USF Libraries
USF Digital Collections

An empirical analysis of the role of mitigation in capital sentencing in North Carolina before and after Mckoy v. North ...

MISSING IMAGE

Material Information

Title:
An empirical analysis of the role of mitigation in capital sentencing in North Carolina before and after Mckoy v. North Carolina (1990)
Physical Description:
Book
Language:
English
Creator:
Kremling, Janine
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
Sentencing outcome
Arbitrariness
Capital punishment
1st degree murder
Mitigating circumstances
Dissertations, Academic -- Criminology -- Masters -- USF   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: This study focuses on the influence of mitigating circumstances on the sentencing outcome before and after the McKoy (1990) decision. In McKoy (1990) the Supreme Court decided that the jurors did not have to find mitigating circumstances unanimously. Results are reported based on a sample of North Carolina first-degree murder cases where the state sought the death penalty. Logistic regression is used to determine the importance of mitigating circumstances as predictors of jury decision-making in North Carolina, controlling for the variety of other factors that influence that decision. The descriptive statistics show that the average number of mitigating circumstances submitted and accepted had doubled in the post-McKoy cases. At the same time, the number of aggravating circumstances presented and submitted stayed about the same.
Thesis:
Thesis (MA)--University of South Florida, 2004.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Janine Kremling.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 50 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001681130
oclc - 62792943
usfldc doi - E14-SFE0000615
usfldc handle - e14.615
System ID:
SFS0025305:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001681130
003 fts
005 20060215071302.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 060106s2004 flu sbm s000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000615
035
(OCoLC)62792943
SFE0000615
040
FHM
c FHM
049
FHMM
090
HV6025 (Online)
1 100
Kremling, Janine.
3 245
An empirical analysis of the role of mitigation in capital sentencing in North Carolina before and after Mckoy v. North Carolina (1990)
h [electronic resource] /
by Janine Kremling.
260
[Tampa, Fla.] :
b University of South Florida,
2004.
502
Thesis (MA)--University of South Florida, 2004.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 50 pages.
520
ABSTRACT: This study focuses on the influence of mitigating circumstances on the sentencing outcome before and after the McKoy (1990) decision. In McKoy (1990) the Supreme Court decided that the jurors did not have to find mitigating circumstances unanimously. Results are reported based on a sample of North Carolina first-degree murder cases where the state sought the death penalty. Logistic regression is used to determine the importance of mitigating circumstances as predictors of jury decision-making in North Carolina, controlling for the variety of other factors that influence that decision. The descriptive statistics show that the average number of mitigating circumstances submitted and accepted had doubled in the post-McKoy cases. At the same time, the number of aggravating circumstances presented and submitted stayed about the same.
590
Adviser: Smith, Dwayne.
653
Sentencing outcome.
Arbitrariness.
Capital punishment.
1st degree murder.
Mitigating circumstances.
0 690
Dissertations, Academic
z USF
x Criminology
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.615



PAGE 1

An Empirical Analysis of the Role of Mitiga tion in Capital Sentencing in North Carolina Before and After McKoy v. North Carolina (1990) by Janine Kremling A thesis submitted in partial fulfillment of the requirements for the degree of Master of Arts Department of Criminology College of Arts and Sciences University of South Florida Major Professor: M. Dwayne Smith, Ph.D. John Cochran, Ph.D. Kathleen Heide, Ph.D. Tom Mieczkowski, Ph.D. Date of Approval: July 9, 2004 Key words: Mitigating Circumstances, 1st Degree Murder, Capital Punishment, Arbitrariness, Sentencing Outcome Copyright 2004, Janine Kremling

PAGE 2

i Table of Contents List of Tables................................................................................................................. ........ii Abstract....................................................................................................................... ..........iii Chapter One Introduction....................................................................................................... 1 Chapter Two Legal History of Capital Punishment from Furman (1972) to McKoy (1990)......................................................................................................................... ..........6 Chapter Three Research on the Administra tion of the Death Penalty in the United States in the postGregg era...............................................................................................14 Research on the Issue of Arbitr ariness: Aggravating Factors ..................................14 Research on the Issue of Arbitrariness: Both Aggravating and Mitigating Factors.................................................................................................................. 16 Chapter Four Method and Analytical Procedures.................................................................21 Data.......................................................................................................................... .21 Statistical Analysis....................................................................................................27 Chapter Five Results........................................................................................................... ..29 Chapter Six Discussion......................................................................................................... 37 References..................................................................................................................... ........44

PAGE 3

ii Lists of Tables Table 1 Number of Aggravating and Mi tigating Circumstances Presented and Accepted Preand PostMcKoy .............................................................................30 Table 2 Logistic Regression Model....................................................................................33

PAGE 4

iii An Empirical Analysis of the Role of Mitig ation in Capital Sentencing in North Carolina Before and After McKoy v. North Carolina (1990) Janine Kremling ABSTRACT This study focuses on the influence of m itigating circumstances on the sentencing outcome before and after the McKoy (1990) decision. In McKoy (1990) the Supreme Court decided that the jurors did not have to fi nd mitigating circumstances unanimously. Results are reported based on a sample of North Carolin a first-degree murder cases where the state sought the death penalty. Logistic regression is used to determine the importance of mitigating circumstances as predictors of jury decision-making in North Carolina, controlling for the variety of other f actors that influence that decision. The descriptive statistics show that the average number of mitigating circumstances submitted and accepted had doubled in the postMcKoy cases. At the same time, the number of aggravating circumstances presen ted and submitted stayed about the same. The analysis then moved to the consideration of the impact of mitigating circumstances, and whether there had been a change between the tw o eras. Separate logis tic regression analysis revealed that there had indeed been a shift in the effects of aggravation and mitigation, but no in the manner that might have been anticipated, Specifically, in the postMcKoy era, mitigating circumstances had a diminished impact on the probability of a death sentence while, conversely, aggravating circumst ances carried an increased impact.

PAGE 5

1 Chapter One Introduction and Statement of the Research Issue Introduction In 1972, the United States Supreme Court (h ereafter referred to as the “Court”) decided in the historical Furman v. Georgia (1972) case that capit al punishment statutes allowing unguided discretion were unconstituti onal because they produced arbitrary and capricious sentencing outcomes. Four years la ter, the Court ruled that mandatory death penalty statutes were also unconstitutional b ecause they did not allow for individualized sentencing ( Woodson v. North Carolina 1976). Since then, 36 states have incorporated systems of capital punishment that employ different versions of a guided discretion model that was approved in Gregg v. Georgia (1976) and, in somewhat different form, in Proffitt v. Florida (1976). Texas and Oregon utilize ye t another guided discretion model approved in Jurek v. Texas (1976). From 1976 until today, the Court has focuse d on two principles regarding capital punishment that are part of the Eighth Amen dment prohibition against cruel and unusual punishment. First, discretion must be guided so that the group of death-eligible persons is narrowed. Second, the sentencer must consid er all mitigating evidence and allow for individualized sentencing that hypothetically takes into account the full context in which the crime occurred (Kirchmeier, 1998, p. 3). The general trend of Court decisions has been to systematically increase the disc retion of jurors regarding mitigating and aggravating circumstances, both of which are key elements in the guided discretion models employed in capital murder cases (i.e ., those eligible for th e death penalty) during the penalty phase of the trial. Briefly, aggravating circumstances are situations established through state legisl ation by which a murder becomes eligible for the death penalty, and states vary considerably in th e number and types of situations that are statutorily defined (Coyne and Entzeroth, 2001, p. 329). In contrast, mitigating factors are circumstances presented by defendants, the purpos e of which is to argue that they should

PAGE 6

2 be spared the death penalty. In most states, if one or more statutory aggravators are found to exist, they must be “weighed” against mitigating circumstances. However, the Court has not specified any rules on how to weigh aggravating against mitigating factors, so jurors (or, in some states, judge s) are left with rather broad discretion in their decision to give the defendant a life or death sentence. One of these Court decisions was McKoy v. North Carolina (1990) whereby the Court ruled that jurors do not have to be unanimous in deciding whether any specific mitigating factor should be taken in account during punishment deliberations. This ruling was particularly important to the North Ca rolina system of capital punishment because juries have the final senten cing authority. The result of McKoy was to overturn a number of death sentences in North Carolina and, of longer-term consequence, to mandate a new procedure in which jurors respond to mitiga ting factors presented by the defense. The essence of this change was that jurors no l onger had to agree unanimously that a specific mitigating factor existed in responding “yes ” on an “Issues and Recommendation” form that juries are required to complete as a reco rd of their deliberations regarding sentencing. The revised procedure that remains in use toda y simply requires that one or more jurors believes that a mitigating factor exists for the jury to answer “yes” on the form. In effect, the meaning of a “yes” response has changed; in its extreme ramification, a lone juror rejecting a mitigator prior to McKoy would have resulted in a “no” being recorded, while after McKoy a lone juror accepting a mitigator would mandate a “yes” being recorded. Background to the McKoy Case A bit of background on the McKoy case may be helpful in understanding how the case unfolded. Dock McKoy, Jr. (also known by several aliases), age 65, was a resident living outside of Wadesboro in Anson County, North Caro lina. On the afternoon of December 22, 1984, the Anson County Sheriff’ s office received reports that McKoy was discharging a firearm outside hi s house in an area that was dangerously close to other residents. Two deputies responded, conferred wi th the person calling in the complaint, and then talked with McKoy. Noting that he was highly intoxicated, they asked him to stop the shooting. A short time later, two ot her deputies were sent back to McKoy’s house because of complaints that the shoo ting had resumed. When the officers arrived, McKoy retreated inside his house, rejected their requests that he come outside, and

PAGE 7

3 threatened to kill them if they did not leave. Other officers began to arrive, and one deputy walked to the back of McKoy’s house to see if entr y could be gained. Unexpectedly, McKoy pushed open his screen door and fired a shot at Deputy Kress Horne, who was braced against a patrol car. The shot hit Deputy Horne in the face, a wound from which he died later that even ing. A brief gun battle broke out between McKoy and the officers on site, finally ending when a wounded McKoy surrendered. Because of intense publicity in the relatively small county, McKoy’s trial was moved to neighboring Stanly County. The stat e announced that it w ould seek the death penalty, and following McKoy’s conviction for first degree murder, presented two aggravating circumstances, both of which th e jury accepted: (1) that McKoy had been previously convicted of a felony involving vi olence to the person (he had been convicted of two counts of assault with a deadly weapon in 1977, crimes for which he served five years in prison), and (2) that the murder was against a law enforcement officer performing his professional duties. The defense presented seven mitigators, including a “catch-all” factor that allowed the jury to c onsider any other mitigating factor it believed warranted consideration (as mandated in Locket v. Ohio [1978]). A unanimous vote was required to indicate acceptance of each mitigator resulting in the jury indicating “yes” to two mitigators, “no” to four others presented by the defense, and indicating in response to the “catch-all” that they had no other mitigat ors they wished to consider. On August 8, 1985, the jury returned a sentence of death. In their appeal to the North Carolina S upreme Court, McKoy’s attorneys raised a number of issues, many of which centered on his intoxicated state at the time of the offense. One appeal, however, questioned wh ether requiring unanimity from juries in accepting mitigating factors led to jurors believin g that they could not consider that factor when they made their individual decisions re garding the sentence. In other words, the question raised was whether the unanimity requirement could l ead to a single dissenting juror neutralizing a strong feeling among othe rs that a specific mitigator (especially McKoy’s intoxication) should be taken into account. The charge was that jurors might interpret this guideline to mean that they could not ta ke this factor into account when making their individual decision for a life or death sentence. The North Carolina justices were not persuaded by this, or any of the ot her arguments presente d on appeal, and thus

PAGE 8

4 upheld McKoy’s conviction and death sentence ( State of North Carolina v. Dock McKoy, Jr. 1988). McKoy’s attorneys continued to press his appeals, and eventually got a hearing before the U.S. Supreme Court. There, basi ng their argument on the recent Court decision in Mills v. Maryland (1988), they pressed the argument that their client might have received a death sentence because some juro rs believed that they could not consider issues presented in the four mitigators that were not unanimously found by the jury. The state’s attorneys argued that North Carolina juries were not compelled to assess a death sentence even in the absence of finding any mitigators, and unlike the Maryland procedures struck down in Mills the jury instructions made this clear. Ultimately, by a 6-3 vote, the Court’s majority justices we re persuaded by the arguments on behalf of McKoy. Although the Court upheld his convict ion, it vacated McKoy’s death sentence and ordered a new sent encing phase trial ( McKoy v. North Carolina 1990). As an ironic ending to the case, Dock McKoy was declared incompetent to be retried because of his deteriorated mental condition following longterm alcohol abuse. Now 84 years old, he remains incarcerated as of this writing. The North Carolina Department of Correction website indicates that his current st atus is no longer a matter of public record, suggesting that he is confined to the mentally disabled ward of the state’s main prison in Raleigh, North Carolina. Statement of the Research Issue As will be discussed in the next chapter, a variety of Court decisions have demanded a somewhat standardized procedur e of capital punishment that nevertheless allows for individualized sentencing decisi ons. That is, while seeking to eliminate arbitrariness (meant as the absence of legally relevant factors as predictors of death sentences), the Court recognizes that seemi ngly similar cases may have characteristics that will cause jurors (and, in some states, judges) to vary in the sentences they assign. This thesis addresses th e question of whether the McKoy (1990) decision has influenced the way in which mitigating factors are c onsidered by North Carolina capital juries. Specifically, a large sample of capital cases in North Carolina will be analyzed to determine whether there is a difference before and after the McKoy decision in the role of

PAGE 9

5 mitigating factors in predicting juries’ sent encing recommendations. Three possibilities exist: (1) There will be no difference in the two periods; (2) Because unanimity was required preMcKoy mitigating factors will have a stronger influence; or (3) If jurors were indeed dissuaded from considering mitigating factors unless there was juror unanimity, freeing them from this constraint might result in mitigating factors very significantly related to se ntencing recommendations. Organization of the Study To explore the issue of whether the McKoy decision affected the role of mitigation in North Carolina capital sentenci ng patterns, this study is organized in the following manner. Chapter 2 provides an ev aluation of the decisions of the Court regarding its jurisprudence on how aggravating and mitigating circumstances are to be taken into consideration by the sentencer in order to avoid arbitr ary death sentences. Included in this discussion are changes in cas e law that have influenced North Carolina’s death penalty system towards increased juror discretion. Against this background, I consider the extent to which the Court’s postFurman jurisprudence may be inconsistent with its decision in Furman and Gregg that capital sentencing should be standardized to avoid arbitrary and caprici ous sentencing outcomes. In Chapter 3, a review of the empirical l iterature is presented regarding issues of arbitrariness in capital puni shment decision making after Furman (1972). During the past 30 years, researchers have conducted extensive research that addre sses whether continued arbitrariness in the administration of the de ath penalty exists. The general findings from this literature are discussed, with a speci al focus on the roles of aggravating and mitigating circumstances as statistical predictors of sentencing outcomes. Chapter 4 explains the methodology and st atistical analysis of the current study, while the results of the study’s analyses are pr esented in Chapter 5. Chapter 6 contains a discussion of the findings and their impli cations, and concludes the thesis with suggestions for further research in this area.

PAGE 10

6 Chapter Two Legal History of Capital Punishment from Furman (1972) to McKoy (1990) In McGautha v. California (1971), the Court consider ed for the first time the constitutionality of unguided discretion in capit al sentencing. In a 63 vote, the justices ruled that unguided discretion models do not violate the Fourteenth Amendment due process clause. However, only a year later, th e Court appeared to reverse itself in the case of Furman v. Georgia (1972) when a deeply divided cour t ruled 5-4 that the then-current practice of capital punishment was unconsti tutional. Even though the five majority justices issued separate opi nions, a common thread across th eir discussions was that the death penalty was administered in an ar bitrary and capricious manner due to the unfettered discretion granted to capital ju ries. They emphasized an underlying premise that “death [sentencing] is different.” They believed that the arbitrariness and capriciousness that seemed to characterize Georgia’s death sentencing was unacceptable. Consequently, the prevailing opi nion was that the decisions of the sentencer must be guided by clear rules so as to create a sta ndardized sentencing model. The primary requirements for those states wishing to re sume capital punishment were to “channel the sentencer’s discretion by clea r and objective standards that provide specific and detailed guidance and that make rationally reviewab le the process for imposing a sentence of death“ ( Furman v. Georgia p. 428) and to narrow the class of offenders that are death eligible. In regards to narro wing the group of offenders elig ible for death sentencing, the justices wanted to overcome the problem of “overinclusion,” meaning the failure to differentiate those whose crimes most merite d the death penalty from those whose crimes were of a less egregious natu re (Striker and Steiker, 2003). Following the Furman (1972) decision, most state legislatures rewrote their capital punishment procedures. In 1976, severa l different approaches came before the Supreme Court. In Gregg v. Georgia (1976), Jurek v. Texas (1976), and Proffitt v.

PAGE 11

7 Florida (1976) the Court approved, by 7-2 votes, three separate guided discretion models. All three employed a bifurcat ed trial procedure whereby separate guilt and sentencing phases of the trial were conducted. In the penalty phase, aggravating and mitigating circumstances were to be considered in or der to guide the sent encer’s discretion and thereby eliminate arbitrary and capricious sentencing decisions.1 These models also provided guidance in narrowing the class of death-eligible offenses by limiting death eligibility to first-degree murders with at least one aggravating circumstance, and required a proportionality review to dis cern whether a given death sentence was proportionate compared to other similar case s. In approving these models, the majority justices were optimistic that the concern for fairness expressed in Furman (1972) four years earlier would be adequately addressed. While the guided discretion models were declared constitutional, mandatory death penalty models introduced by Woodson v. North Carolina (1976) and Roberts v. Louisiana (1976) were rejected and said to viol ate the Eighth Amendment. The reasoning of the Court was that mandatory schemes pr ovided no guidance to the sentencer, did not address the problem of overinc lusion, and did not allow fo r the consideration of the character of the crime and the defendant. The Court was also concerned with jury nullification, which refers to the reluctance of juries to convict defendants whom they believe to be guilty, but not deservin g of death (Coyne and Entzeroth, 2001). Interestingly, the decisions of the Court in Gregg, Proffitt, and Jurek and those in Woodson and Roberts are somewhat contradictory. On the one hand, the Court required standardized sentencing in Gregg Proffitt and Jurek to address the concerns raised in Furman ; on the other hand, th e Court rejected Woodson and Roberts because the statutes did not allow for more individualized sentenci ng. Critics have said that a precedent was set in favor of individuali zed sentencing, and the Court’ s focus shifted from guided discretion to simply narrowing the class of death-eligible defendants (Kirchmeier, 1998). In most states, including North Carolina, as long as jurors find at least one statutory aggravating factor to exist, th ey have broad discretion to retu rn either a life or a death sentence. Subsequent decisions of the Court make this contradiction more evident. Two years after Gregg and Woodson Lockett v. Ohio (1978) increased the discretion of the

PAGE 12

8 sentencer when the Court ruled that any mitigating circumstance supported by the evidence can be submitted by the defense, not just those specified in the state’s capital punishment statutes. This decision was confirmed and expanded in Eddings v. Oklahoma (1982) when the Court held that the sentence r cannot refuse to ta ke into account any mitigating factor supported by the evidence. The rulings in Lockett and Eddings drew on Woodson and Roberts in specifying that sentencing must be individualized and related to the culpability of the offender, but appeared contrary to the Furma n and Gregg decisions that standardized discretion models must channel discretion. Sundby (1991) argues that the ultimate impact of the Lockett decision, however unintended, was to provide capital juries with an increased opportunity to ma ke the kinds of arbitrary and capricious sentencing decisions that were objected to in Furman In three decisions in 1983, the Court fu rther expanded the discretion of capital juries regarding their consideration of a ggravating circumstances. The decisions in Stephens v. Zant (1983), Barclay v. Florida (1983), and California v. Ramos (1983), provided juries with more flexibility to retu rn a death sentence by expanding the scope of aggravating factors that could be considered. In Zant the Court decided that a death sentence is constitutional even if one of the several aggravating circumstances found by the jury is later held to be invalid (but, at least one must be valid). The Court also allowed the consideration of non-statutory aggravat ing circumstances if the jury unanimously found at least one statutory aggr avating factor to exist. Ho wever, the decision’s impact was limited to states where the jury did not have to “weigh” aggravating against mitigating factors. In Barclay the Court closed the hole that it had left open in Zant when it stated that if there are other valid aggravating factor s, one invalid aggravating circumstance does not support the claim of a ne w sentencing hearing even when the jury or judge is supposed to weigh the aggrav ating circumstances against the mitigating circumstances. Further expanding discretion, the Court ruled in Ramos that jurors are allowed to consider unrelated aggravating factors, such as a possibility of a commutation from a life sentence without the possibility of parole to a life sentence with the possibility of parole, in their final sentencing decisions. Critics such as Justice Thurgood Marshall have said that the decisions in Locket/ Eddings and Zant/Barclay/Ramos served to undermine a fair and consistent sentencing

PAGE 13

9 procedure and violated the Furman requirement of guided di scretion. Justice Marshall expressed this concern in his dissent from the Zant decision by saying th at "[o]nce [the threshold] finding [of one statut ory aggravating circumstance] is made, the jurors can be left completely at large, with nothing to guide them but their wh ims and prejudices" ( Zant v. Stephens 1983, p. 2760). The growing discretion that alarmed Justic e Marshall was widened even further in Mills v. Maryland (1988) when the Court removed th e barrier of jury unanimity in considering mitigating circumstances. Mills reaffirmed and expanded Lockett by ruling that the vote of a single juror is sufficient to find the existence of a mitigating factor. More precisely, that juror cannot be prevente d from taking that mitigator into account when making the sentencing decision, even if others on the jury elect not to do so. Importantly for the present research, the Mills (1988) decision was the basis for a change in North Carolina’s death penalty stat ute. Two years later, the appeal of Dock McKoy was ruled on. A key element of the appeal had been to use Mills to challenge North Carolina’s requirement that each mitigating factor being considered by the jury must be approved unanimously. Attorneys for the state argued that what distinguished North Carolina from Maryland was that the Nort h Carolina statute clear ly stated that the jury was not required to impose a death se ntence when no mitigating evidence was found to exist. Contrary to North Carolina, the se ntencing instructions in Maryland were so ambiguous that they may have misled the jury to believe that finding no mitigators mandated a death sentence. In a 6-3 decision, the justices re jected the state’s argument in McKoy v. North Carolina (1990) and ruled that “North Carolina ’s unanimity requirement impermissibly limits jurors’ consideration of mitigating eviden ce and hence is contrary to this Court’s decision in Mills ” ( McKoy v. North Carolina 1990, p. 2). The majority of justices argued that it was unconstitutional for one holdout juror to prevent th e other jurors from considering mitigating evidence, especially since this may inadvertently slant the subsequent decision toward a death. Justice Antoine Scalia delivered the di ssenting opinion of the court, one which Chief Justice William Rehnquist and Justice Sandra O’Connor joined. These justices agreed that the State must gui de the sentencers’ discretion and provide the opportunity to

PAGE 14

10 rationally assess their decision in order to retu rn a death or a life sentence. However, they believed that “there is little guidance in a sy stem that requires each individual juror to bring to the ultimate decision his own idios yncratic notion of what facts are mitigating, untempered by the discipline of grou p deliberation and agreement” ( McKoy v. North Carolina 1990, p. 21). They believed that a capital jury is supposed to return a final sentence as a “body” that deli berates and decides together th e appropriate sentence. Their dissent expressed concern that McKoy will separate the jurors and force them to reach an isolated decision, an outcome not supported by Lockett (1978), Eddings (1982), or Mills (1988). In sum, the McKoy decision, according to the dissenting justices, violated the rules of guided discretion, thereby u ndermining “sound jury deliberations” ( McKoy v. North Carolina 1990, p. 23). The pragmatic effect for North Carolina wa s to immediately alter the wording of Issue Two in the Issues and Recommendation Sheet that juries must complete as a record of their deliberations. The wording was changed to (and remains) the following instruction: “Do you find from the evidence the existence of one or more of the following mitigating circumstances?” followed by “Befor e you answer Issue Two, consider each of the following mitigating circumstances. In the space after each mitigating circumstance, write ‘yes,’ if one or more of you [my emphasis] finds that circumstance by a preponderance of the evidence. Write ‘ no’ if none of you find that mitigating circumstance.” This replaced the wording that was in effect: “Do you unanimously find from the evidence the existence of one or more of the following mitigating circumstances?” followed by “Before you answ er Issue Two, consider each of the following mitigating circumstances. In the space after each mitigating circumstance, write ‘yes,’ if you unanimously find that mitigating circumstances by the preponderance of the evidence. Write ‘no,’ if you do not unanimously find that mitigating circumstance by a preponderance of the evidence.”2 As a result of the McKoy ruling, all individuals w ho received a death sentence under the unanimity requirement had their death sentences rema nded. However, those cases where the jury unanimously accepted all submitted mitigators were not considered to have the basis for an appeal. A review of death se ntences prior to McKoy revealed that 41 individuals whose cases were originally upheld by the North Carolina Supreme Court

PAGE 15

11 had their sentences remanded for retrial (Rhee, 1993). Of these, my analysis shows that 24 received death sentences in subsequent re-trials of the sentencing phase using the new guidelines. As discussed at the beginning of this chapter, the Court’s concern in Furman was that the death penalty was imposed in an arbitrary and capricious manner. With its decisions following Gregg (1976), the Court may have opened the doors once again for unfettered discretion and invited exactly the arbitrary decision making that was the focal concern of Furman In reality, subsequent rulings have created a situation where there is not much guidance left for jurors, because th ey can consider anything in mitigation and anything in aggravation, as long as one statutory aggrava ting factor is found exist. What has emerged in practice from this situation? A large body of empirical literature has been devoted to patterns of capital sentencing following the resumption of executions in 1977. The next chapter review s this literature to determine whether arbitrariness and capriciousness remain as pa rt of the contemporary character of death sentencing in the United States. Endnotes 1The differences in the models are as follows: ( Gregg v. Georgia 428 U.S. 153, p. 1) The new Georgi a “statutory provisions with regard to imposition of the death penalty for the crime of murder and other offenses, (1) guilt or innocence is determined, either by a jury or the trial judge, in the first stage of a bifurcated trial, with the judge being required to charge the ju ry as to any lesser included offenses when supported by the evidence, (2) after a verdict, finding, or plea of guilty, a presentence hearing is conducte d, where the jury (or judge in a case tried without a jury) hears argument and additional evidence in mitigation or aggravation of punishment, (3) at least one of ten aggravating circumstances spec ified in the statutes must be found to exist beyond a reasonable doubt, and must be designated in writing, before the jury (or judge) may impose the death sentence on a defendant convi cted of murder, the trial judge in jury cases being bound by the jury's recommended se ntence, (4) on automatic appeal of a death sentence, the Supreme Court of Georgi a must determine whether the sentence was imposed under the influence of passion, prejudi ce, or any other arbi trary factor, whether the evidence supported the finding of a statut ory aggravating circum stance, and whether

PAGE 16

12 the death sentence was excessive or dispropor tionate to the penalty imposed in similar cases, considering both the crime and the defendant, and (5) if a death sentence is affirmed, the decision of the Georgia Suprem e Court must include reference to similar cases that the court considered“. ( Jurek v. Texas 428 U.S. 262, p. 1) Under the new Texas statutes, (1) capital homicides are limited to intentional and knowing murders committed in the five specified situations of murder of a peace officer or fireman, mu rder committed in the course of kidnapping, burglary, robbery, forcible rape, or ar son, murder committed while escaping or attempting to escape from a penal institution, murder committed for remuneration, and murder committed by a prison inmate when the victim is a prison employee; (2) if a defendant is convicted of a capital offense, a separate presentence hearing must be held before the jury, where any relevant eviden ce may be introduced and arguments may be presented for or against the death sentence; (3) the jury must answer the questions (a) whether the defendant's conduct that cause d the death was committed deliberately and with the reasonable expectation that the death of the deceased or anot her would result, (b) whether there is a probability that the defendant would commit criminal acts of violence that would constitute a continuing threat to society, and (c) if raised by the evidence, whether the defendant's conduct in killing the deceased was unreasonable in response to the provocation, if any, by the deceased; (4) if the jury fi nds that the state has proved beyond a reasonable doubt that the answer to each of the pertinent questions is yes, then the death sentence is imposed, but if the jury finds that the answer to any question is no, then a sentence of life imprisonment results; and (5) death sentences are given expedited review on appeal”. ( Proffitt v. Florida 428 U.S. 242, p. 1) “Under the new Florida statutes, (1) if a defendant is found guilty of first-degree murder a separate presentence hearing is held before the jury, where arguments may be presented and where any evidence deemed relevant to sentencing may be admitted and must include matters relating to eight aggravating and seven mitigating circumstances specified in the statutes, (2) the jury is directed to weigh such circumst ances and return an advisory verdict as to the sentence, to be determined by a majority vote, (3) the actu al sentence is determined by the trial judge, who is also directed to weigh the statutory aggravating and mitigating circumstances, (4)

PAGE 17

13 if a death sentence is imposed, the trial court mu st set forth in writing its fact findings that sufficient statutory aggravating circumstances exist and are not out weighed by statutory mitigating circumstances, and (5) a death sentence is automatically reviewed by the Supreme Court of Florida, which considers its function to be to guarantee that the aggravating and mitigating reasons present in one case will reach a similar result to that reached under similar circumstances in another case.” As a note, the capital punishment st atutes ultimately adopted by the North Carolina legislature seems to be consistent with the Florida statute developed in Profitt However, one major exception exists in that th at the judge has no authority to overturn a jury decision in North Carolina; in essence, the jury is the sentencer in North Carolina while final authority rests with the trial judge in Florida. 2The form in place at the time came about through a North Carolina Supreme Court decision, State of North Carolina v. Clinton Rondale Kirkley (1983), in which a revision of the original postGregg Issues and Recommendation form was ordered. The essence of the revision was that the state, in Issue Two, had been submitting mitigating factors as a group. The jury ha d to indicate whether or not it unanimously found one or more of them to exist, but did not require the jury to indicate which mitigators it did, or did not, approve. The Kirkley decision led to a revised fo rm that required unanimous approval for each separate mitigator. This revised form was the object of appeal in the McKoy decision.

PAGE 18

14 Chapter Three Research on the Administrati on of the Death Penalty in the United States in the PostGregg Era Researchers have conducted many studies since the re-instatement of capital punishment in Gregg v. Georgia (1976). The major part of research has focused on whether the issues raised in Furman v. Georgia (1972) have been addressed, namely the presence of arbitrariness (the absence of lega l factors as predictors of death sentencing) and discrimination (whether non-legal factors are found as predictors). The United States General Accounting Office (GAO) analyzed 28 studies that had been conducted in the first decade postGregg and attempted to summarize the findings emanating from this body of research. The ensuing report ( GAO, 1990) concluded that even though the studies varied in their method and overall qua lity, a consistent finding was that factors related to aggravation were significant predicto rs of death senten ces, although their relative weights varied consid erably among the states that were sources of the data. Interestingly, the report was not as explicit in its assessment of mitigation in predicting death sentencing. What may have contributed to this absence is that a number of the studies reviewed in the GAO repor t did not include some aspect of mitigation as part of the analysis.1 The following sections concentrate on subs equent literature not covered in the GAO (1990) report because of the issue of arbi trariness, specifical ly how a change in North Carolina juries’ consideration of mitigation may have changed its role in predicting death sentences, The discussion is divided into two sections, one concentrating on research that included mitigating factors in the analysis and the other on those studies which considered both aggravating and mitigatin g factors as legal factors in predicting death sentences.

PAGE 19

15 Research on the Issue of Arb itrariness: Aggravating Factors Almost all death penalty studies since the GAO (1990) report have examined the impact of aggravating circumstances on sent encing outcomes, treating them as legally relevant variables that should be expected to influence thos e decisions. Some researchers have included separate aggravators accepted by juries as dummy variables while others have employed aggravating factors as a summ ed variable (the number of aggravating circumstances found by the jury). Also, a few researchers have created a so-called culpability index as a method for grouping cases on a seriousness scale. The researchers then compared the sentencing outcomes of “sim ilar” cases (proportiona lity review). The vast majority of recent studies including aggr avating factors as dummy variables find that certain single aggravating circumstances can significantly influence the outcome of capital trials. Klein and Rolph’s (1991) study in California showed that murders involving the victim being subjected to sodom y led to a death sentence in 92% of the cases and murders that included torture led to a death sentence in 100% of the cases. Other aggravating circumstances that signifi cantly increased the pr obability of a death sentence were murder to avoid arrest, murder for pecuniary gain, and prior prison record. Keil and Vito (1995), analyzing death se ntencing in Kentucky, included four legally relevant variables (felony circumst ances, multiple victims, silenced victim [murder to avoid prosecution]) in their analysis. They also included a variable for “case seriousness” that indicated wh ether more than one aggravating circumstance existed. The results indicated that multiple victims, silenced victim, and case seriousness all significantly influenced penalty trial outcome s towards a death sentence. However, the authors found evidence that juries treate d aggravating circumstances differently depending on the race of the defendant and th e race of the victim, namely that black defendants who killed a white victim were mo re likely to receive a death sentence. Brock, Cohen, and Sorensen (2000) found in Texas that felony circumstances (the murder was committed in the course of another crime) were the best predictor of a death sentence. Among the felony circumstances, sexual assault was overrepresented among death sentences. The probability of a death sentence was also higher for multiple offenders, multiple victims, murders by st rangers, and murders by men. Additionally, physical weakness of a victim (females, children, elderly) al so increased the probability

PAGE 20

16 of a death sentence. Further, Brock et al. cr eated a scale of case se riousness that included felonies by strangers, death by weapons other than guns, multiple victims, and helplessness of the victim when murdered. The scale was created by assigning scores based on the existence or absence on these four factors. The authors were able to show that with an increasing level of case seriousne ss the probability of a death sentence also increased. In other work, a study by Holcom b and Williams (2001) in Ohio found that felony circumstances (the murder committed in the course of another crime) were the best predictor of a death sentence, along with multiple victims being murdered. Similarly, Pierce and Radelet’s (2002) st udy of Illinois showed that prior murder record, felony circumstances, and multiple victims were significantly related to a death sentence. Even though the studies discussed here employed very different methodologies and were based on research in different states a general finding to em erge is that felony circumstances, prior violent record, and multiple victims were consistently and significantly related to death sentencing. Howe ver, based on the model to emerge from the Gregg decision and subsequent U.S. Supreme Court rulings, mitigating circumstances must also be considered as legal factors that can legi timately influence sentencing decisions. Therefore, the next section will revi ew studies that have assessed the influence of both mitigating as well as aggravating ci rcumstances in determining predictors of death sentencing. Research on the Issue of Arbitrariness: Both Aggrav ating and Mitigating Factors In 1990, Baldus, Woodworth, and Pulask i published a now-famous study about the administration of the death penalty in Georgia.2 In their Procedural Reform Study (PRS), Baldus and his colleagues compared the administration of the death penalty in Georgia before and after Furman. They found that the probability of a death sentence increased with an increasing number of accepted aggravating circumstances. However, the death sentencing rate for the most aggravated cases was only .63. The authors concluded that “the st atutorily designated aggravating circumstances in Georgia’s postFurman law do not serve in practice to distingui sh murder cases in which death sentences are routinely imposed from those that normally result in a life sentence” (Baldus et al., 1990, p. 97).

PAGE 21

17 Additionally, Baldus et al. (1990) di scovered that single statutory and nonstatutory aggravating and mitigating circum stances had a significant influence on the sentencing outcome.4 An important component of thei r analyses of whether Georgia’s death penalty operates arbitrarily was the construction of a “culpability scale” that included seventeen legitimate factors (aggr avating and mitigating circumstances). The culpability index consisted of five levels: 1) “very mitigated,” 2) “some mitigation,” 3) “neither aggravated nor mitigated,” 4) “som e aggravation,” and 5) “very aggravated.” The hypothesis was that “in a random or highl y capricious death-sentencing system, there would be little or no relationship between th e perceived culpability of those sentenced and the sentences they received” (p. 59). The results suggested that the level of culpability was related to the sentencing outcome, namely the higher the level of culpability, the greater the proba bility of a death sentence. Ho wever, in the midrange of aggravation, the death sentencing rate vari ed between .08 and .80. This finding indicated to the authors that the death penalty in Georgia in the postFurman era was not reserved for the most aggravated cases and that the death penalty was not being applied consistently. Baldus et al.’s (1990) anal ysis shows that even though the administration of the death penalty may be less arbitrary a nd more evenhandedly applied in the postFurman era than in the pre-F urman era, legitimate case characteri stics do not fully explain the sentencing outcome and death sentences ar e only applied evenhandedly in 50 to 60 percent of all cases.4 In sum, the results imply that the Georgia death penalty system still operates in a somewhat arbi trary and excessive manner. In 1998, Baldus, Woodworth, Zuckerman, Weiner, and Broffitt published the results of another study, this one from the c ity of Philadelphia. The researchers used different methodologies to measure defendant cu lpability and their re sults suggested that aggravating factors significantly influenced juries’ decisions. In contrast, mitigating factors appeared to have little influence. In further wor k, a study of Nebraska capital sentencing by Baldus, Woodworth, Grosso, and Christ (2002) found support for their Philadelphia findings. Whereas six of the st atutory aggravating circumstances were significantly related to sentencing outcomes, none of the statutory mitigating factors was significant.3 When the jury found one aggravating ci rcumstance to exist, mitigation had a

PAGE 22

18 slight effect. When the jury found two aggrav ating factors, the probability of a death sentence decreased when the number of mitigating factors increased. When a jury found three or more aggravating factors, mitigating factors did not matter. In sum, the influence of the number of mitigating factors on the sentencing outcome varied depending on the number of aggravating factors. Overall, aggravating factors had a much stronger influence on the sentencing outcome than mitigating circumstances. Unah and Boger (2001) conducted a later st udy in North Carolina that began with all murders in North Carolina during 1993 – 1997,5 analyzing the likelihood of a case receiving a death sentence at several levels of the criminal justice process (e.g., decision to charge first degree murder, decision to s eek the death penalty, etc.). For those cases that were tried capitally, the aggravating circumstances of prior violent felony, felony circumstances, killing of a law enforcement officer, an especially heinous killing, the murder posing a great risk to other persons, a nd violence against another victim were all found to be significantly rela ted to a death sentence. C onversely, acceptance of two statutory mitigating factors, “capacity to appreciate criminality impaired” and “defendant’s age,” were significantly related to the defendant receiv ing a life sentence. The authors also created a variable “nonsta tutory mitigating factors” and a variable “nonstatutory aggravating circumstance of vi ctim”. Both variable s were significantly related to the sentenci ng outcome. However, the authors provided no information as to which factors were counted as non-statutory aggravating and mitigating circumstances. More recently, Lenza, Keys, and Guess ( 2003) analyzed data from Missouri and found that prior convictions, an aggravating fa ctor, had a significant effect on defendants receiving a death sentence. Age of the defe ndant (younger than 21), a mitigating factor, did not significantly decrease the probab ility of receiving a death sentence. Conversely, defendants younger than 30 years were more lik ely to be sentenced to death. The authors speculated that jurors saw young people as ha ving less social value and therefore being more “executable” than older offenders. Paternoster (2003) concentrated mainly on the influence of geography and race of victim, as well as race of defendant, when he examined the application of the death penalty in Maryland. Paternoster used aggr avating and mitigating circumstances as control variables to determine whether the capital punishment system of Maryland shows

PAGE 23

19 discriminatory patterns. Numerous aggrav ating and mitigating circumstances had a significant influence on the sentencing outcome when estimating the effects of geography (counties). The influential statutory and non-st atutory aggravating factors are “multiple victims,” “victim sexually abused,” “victim killed execution style,” and “prior felony conviction.” The influential statutory and nonstatutory mitigating circumstances were “defendant physically abused as a child” and “defendant made full confession to aggravating factors.” These aggravating and mitigating circumstances were also significantly related to the sent encing outcome when the eff ects of the defendant’s race were assessed. Additionally, the statutory and nonstatutory aggravating factors of “defendant implicated in other killings,” “def endant forced his way into the place of the victim,” “victim suffered multiple trauma s,” “victim killed execution style.” and “defendant tried to hide or dispose body” and the statutory and nonstatutory mitigating circumstances “defendant had history of dr ug abuse,” “defendant sexually abused as a child” and “defendant confessed to aggravating circumstances” were significantly related to the sentencing outcome when effects of victim race were evaluated (Paternoster, 2003). The Purpose of the Study Reiterated Although stated in Chapter 1, the foregoi ng discussion may have clarified the intended objectives of the present study. To re iterate, the purpose is to explore whether the Court’s McKoy v. North Carolina (1990) decision affected the role of mitigation in capital sentencing in North Carolina. A larg e set of capital cases in North Carolina will be analyzed to determine whether ther e is a difference before and after the McKoy decision in the role of mitigating factors in predicting juries’ sentencing recommendations. A secondary consideration is whether the predictiv e models of death sentencing show less or more arbitrariness between the two periods. Three possibilities exist: (1) There will be no difference in the two pe riods, with predictors of death sentencing being essentially the same during both time periods; (2) Because unanimity was required preMcKoy mitigating factors will have a stronger influence; or

PAGE 24

20 (3) If jurors were indeed dissuaded from considering mitigating factors unless there was juror unanimity, freeing them from this cons traint could enhance the relationships of mitigating factors with sentencing recommendations. Endnotes 1Another finding of the GAO report, one that received greater attention that the evidence concerning arbitrariness, was the consis tent result that white victim cases were significantly more likely to re ceive a death sentence than t hose involving the murders of non-white victims. While the presence of non-le gal factors is not a focus of the present research, variables are included as controls in the analysis that will allow for an assessment of their roles as predictors. 2This work was the centerpiece of the defe nse’s claim of systemic discrimination in the U.S. Supreme Court ruling McCleskey v. Kemp (1987). The findings showed that the murder victim being white was a statisti cally significant predictor of death sentencing in Georgia. While the legacy of this work is the finding concerning discrimination, the results contain valuable informati on about aggravation and mitigation. 3The six aggravating circumstances found to have a significant effect were “record of murder, terror, or serious assault,” “contract murder,” “murder was heinous, atrocious, or cruel,” “multiple victims,” “great risk of death to others,” and “murder as committed in an effort to conceal the commission of another crime.” 4The aggravating circumstances “murder committed during the course of another felony,” “previous convicted of violent crime,” “defendant created great risk of death to others,” “murder for money,” “murder for hire ,” “murder wanton or vile” and “defendant was prisoner or escapee” (Baldus et al., 1990, p. 657). A number of mitigating circumstances was also significantly rela ted to the sentenci ng outcome, namely “defendant 16 or younger,” “no conviction for a previous violent felony,” “defendant was underling in murder” and “defendant wa s provoked” (Baldus et al., 1990, pp. 656-657). 5Undoubtedly, the authors were cognizant of the fact that juri es’ responses to mitigation had different meanings in years before the McKoy decision, so they elected to focus on cases during a relatively limited time span in the postMcKoy era. This study is available on-line as a repor t from the agency that sponsored the research, but, surprisingly, does not appear to have been published in a scholarly journal.

PAGE 25

21 Chapter Four Method and Analytical Procedures Data Description of the samp le and case materials The analysis is based on information from reviews of capital murder tr ials in North Carolina. These cases were located through LexisNexis searches of North Carolina Supreme Court and Court of Appeals cases. In these trials, the defendants were convicted of, or pled guilty to, first degree murder, the state sought the death pe nalty, the trial progressed to a sentencing phase whereby the jury heard evidence concerning aggravating and mitigating factors, and the jury issued a binding recommendati on for a sentence. In making a sentencing recommendation, North Carolina capital juries ha ve only two options, a death sentence or a sentence of life in prison, cu rrently one without the possibi lity of parole except by the governor’s clemency. Included in the analyses are cases where the sentencing phase was conducted, but the jury declared that they could not reach the required unanimous decision regarding a sentence (in essence, a “deadlocked jury”), resulting in the default sentence of life in prison. Reviews of capital trials were deri ved from public records materials that accompany decisions regarding appeals of capital murder convictions rendered by the North Carolina Supreme Court and the North Ca rolina Court of Appeals. These materials include defendant and state briefs, as well as a form completed by the jury that records its responses to aggravating and mitigating factor s, and concludes with the jury’s sentencing recommendation. Historically, these material s have been published in hard-copy form and placed in two university law libraries in North Carolina (the University of North Carolina – Chapel Hill and Wake Forest University), while other locations have microfilm copies. Beginning with decisions returned from cases tried in 1999, hard copies have not been made available, but ma terials are accessible vi a an electronic data file (http://www.ncappellatecourts.org).

PAGE 26

22 There are 818 cases in the dataset fr om trials held during the period 1979-2000. The initial year of 1979 represen ts the first year following the Gregg v. Georgia (1976) decision that death sentences were likely to be sustained upon appeal to the North Carolina Supreme Court. The year 2000 represen ts the latest year for which Supreme or Appeals Court decisions have been issued for th e substantial majority of appeals filed. Of these, 741 cases are original trials while 86 ar e retrials following a vacating of either the defendant’s conviction a nd/or death sentence. Because there is no centralized source of information regarding capital murder trials in North Carolina, it is impossible to determine the precise number of all capital murder trials conducted during the period under investigation. However, appeals of death sentences are automatically referred to the state Supreme Court. Also, a large proportion of defendants receiving a life sentence appeal their first degree murder convictions to the state Court of Appeals. If the Court of App eals decision is not in their favor, defendants may appeal to the state Supreme Court, but th at court has the option of declining to hear the case. Given that the substant ial majority of capital cases are appealed to at least one of these courts, we estimate that the availa ble data contain revi ews of 80-90% of all sentencing recommendations made by juries during this period. There are two instances were defendants are unlikely to appeal, and therefore not be included in the dataset. First, if they pl ed guilty and received a life sentence, there is little basis for appeal. Second, some defendant s’ convictions are uphe ld, but their death sentences vacated. If, upon retria l of the penalty phase, they receive a life sentence, there is no basis for appeal. Both of these situations result in cases that are difficult to discover, especially if the trials were held in smaller rural counties without a major news outlet. A much smaller basis for some trials not incl uded in the dataset involved those that were actually identified, but their case materials we re not available because hard copies were missing from both libraries or not yet pos ted in electronic form (n = 12 cases). Of the 818 cases reviewed, 598 had comp lete information necessary for the analyses. The preMcKoy dataset consists of 210 cases and the postMcKoy dataset consists of 388 cases. Several sources of missing data have been determined that resulted in cases being excluded from the working dataset. These sources include:

PAGE 27

23 Cases that did not have a full set of mate rials necessary for review. Specifically, a number of appeals in cases where the i ndividual received a life sentence did not include the jury recommendation form (termed “Issues and Recommendation”) among the case materials. Therefore, it wa s impossible to determine the specific aggravating circumstances and mitigating factors submitted for jury consideration. Also excluded from the analysis are tria ls that involved two types of situations emerging from the jury deliberations. Firs t, the jury did not find an aggravating factor. Second, the jury found an aggravating circumstance to exist, but judged that it did not merit the death penalty. In either cas e, the sentencing decisi on defaults to life, and the deliberations conclude prio r to considering mitigating evidence. In the early postGregg years of capital trials in Nort h Carolina, the juries of some counties were submitted a set of mitigators and were asked if they accepted any of those listed. Thus, the acceptance or re jection of individual mitigators was not required, rendering these cases invalid for this analysis Finally, some appeals were prepared in a manner that did not allow for coding of all variables used in the analysis. That is, descriptions of the crime were lacking in detail, or materials were excluded that were necessary to complete some codings. Defendant information Defendants’ age, race, and sex were available from the North Carolina Department of Corrections we bsite, http://www.doc.state.nc.us/offenders. Victim information Through 1996, victims’ age, race, and sex were taken from a commercially available CD-ROM, North Carolina Vita Records: Deaths 1968-1996 For 1997-2000, victims’ demographic information was determined from some combination of court material (such as reference to the vi ctim in the state’s or defendant’s appeals briefs), newspaper accounts, or obituarie s obtained through Worl d Wide Web search engines. Cases for which this information co uld not be obtained are not included in the dataset. Description of Variables Used in the Analysis Dependent variable The dependent variable for the analysis is whether or not the jury assessed the defendant a sentence of deat h. For all cases, life sentences are coded as “0” while death sentences are coded as “1.”

PAGE 28

24 Focus variables: Aggravati ng and Mitigating Circumstances The main focus of this study is on the effect of the McKoy v. North Carolina decision (1990), which changed the unanimity requirement for mitigating circumstances, and how it affected the sentencing outcomes in capital murder trial. North Carolina is a “weighing” state in which the jury must weigh aggravating circumstances against mitigating circumstances in making a sentencing decision. Mi tigating circumstances are de fined as: “a fact or group of facts which do not constitute any justificati on or excuse for killing or reduce it to a lesser degree of the crime of first-degree murder, but which may be considered as extenuating, or reducing the moral culpability of the killing, or making it less deserving of the extreme punishment than are other first degree murder s” (North Carolina Capital Punishment Statutes, p. 148). The North Carolina Capital Punishment St atutes identify 11 st atutory aggravating circumstances in the order as they appear in the statutes. These are: 1. The capital felony was committed while th e defendant was engaged, or was an aider or abettor, in the commission of, or an attempt to commit, any homicide, robbery, rape or a sex offense, arson, burglary, kidnapping, or aircraft piracy or the unlawful throwing, placing, or di scharging or a destructive device or bomb. 2. The capital felony was comm itted for pecuniary gain. 3. The capital felony was especially heinous, atrocious, or cruel. 4. The murder for which the defendant stands convicted was part of a course of conduct in which the defendant engage d and which included the commission by the defendant of other crimes of violence against another person or persons. 5. The defendant knowingly created a great risk of death to more than one person by means of a weapon or device, which would normally be hazardous to the lives of more than one person. 6. The defendant had been previously convi cted of a felony involving the use or threat of violence to the person or had been previously adjudicated delinquent in a juvenile proceeding for committing an offense that would be a Class A,

PAGE 29

25 B1, B2, C, D, or E felony involving the use or threat of violence to the person if the offense had been committed by an adult. 7. The capital felony was committed by a person lawfully incarcerated. 8. The defendant had been previously convi cted of another capital felony or had been previously adjudicated delinqu ent in a juvenile proceeding for committing an offense that would be a capital felony if committed by an adult. 9. The capital felony was committed for the purpose of avoiding or preventing a lawful arrest or affecting an escape from custody. 10. The capital felony was committed to disrupt or hinder the lawful exercise of any governmental function or th e enforcement of the laws. 11. The capital felony was committed ag ainst a law-enforcement officer, employee of the Department of Correcti ons, jailer, fireman, judge or justice, former judge or justice, prosecutor or fo rmer prosecutor, juror or former juror, or witness or former witness agains t the defendant, while engaged in the performance of his official duties or b ecause of the exercise of his official duty” (North Carolina Capital Punishment Statutes). North Carolina also provides a list with nine statutory mitigating circumstances, which are the following: 1. The defendant has no significant hist ory of prior criminal activity. 2. The capital felony was committed wh ile the defendant was under the influence of mental or emotional disturbance. 3. The age of the defendant at the time of the crime. 4. The capacity of the defendant to appreciat e the criminality of his conduct or to conform his conduct to the requirements of law was impaired. 5. The defendant was an accomplice in or accessory to the capital felony committed by another person and his participation was relatively minor. 6. The defendant acted under the duress or under the domination of another person. 7. The defendant aided in the apprehension of another capital fe lon or testified truthfully on behalf of the prosecut ion in another prosecution of a felony.

PAGE 30

26 8. The victim was a voluntary participant in the defendant’s homicidal conduct or consented to the homicidal act 9. Any other circumstance arising from th e evidence which the jury deems to have mitigating value (known as the “catchall” factor). The prosecution must prove the existen ce of the aggravating circumstance or circumstances beyond reasonable doubt and the aggravating circumstances must be found unanimously by the jury. In contrast to the aggravating circumstances, mitigating circumstances must only be proven to the satisfaction of an i ndividual juror. PreMcKoy the mitigating circumstances had to be found unanimously by the jury. For the analysis, aggravation was measur ed as a summed variable consisting of the numbers of aggravators accepted by the ju ry. Mitigation was measured as the sum of the eight statutory variables (described above ) accepted by the jury and the sum of all non-statutory mitigators accepted by the jury ; the latter count included any “yes” responses to the catchall question.1 The range of aggravators accepted was 1 – 9. The range of mitigators accepted was 1 – 44. Control Variables To determine whether any results would hold when taking into account the effects of other va riables shown by past research to predict death sentencing, a set of control variables employed a number of control variables based on previous research were included. Some of these were additional variables could be interpreted as additional legal factors that could legitimatel y influence jury decisions, either by further aggravating or mitigating the case. These variables were: multiple victims (coded 1 if the case had multiple murder victims) urban or rural county in which the trial was held (coded 1 if an urban county) whether the defendant confessed (cod ed 1 if the defendant confessed). Another set of variables were included to capture the possible effect of non-legal factors influencing sentencing outcomes. Thos e included are, by far, the most prominent non-legal variables st udied in the postGregg literature, and have been found in a majority (but not all) studies to emerge as predictors of death sentencing. Comprised of demographic characteristics of defendants and victims, these variables are: race of the defendant (coded 1 for white defendants, 0 for black and other defendants)

PAGE 31

27 age of the defendant (inter val variable ranging from 15-77)2 race of the victim (coded 1 for white vi ctims, 0 for black and other victims) sex of the victim (coded 1 for male victims, 0 for female victims) age of the victim (interval variable ranging from 0 to 100) Statistical Analysis The analysis consists of two parts, deve lopment of descriptive statistics and a multivariate analysis. The descriptive statisti cs give an overview about the number of aggravating and mitigating circumstances presented to and accepted by capital juries and where they have changed, if at all, as result of the McKoy decision. A logistic regression analysis examines the extent to which thes e changes, if any, have influenced capital sentencing outcomes. The logis tic regression analysis consis ts of three models: (1) a model for the total dataset including 598 cases with a variable in the model indicating whether a case was preor post -McKoy ; (2) a model for the preMcKoy dataset including 210 cases; and (3) a model for the postMcKoy dataset including 388 cases. The logistic regression analysis allows for a comparison of the change in the impact of mitigating circumstances postMcKoy A comparison of the models also allows for assessing the impact of the legally relevant variables of aggravation and mitigation, while controlling for nine other variables on the sentencing outcome preand postMcKoy Logistic regression can be used when the dependent variable (life or death sentence) is a binary variable. Logistic regr ession estimates the imp act of the independent variables on the odds that a de fendant would receive a sentence of death. The odds ratio measures the strength and direction of the independent variables on the probability of a death sentence. The odds refer to the probabi lity of an event occurring (death sentence) divided by the probability of an event not occurring (no death sentence). The odds can take values between 0 and The odds ratio determines whether the probability of a certain event (ex. death sentence) is the same for 2 groups (e.g. white and black defendants). The odds ratio is the ratio of th e odds of an event occurring for one group (e.g. the odds of a death sentence occurring fo r white defendants) divided by the odds of an event occurring for another group (e.g. th e odds of a death sentence occurring for black defendants). An odds ratio of a 1 indica tes that an event is equally likely in both groups. When an odds ratio is greater than 1 it indicates that the odds of getting a 1 on the

PAGE 32

28 dependent variable increases when the inde pendent variable increases. When an odds ratio is between 0 and 1 it indicates that th e odds of getting a 1 on the dependent variable decreases when the independent variable increases (Menard, 1995). The size of the effects of the independen t variables on the probability of a death sentence is reported through the coefficients (b) of the independent variables. For instance, if the coefficient (b) for the numb er of aggravating circumstances accepted is .857 and the odds ratio is 2.356, this mean s that every additional aggravating circumstance accepted increases the probability of a death sentence by 240 percent (2.356 rounded). If the coefficient has a negati ve value, the odds ratio would mean that with every additional circumstance accepte d the probability of a death sentence decreases by 240 percent. An independent variable is said to have a statistically significant effect if that variable contributes to the prediction of the probability of a death sentence beyond a contribution that might be due to chance. A probability of .05, or 1 in 20, is low enough to be thought of as statistically significant. The interpretation of th e results will be based on whether the legally relevant variables are shown to be statistically significant and, based on their relative probabiliti es, whether they are more or less important as predictors of death sentencing during the periods before and after the McKoy decision. Endnotes 1The influence of the single aggravating and mitigating factors was tested, but due to the small number of submitted and accepte d single aggravating and mitigating factors we created a summed variable (numbe r of aggravating circumstances). 2Arguably, this variable could also serve as a mitigator because juries may decide that the youth (or even older ages)of the defe ndant serves to lessen culpability. However, all other things being equal, as they are in tr eated in the statistical analysis to follow, it could also be argued that age, like race and/or sex, is an ascr ibed status that should not be a determinant of sentencing decisions.

PAGE 33

29 Chapter Five Results The first aspect of the pe riods before and after the McKoy v. North Carolina (1990) decision is whether their has been an apparent shift in the total number of mitigating circumstances both presented by the defense and accepted by the jury. An assumption would be that eliminating the unanimity requirement would present the defense with greater latitude in presenting mitigators, knowing that they only have to convince one juror of that fact or’s relevance to the case. This assumption is supported by the results shown in Table 1 where the aver age number of aggravating and mitigating circumstances presented to and accepted by the jury for 210 preMcKoy cases and 388 postMcKoy cases is presented. While the aggravating circumstances presented and accepted have increased only slightly in the postMcKoy era, the number of mitigating circumstances presented has more than doubled postMcKoy increasing from 9.12 to 20.6, and likewise, the average number accepted increased from 4.31 to 10.46.

PAGE 34

30 Table 1 Average Number of Aggravating and Mitigating Circ umstances Presented and Accepted Preand Post-McKoy Presented Accepted Pre-McKoy Post McKoy Pre-McKoy Post McKoy Total Number of Aggravating Circumstances 2.04 2.45 1.81 2.25 Total Number of Mitigating Circumstances 9.12 20.61 4.31 10.46 a) Statutory Mitigating Circumstances 2.3 2.86 0.97 1.4 b) Non-Statutory Mitigating Circumstances 6.82 17.75 3.34 9.06

PAGE 35

31 Table 1 also presents the average numb er of statutory versus non-statutory mitigating circumstances presented and accepted. PreMcKoy defense counsel presented on average 2.3 statutory mitigating circumstances and the jury, under the preMcKoy unanimity requirement, accepted .97. Furthermore, postMcKoy 2.86 statutory mitigating circumstances were presented and 1.4 were accepted. This finding demonstrates a slight increase in the number of statutory mitigati ng circumstances both presented and accepted postMcKoy Regarding non-statutory mitigating circumstances, preMcKoy defense counsel presented on average 6.82 and the jury accepted 3.34 of them. In contrast, postMcKoy defense counsel introduced on average 17.75 nonstatutory mitigating circumstances and the jurors accepted 9.06 of them These results show clear increases in the number of non-statutory mitigating circumstances presented and accepted postMcKoy Interestingly, Table 1 also demonstrates an increase in the average number of aggravating circumstances both introduced and accepted postMcKoy despite the fact that the Court’s decision did not ch ange the interpretation of how aggravating circumstances should be considered by the jury. One possible explanation for this jump in the average number of aggravating ci rcumstances presented postMcKoy by prosecutor’s may be a strategic attempt to combat the larger number of mitigating circumstances being presented by defense counsels. Because the number of both statutory and non-statutory mitigating circumstances presented and accepted has more than doubled postMcKoy it is possible that mitigating circumstances might have a greater in fluence on capital sentencing outcomes by decreasing the probability of a death sentence Logistic regression was used to model the determinants of capital sentencing, a nd, in turn, test this hypothesis. The results of these logistic regression analyses are pr esented in Table 2. Three models are presented; one is for the total sa mple of capital cases and one each for cases sentenced both preMcKoy and postMcKoy The logistic regressi on model for the total sample includes a dichotomous variable to distinguish between preMcKoy and postMcKoy cases (postMcKoy = 1). Surveying the results, th is model indicates that the McKoy decision had an impact on capital senten cing outcomes because of the effect of

PAGE 36

32 this dichotomous, prevs. postMcKoy variable attains statisti cal significance (b = .871, p < .001). However, the effect of this variable is positive and indicates that the odds of receiving a death sentence were 2.4 times higher in the postMcKoy era. While the McKoy decision theoretically presented defenda nts with a greater opportunity to sway jurors through mitigation, the results indicate that, all other things being equal, postMcKoy defendants were actually at greater risk of receiving a death penalty.

PAGE 37

33Table 2 Logistic Regression Model Variable Total Pre-McKoy Post-McKoy Differences in Effects b se(b) Odds Ratio b se(b) Odds Ratio b se(b) Odds Ratio Z p Post-McKoy cases .871** .227 2.389 N.A. N.A N.A. N.A. N.A. N.A. N.A. N.A. Aggravators Accepted .857** .119 2.356 .475* .194 1.608 1.059** .164 2.885 2.30* .022 Mitigators Accepted -.098** .014 .906 -.181** .044 .835 .098** .016 .907 1.77 .076 Multiple Victims .034 .226 1.034 .103 .372 1.108 -.069 .308 .993 -.356 .718 Confession .079 .051 1.083 .416* .152 1.515 -.011 .061 .989 -2.61* .008 Defendant Age .028* .011 1.028 -.005 .018 .995 .055** .017 1.057 2.42* .016 Defendant Race (NonWhite) -.153 .232 .858 -.152 .381 .859 -.127 .311 .881 .051 .960 Victim Age -.016* .005 .985 -.015 .008 .985 -.017* .007 .983 -.188 .850 Victim Sex .405* .202 1.500 -.009 .329 .991 .678* .269 1.971 1.62 .106 Victim Race (White) .331 .247 1.392 .258 .415 1.294 .369 .320 1.447 .212 .842 County -.425* .205 .653 -.309 .335 .734 -.526 .280 .591 -.497 .624 Intercept -1.220 .558 .295 .623 .909 1.864 -1.389 .737 .059 -2 Log Likelihood intercept 783.446 286.227 492.066 model 622.597 235.378 359.634 Model X2 160.849 50.849 132.432 Corrected R2 .323 .289 .402 N 598 210 388 *p < .05; **p < .001

PAGE 38

34 The logistic regression model for the total data set also shows that the number of aggravating and mitigating circumstances accepted by the jury are related to capital sentencing outcomes in a manner consistent with their legal function. Th at is, the greater the number of aggravating circumstances accepted by the jury, the greater the likelihood of a death sentence (b = .857, p < .001). Convers ely, the greater the number of mitigating circumstances accepted by the jury, the lower the likelihood of a death sentence (b = -.098, p < .001). In fact, each additional aggravating ci rcumstance accepted by the jury increases the odds of a death sentence by a multiple of 2.36, while each additional mitigating circumstance accepted by the jury reduces the odds of a death sentence by approximately 9%. The results from this first model presented in Table 2 suggest that several extra-legal variables also significantly in fluence sentencing outcomes. First, defendants in urban counties are 34.7% less likely to receive a death sentence than defendants in rural counties. Second, murderers of female victims were 1.5 time s more likely to receive a death sentence than murderers of male victims. Third, killer s of younger victims have a higher probability of receiving a death sentence (b =.016, p < .05). The age of the defendant also significantly influences sentencing outcomes -the younger the defendant, the less likel y the probability of a death sentence (b = .028, p < .05). The effects of multiple victims, de fendant race, victim race, and confession failed to at tain statistical significance. While these data do not measure the weight s given to the various aggravating and mitigating circumstances accepted by capital juries in North Carolina, they do show that numerous mitigating circumstances may be re quired to overcome even a small number of aggravating circumstances. Because the McKoy decision reduced the burden to the defense at penalty phase for jurors accepting mitigating circ umstances, it is possible that the influence of aggravating and/or mitigating circumstance on capital case sentencing outcomes in North Carolina varies across these two capital sentencing eras To address this possibility, two additional logistic regression models are presented in Table 2. To reiterate, the preMcKoy model is based on 210 capital cases sentenced while the postMcKoy model is based on 388 capital cases sentenced. The preand postMcKoy models presented in Table 2 bot h show that increases in the number of mitigating factors accepted by capital jurors decrease the probability of a death

PAGE 39

35 sentence. However, contrary to what might be expected, mitigating circumstances have less impact on sentencing outcomes postMcKoy In the preMcKoy era, each additional mitigating circumstance accepted by the jury decreas ed the probability of a death sentence by 16.5%. In the postMcKoy era, each additional mitigating circumstance accepted decreased the probability of a death sentence by only 9.3%. Table 2 also repor ts the results of a test for the equality of preand postMcKoy model coefficients (Brame, Paternoster, Mazerolle, and Piquero, 1998). The differences in the effect s of the impact of number of mitigating circumstances accepted preand postMcKoy are statistically signi ficant (difference = 1.77), reinforcing the finding that mitigating factors appear to have less of an effect in the postMcKoy era. Conversely, the impact of the number of aggravating circumstances accepted on sentencing outcomes was much stronger postMcKoy than preMcKoy For both the preand postMcKoy model, as expected, an increasing number of aggravating circumstances accepted were significantly associated with a greater probability of a death sentence. In the preMcKoy model, each additional aggravating circum stance accepted increased the odds of a death sentence by a multiple of 1.6. In the post -McKoy model each additional aggravating circumstance accepted increased the odds of a death sentence by a multiple of 2.89. The difference in these effects (differenc e = 2.30) was statis tically significant. Even though the number of mitigating circumstances doubled postMcKoy the impact of mitigating circumstances on sent encing outcomes decreased; moreover, the influence of aggravating factors and the probabi lity of a death senten ce both increased during the postMcKoy period. These findings are contrary to any expectation th at the increased latitude in having mitigators accepted may have resulted in decreased death sentencing. In practical terms, the McKoy decision may have inspired de fense counsel to submit more mitigating factors at penalty phase, and it ma de jury acceptance of these mitigators more likely by eliminating the unanimity requirement but the weight or influence of these additional mitigators appears to have been very slight. Moreover, their presence may have diluted or diminished the influence of other more salient mitigators. In turn, the influence of aggravating circumstances may have increase d and the odds of a d eath sentence became more likely as a consequence.

PAGE 40

36 These preand postMcKoy models presented in Table 2 also provide additional evidence of the extent to which the death pena lty may have been applied arbitrarily and/or capriciously in North Carolina. In the preMcKoy model, only the effects of three legally relevant variables attained statistical signi ficance: number of aggravating circumstances accepted, number of mitigating circumstances accepted, and confession. None of effects of the extra-legal variables attained statis tical significance. Conv ersely, in the postMcKoy model, five variables attain significance. Three of these variables are legally relevant to the sentencing decision, the number of both aggr avating and mitigating circumstances accepted, and defendant age. The other two variables th at attained significance were quasi-legal variables. First, as victim’s age increased, the odds of a death se ntence decreased. Second, murderers of female victims were twice as likel y to receive a death sentence as murderers of male victims. These findings suggest that the death penalty in North Carolina shows no clear and convincing evidence of arbitrariness nor capriciousness during either the preMcKoy or the postMcKoy eras. The results shown in Table 2 reveal that lega l factors are the primary predictors of death sentencing. Although th e results make clear that aggravating circumstances are much stronger predictors of a death sentence compared to mitigating. While both defendant and victim age are included among the stat istically significant variables these can be argued as being releva nt to the sentencing decision and cannot be necessarily be construed as evid ence of caprice relevant to th e sentencing decision and could not be construed as evidence of caprice. Howe ver, the significant effect of victim’s sex during the postMcKoy era suggests that some level of gender-based bias may be present. Although beyond the scope of the present study, this effect might be attributed to femalevictim murders in which rape was involved, a sing le aggravator that ma y carry special weight in tilting the jury’s decision toward a death sentence.

PAGE 41

37 Chapter Six Discussion and Recommendations for Future Research Summarizing the Results In McKoy v. North Carolina (1990), the Court increased jurors’ di scretion in capital sentencing by holding that jurors do not have to find mitigating circumstances unanimously, but that each juror can find the mitigating circumstances he/she believes to be present. This decision provided the basis for our primary research question, which was whether the McKoy (1990) decision influenced the way in which aggravating and mitigating circumstances are processed in North Carolina. The focus of the Court on individualized sentencing also represents the basis for a relate d question as to whether or not the death penalty system in North Carolina operates arbitrarily, meaning the le gally relevant variables of aggravation and mitigation fail to determine the probability of a death sentence. Since each juror can now find mitigating ci rcumstances independent of the other jurors in the postMcKoy era, it was expected that the number of mitigating circumstances presented and accepted would increase. This ex pectation is supported by the descriptive statistics, which showed that the average number of mitigating circumstances presented and accepted doubled in the postMcKoy cases. At the same time, the number of aggravating circumstances presented and submitted stayed about the same. The analysis then moved to a consideration of the impact of mitigating circumstances, and whether there had been a change between the two eras. Separate logistic regression analyses re vealed that there had indeed been a shift in the effects of aggravat ion and mitigation, but not the manner that some might have anticipated. Specifically, in the postMcKoy era, mitigating circumstances were found to have a diminished impact on the proba bility of a death se ntence while, conversely, aggravating circumstances car ried an increased impact.

PAGE 42

38 Discussion of the Results The findings of this study suggest that even though the number of mitigating circumstances accepted has doubled postMcKoy they have less influence on the probability of a death sentence. One possibl e explanation could be a diffusion effect. In essence, even though jurors accept twice as many mitigating circumstances postMcKoy the accepted mitigating circumstances seem to carry less weight. Jurors are presented with on average 20 mitigating circumstances by the defense counsel in the postMcKoy era. The jurors might have the impression that defense attorneys pr esent anything they possibly can because there is not a specific reason why the defendant does not deserve to die. As an extension of this thinking, jurors might believe that if the defendant has a good r eason why he should not die, he would not need to submit so many mitigating circumstances. Ironically, the decreased relevance of mitigating circumstances postMcKoy could also be caused by the abandonment of th e unanimity requirement for mitigating circumstances that was dictated in McKoy v. North Carolina In the preMcKoy era, jurors had to find mitigating circumstances unanimous ly and when they did, these circumstances were probably very salient and carri ed significant weight. In the postMcKoy era, jurors do not have to agree on the same circumstance and therefore it may be that the “accepted mitigating circumstances” are not given the same weight. Jurors may engage in fewer discussions about whether or not to accept a cer tain mitigating circumstance; therefore, each juror is left to decide for her/himself self whether to accept the mitigator, and the approval of the other jurors is not necessary. If the jurors do not discuss the importance of a mitigating factor together, those who do not initially agree with it are un likely to consider it when it comes time to vote for the sentence. A second research question referred to the issue of arbitrarine ss and whether legally relevant variables determine the sentencing outcome. The logistic regression analysis indicates that North Carolina sentencing outco mes appear to have been significantly influenced by the legally relevant variable s of aggravation and mitigation. The analysis, however, also suggests that even though the number of mitigating circumstances accepted significantly decreases the proba bility of a death sentence, the impact of mitigating circumstances is only marginal when compared to the impact of aggravating circumstances, especially postMcKoy In the absence of guidance fo r jurors on how to weigh the

PAGE 43

39 aggravating against the mitigating circumstances and when everything counts in aggravation and mitigation, subjectivity appears to influen ce sentencing outcomes and invites arbitrary decision-making (Steiker and Steiker, 1995). In both time periods, jurors were given the opportunity for arbitrary decision-making by the state being allowed to submit vague statut ory aggravating circumstances. Some of the statutory aggravating circumstances are so broa d that they could be submitted in any firstdegree murder case. North Carolina employs the aggravating circumstance “murder was heinous and/or cruel,” which means that in some way the murder was worse than another and the defendant therefore deserves the death pe nalty. The terms “heinous and cruel” are very subjective and very open to interpretati on. A similar broad statutory aggravating circumstance is “murder for pecuniary gain,” which can include anything from a robbery to a murder for hire, so has great latitude in its meaning. The statutory aggravating circumstance “capital felony was committed while the defendant was engaged in another felony” can also be applied to many other non-capital first degree mu rders. These examples show that there is a broad prosecutorial discretion to seek the de ath penalty among theoretically death-eligible cases and, when sought, if the jury accepts one or more aggravating circ umstances, they have virtually unbridled discretion to impose a life or death sentence. Just ice Potter Stewart wrote in Gregg v. Georgia (1976, p. 46) that “a system could ha ve standards so vague that they would fall adequately to channel the sentencing decision patterns of juries with the result that a pattern of arbitrary and capricious sentencing like that found unconstitutional in Furman could occur.” The goal of narrowing the class of death eligible defendants and provide a consistent application of the death penalty is hard to achieve when almost any first-degree murder can be death-eligible by employing va gue aggravating circumstances that make a defendant death-eligible. The current death penalty system may be deceptive because it creates the impression (some would say “illusion”) that the death penalt y is administered fairly and non-arbitrarily. The Court has, over the past three decades, creat ed a death penalty system that is highly discretionary, but generates the impression of a greatly regulated, supervised system. Steiker and Steiker (1995, p. 3) propose that the death penalty today “… is, perversely, both overand under-regulated. The body of doctrine produced by the Court is enormously complex and its applicability to sp ecific cases difficult to discern; yet, it remains unresponsive to the

PAGE 44

40 central animating concerns that inspired the C ourt to embark on its regulatory regime in the first place. Indeed, most surprisingly, the ove rall effect of twenty-odd years of doctrinal head-banging has been to substantially reproduce the preFurman world of capital sentencing.” That said, one particular finding merits discussion because of the absence of an effect. As discussed in Chapter 3, one of the most consistent findings in the postGregg death penalty literature concerning discrimination is a persistent race-of-vic tim effect found across a variety of studies. The impact of this much-reported effect is to increase significantly (although the magnitude varies from state to stat e) the likelihood of de fendants receiving a death sentence when the victim is white. In cont rast to this rather voluminous literature, none of the three models presented in Table 2 demons trate this effect. It is noteworthy that the only other systematic analysis of North Caro lina death sentencing in recent years is the research of Unah and Boger (2001) (see Ch apter 3, especially Endnote 5). Their work attracted considerable attention upon its re lease because it purporte d to show a pronounced race-of-victim effect in North Carolina deat h sentencing practices whereby murderers of whites were more likely to receive a death sentence, even when controlling for legal factors of the cases. Careful scrutiny of their results re veals that this effect is found at the decisionmaking levels of deciding to pursue a first de gree murder charge and to seek the death penalty. However, in a little discussed portion of their report, the race-of-victim effect disappears at the trial level. To re-emphasi ze Chapter 3, Endnote 5, the years of Unah and Boger’s analysis were somewhat limited (1995-1999) and were all postMcKoy It is of considerable interest that the similar findings di scussed in this thesis are from much larger set of capital trials across a much br oader span of years. This effect is more remarkable in the light of the fact that, if Unah and Boger’s fi ndings are generalizable to the larger set of capital trials discussed here, juries have been, in effect, handed a set of cases in which a raceof-victim effect is already embedded. Explanat ions why this effect disappears at the trial level – arguably the stage of the criminal just ice process most subject to arbitrariness and capriciousness – are quite elusive, and will pose a considerable challenge for future researchers. Finally, one possibility exists that is diffi cult to ascertain, but the influence of which cannot be ruled out. Faced with the challenge of seeking the death penalty for defendants

PAGE 45

41 who are entitled to more latitude in having the potentially mitigating aspects of their crimes considered, state prosecutors may have become more selective in their cases in the postMcKoy era. That is, they are less likely to go fo rward with single-aggravator cases, and to pay more attention to what mitigation the defe ndant might present so that it can be more aggressively countered in the sentencing phase of the trial. In addition, prosecutors may have gone to greater lengths to presen t strong cases to the jury so th at the aggravators they present will carry more weight. In essence, the McKoy decision may have resulted in a conscious change in prosecutors’ presentation of aggravat ion, leading to an enhanced effect of these legal aspects of the cases. Suggestions for Future Research An acknowledged weakness of empirical de ath penalty studies using secondary analysis is that there are undoubt edly subtle variables that may influence decisions for which no controls are feasible. These include, but ar e not limited to, the demeanor of the defendant during the trial, the sheer quality of the pros ecution and defense presen tations, and the sociopolitical environment surrounding the trial (e.g., the O. J. Simpson trial). However, there are some other avenues for future suggested by this research that could be pursued, some of which are discussed in the paragraphs that follow. The decision-making of jurors has been a ssessed by the Capital Jury Project (“CJP”; see Bowers, 1995), a consortium consisting of criminologists, law faculty, and social psychologists. As part of the CJP, researcher s interviewed between 80 and 120 former capital jurors from each of 14 states, including North Carolina (Luginbuhl and Howe, 1995). The interviews are open-ended about the “jurors ex periences and decisionmaking over the course of the trial, identify points at which various infl uences come into play, and reveal the ways in which jurors reach their final sentencing deci sion” (Bowers, Fleury-Steiner, and Antonio, 2003, p. 423). A major finding of the CJP is that ju rors often do not correctly understand the sentencing instructions provide d to them by the court, especially how the rules for considering mitigating circumstances are different from the rules for considering aggravating circumstances. Because the study is spread over a number of stat es, the numbers of cases studied in each state is relatively limited. A possibly useful avenue of exploration is to determine how widespread the misunderstandings of jurors were across a broad range of capital cases in North Carolina, and how this may have impacted their considerations of

PAGE 46

42 aggravating and mitigating circumstances. Th e study could replicate some of the CJP methodology, but be tailored more specifically to the possible confusion generated by the seemingly obtuse wording that characterizes North Carolina’s “Issues and Recommendation” form.1 If such misunderstandings are indeed widespread, arbitrarine ss could enter in by virtue of the jurors erroneously belie ving that their options for vo ting life or death were limited, thus undermining the impact of the legal factor s of the case. Further, more extensive, research is needed to thor oughly explore this possibility. As a policy recommendation that would ope n up numerous opportunities for research, states could clarify the requi red weighing process by quantifyi ng the weights given to single aggravating and mitigating circumstances. Alth ough resolving the details would present a formidable challenge that is well beyond the sc ope of this thesis, a sy stematic quantification of the weighing process would give capital jurors consider ably more guidance in their sentencing decision, thereby furt her reducing the unfettered disc retion that was condemned in Furman v. Georgia (1972). Conceptually, th e jurors would compare their scores assigned to aggravating to those assigned to mitigati on and sentence accordingly. By quantifying the weighing process, the state could record not only which aggravating and mitigating factors were accepted, but also how much weight the ju rors gave to each factor. That information would be useful for several reasons. First, the information about how jurors weighed the single aggravating and mitigating circumstances could be used by the state Supreme Court during the automatic appeal and proportionality review to ensure that the death penalty was not imposed arbitrarily. Second, in states wh ere the judge has the right to override the sentencing recommendation of the jury (not th e case in North Carolina), judges would be able to review how much weight was given to the single aggravating and mitigating factors, thus better informing thei r decisions to affirm or override the jury’s decision. Finally, researchers could also use the information gathered from quantitativelyoriented jury studies of this nature to gain a better unde rstanding of the decision-making process of the jurors. That information would also promote our understa nding whether jurors react negatively to a high number of mitigat ing circumstances presented by defense counsel. Jurors may have the feeling that defense counsel admits anything in mitigation they can think of. Consequently, jurors may have the impr ession that the mitigating circumstances

PAGE 47

43 submitted are meaningless and that the defense thinks that quantity is more important than quality. The suggested analysis would also promot e our understanding whether jurors react negatively to a high number of mitigating circumstances presented by defense counsel. Jurors may have the feeling that defense counsel admits anything in mitigation they can think of. Consequently jurors may have the impression that the mitigating circumstances submitted are meaningless and that the defense thinks that quantity is more important than quality. Another avenue to explore is whether it is more effective for the defense counsel to concentrate on a few important mitigating factor s that most of the jurors could agree upon, thereby maximizing their effect. Research of th is nature could focus on whether the number of mitigators rejected serves as a predictor of death sent encing, or alternatively, whether the using ratio of submitted to accepted mitigators yields meaningful results. Concluding Thought As these suggestions indicate, there is s till much work to be done in better determining capital punishment sentencing practi ces and in developing rational explanations for their existence. What cannot be resolved through empirical work alone is whether the system can ever be made just and fair e nough, despite numerous efforts by state and federal courts, to continue a practice that has b een abandoned by most of the industrial world. Endnote 1Issue Three of the form asks “Do you una nimously find beyond a reasonable doubt that the mitigating circumstances or circum stance found by you is or are insufficient to outweigh the aggravating circumstance found by you?” while Issue Four poses the question “Do you unanimously find beyond a reasonable d oubt that the aggravating circumstance you found is sufficiently substantial to call for imposition of the death penalty when considered with the mitigating circumstance or circumstan ces found by one or more of you?”. As a point of clarification, answer ing “yes” to Issue IV does not require the jury to impose a death sentence. But, answering “no” to either III or IV defaults to a life sentence.

PAGE 48

44 References Baldus, D.C., Woodworth, G., & Pulaski, C.A. Jr. (1990). Equal justice and the death penalty Boston: Northwestern University Press. Baldus, D. C., Woodworth, G ., Zuckerman, D., Weiner, N.A., & Broffitt, B. (1998). Racial discrimination and the de ath penalty in the post-Furman era: An empirical and legal overview, with recent findings from Philadelphia. Cornell Law Review 83 1683-1770. Baldus, D. C., Woodworth, G ., Grosso, C., & Christ, C. (2002). Arbitrariness and discrimination in the administration of the deat h penalty: A legal and em pirical analysis of the Nebraska experience (1973-1999). Nebraska Law Review 81 486-756. Bedau, H.A. (1982). The death penalty in America New York: Oxford University Press. Bowers, W. J. (1995). The capital jury pr oject: Rationale, design, and a preview of early findings. Indiana Law Journal 70 1043-1102. Bowers, W. J., Steiner, B. D., & Ant onio, M. E. (2003). The capital sentencing decision: guided discretion, reasone d moral judgment, or legal fict ion. In J. R. Acker, R. M. Bohm,. & C. S. Lanier, (2nd Ed.). Americas experiment with capital punishment (pp. 413468). Durham, NC: Carolina Academic Press. Brame, R. Paternoster, R. Mazerolle, P., & Piquero, A. (1998). Testing for the equality of maximum-likelihood regression coe fficients between two independent equations. Journal of Quantitative Criminology 14 245-261. Brock, D., Cohen, N. & Sorensen, J. (2000). Arbitrariness in the Imposition of Death Sentences in Texas: an Analysis of Four Counties by Offense Seriousness, Race of Victim, and Race of Offender. American Journal of Criminal Law, 28 43-71. Coyne, R., & Entzeroth, L. (2001). Capital punishment and the judicial process. Second Edition Durham, NC: Carolina Academic Press. General Accounting Office (1996). Death pe nalty sentencing: research indicates pattern of racial dispari ties. In H.A. Bedau (Ed.), The death penalty in America: Current controversies (pp.268-274 ). New York: Oxford University Press.

PAGE 49

45 Keil, T. J., & Vito, G. F. (1995). Race a nd the death penalty in Kentucky murder trials. American Journal of Criminal Justice, 20 (1) 17-36. Klein, S., & Rolph, J. (1991). Relationship of offender and victim race to death penalty sentences in California. Jurimetrics 32 33-48. Luginbuhl, J., & Howe, J. (1995). Discreti on in capital sentencing instructions: Guided or misguided? Indiana Law Review, 70 1161-1181. North Carolina Capital Punishment Statut e. N.C. Gen. Stat. Sect. 15A-2000 (e) (2004). (on file with the N.C. General Assembly). [Online]. Available from http://www.ncleg.net/Statutes/GeneralSta tutes/HTML/ByChapter/Chapter_15A.html. North Carolina Capital Punishment Statut e. N.C. Gen. Stat. Sect. 15A-2000 (f) (2004). (on file with the N.C. General Assembly). [Online]. Available from http://www.ncleg.net/Statutes/GeneralSta tutes/HTML/ByChapter/Chapter_15A.html. Paternoster, R., & Brame, R. (2003). An empirical analysis of Maryland’s death sentencing system with respect to the influe nce of race and legal jurisdiction. [online]. Available from http://www.urhome.umd.edu/newsdesk. Pierce G. L. & Radelet M. L. (2002). Race, Region, and Death Sentencing in Illinois, 1988-1997." Oregon Law Review, 81, 39-96. Rhee, F. (1993). Does court thwart N.C. death penalty? 12 pe rcent of capital sentences upheld since 1990, but justice says U. S. law is the reason. [online]. Available from http://infoweb.newsbank.com. Steiker, C. S., & Steiker J. M. (1995) Sober second thoughts: Reflections on two decades of constitutional regulation of capital punishment. Harvard Law Review 109 355437. Steiker, C. S., & Steiker J. M. (2003). Judi cial developments in capital punishment law. In J. R. Acker, R. M. Bohm, & C. S. Lanier, (eds.) Americas experiment with capital punishment (pp. 55-84). Durham, NC: Carolina Academic Press. Sundby, S. E. (1991). The Lockett paradox: Reconciling guided discretion and unguided mitigation in capital sentencing. University of California Law Review, 38 11471208. Unah, I., & Boger, J. (2001). Race and the death penalty in North Carolina The Common Sense Foundation: The Na tional Counsel of Churches. United States General Accounting Office (1990). Death penalty sentencing: Resource indicates pattern of racial disparities. Washington, DC: U.S. General Accounting Office.

PAGE 50

46 Case References Barclay v. Florida, 464 U.S. 874, (1983). California v. Ramos, 463 U.S. 992 (1983). Eddings v. Oklahoma, 455 U.S. 104 (1982). Furman v. Georgia, 408 U.S. 238 (1972). Gregg v. Georgia, 428 U.S. 153 (1976). Jurek v. Texas, 428 U.S. 262 (1976). Lockett v. Ohio, 438 U.S. 586 (1978) McGautha v. California, 402 U.S. 183 (1971). McCleskey v. Kemp, 482 U.S. 920 (1987). McKoy v. North Carolina, 494 U.S. 433 (1990). Mills v. Maryland, 486 U.S. 367 (1988). Proffitt v. Florida, 428 U.S. 242 (1976). Roberts v. Louisiana, 428 U.S. 325 (1976). State v. Kirkley, 302 S.E. 2d 144 (1983) State of North Carolina v. Dock McKoy, Jr., a/ka/ Dock McCoy, a/ka/ Dock McKay a/k/a Paul McCoy, 323 N.C. 1 (1985) Woodson v. North Carolina, 428 U.S. 280 (1976). Zant v. Stephens, 462 U.S. 862 (1983).