection with those questions.

Every criminal defendant has a constitutional right to remain silent, and that goes across the board. It's a right that is a very important right, and you must understand that any criminal defendant who elects to employ that right, there can be no inference with regard to that election.

So another way of saying that is the fact that a criminal defendant elects to remain silent is not to be considered in any way as an inference that he is guilty of anything, and I so instruct you. You may proceed.

[8] Section 4209(g)(2) provides as follows:

(2) The Supreme Court shall limit its review under this section to the recommendation on and imposition of the penalty of death and shall determine:

a. Whether, considering the totality of evidence in aggravation and mitigation which bears upon the particular circumstances or details of the offense and the character and propensities of the offender, the death penalty was either arbitrarily or capriciously imposed or recommended, or disproportionate to the penalty recommended, or imposed in similar cases arising under this section.

b. Whether the evidence supports the judge's finding of a statutory aggravating circumstance as enumerated in subsection (e) of this section and, where applicable, § 636(a)(2)-(7) of this title.