![Sample our Law journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5942/TOC-Subject-Sample-2021-Law.jpg"})
Abstract
Age is prominent among theories of criminology and victimology. It is less conspicuous in punishment theory, despite its emphasis in retributive theory and lawmaking. The present study evaluated competing ‘years of life lost’ and ‘vulnerable victim’ hypotheses to examine the influence of victim age in capital sentencing decisions. Using case file data on the population of capital murder trials in the State of North Carolina (1977–2009), our findings produce mixed results. Our quantitative analyses suggest that death sentences are significantly less likely in direct proportion to victim age. Killers of elderly victims are less likely to receive the death penalty; conversely, the odds of a death sentences are slightly greater for killers of child victims. Supplementary qualitative analyses suggest that while many child and elderly victims were not per se ‘vulnerable,’ a substantial subset of each clearly were treated as such. We discuss implications for vulnerable victim research and the role of quasi-legal factors in case outcomes.
Notes
1. Additional forms of statutorily-defined vulnerable victim characteristics include physical, mental, or developmental disability, pregnant women, fetuses, and victims of physical and/or sexual child abuse. Four states identify victim disability as an aggravating factor (Delaware, Florida, New Hampshire, and Wyoming), three states specifically identify pregnant women among their aggravating factors (Colorado, Delaware, and Indiana), two recognize fetuses (Arizona and Indiana), three include case circumstances in which domestic violence and/or restraining orders are a part of the victim-offender relationship (New Jersey, Pennsylvania, and Washington), and six others specify physical or sexual abuse of a child (Maryland, Montana, Oklahoma, Tennessee, Virginia, and Wyoming).
2. Several states have repealed their death penalty statutes or have had the statute declared unconstitutional; these include: Connecticut, Illinois, Maryland, Nebraska, New Jersey, and New York. Several others are under a governor-imposed moratorium on executions: Colorado, Pennsylvania, Oregon, and Washington.
3. However, within its criminal code more generally (see G. S. 15A-1340.16(d)(11)), the state does recognize victim vulnerability – ‘the victim was very young, or very old, or mentally or physical infirm, or handicapped.’.
4. The race of the victim and the offender/victim racial dyad have demonstrated substantial predictive power in capital sentencing outcomes (for reviews, see Baldus & Woodworth, Citation2003; Kavanaugh-Earl, Cochran, Smith, Fogel, & Bjerregaard, Citation2008). Several studies have also established clear evidence of a female victim effect (Gillespie, Loughran, Smith, Fogel, & Bjerregaard, Citation2014; Richards, Jennings, et al., Citation2014; Richards, Smith, et al., Citation2014; Stauffer, Smith, Cochran, Fogel, & Bjerregaard, Citation2006).
5. A number of other studies have also examined the influence of female (or white female) victims on capital charging and sentencing decisions but have produced mixed results (see Gillespie et al., Citation2014; Holcomb, Williams, & Demuth, Citation2004; Jennings, Richards, Smith, Bjerregaard, & Fogel, Citation2014, 2015; Richards, Jennings, Smith, Sellers, Fogel, & Bjerregaard, Citation2016; Richards et al., Citation2014; Stauffer et al., Citation2006; Williams, Demuth, & Holcomb, Citation2007; Williams & Holcomb, Citation2004). While victim gender may be viewed as a potential form of vulnerability; it is not legally recognized as a per se form of victim vulnerability. To be ‘vulnerable,’ the female victim needs to be perceived as having a considerable physical disadvantage relative to her killer.
6. Research focusing on sexual victimization may obfuscate the independent effects of victim age and the effects of victim credibility. Examination of homicide cases may help disentangle these effects for at least two reasons. First, the presence of a homicide victim eliminates doubt about whether a crime has actually occurred. Second, the unavailability of the victim for testimony or cross-examination automatically precludes questions about the credibility of the victim’s account.
7. In analyses not reported here we allowed for multiple victim cases involving at least one vulnerable victim to be categorized as a vulnerable victim case for all of its victims; the effects of victim vulnerability were considerably diminished relative to those reported herein.
8. A variety of other age cut-offs for identifying child and/or elderly status were also examined (e.g. ‘child’ <5, 8, 10, 14, 17 years of age and ‘elderly’> 55, 60, 62, 65, 70, and 80 years of age). Results of these operationalizations are available upon request; substantively these findings largely duplicate what we report herein. In addition, as noted above, the mean and modal age cut-point in the states’ vulnerable victim statutes is under 13, while the mean and modal cut-point which statutorily defines the elderly as vulnerable is 70 or older.
9. Statutorily defined aggravators include: murder committed in the course of another felony; offense for pecuniary gain; offense that was heinous, atrocious, or unusually cruel; offense was part of a course of conduct involving other crimes of violence against the victim and others; defendant created a risk of death to more than one person; defendant had prior felony convictions involving violence; offense occurred while defendant was incarcerated; defendant has prior conviction for a capital offense; offense was committed to avoid arrest or to escape from custody; offense was committed to disrupt or hinder lawful exercise of a governmental function; the murder was of a law enforcement officer.
10. Statutorily defined mitigators include: no significant history of prior criminal activity; defendant under influence of mental or emotional disturbance; defendant was very young or very old; defendant had impaired ability to appreciate criminality of his/her conduct; defendant was a minor accomplice; defendant acted under duress; defendant aided in apprehension of another felon; victim was a voluntary participant in events leading to the murder or consented to the offense. The defense may also submit unlimited non-statutory mitigators.
11. ‘The coder’s subjective interpretation as to whether the crime or resulting crime scene was particularly horrendous’ (North Carolina Capital Sentencing Project (NCCSP), Citation2013). By ‘particularly horrendous,’ the coders examined the case files for evidence of torturous death, lingering death, multiple wounds/injuries, especially bloody death, and other similar characteristics.
12. N = 84 cases deleted with Asian, Indian, Hispanic, or other-race defendants.
13. N = 41 additional cases deleted with Asian, Indian, Hispanic, or other-race victims.
14. The confession may have been to law enforcement or to others who testified at trial to that effect. The latter could include accomplices, friends, or jailhouse ‘snitches.’ Also, the individual may have later recanted the confession, but if evidence of a confession is presented in court, the variable is coded ‘yes.’.
15. Beyond simply confessing, did the defendant cooperate with police in a way that led to a solving of the crime?
16. While the U.S. Supreme Court ruled that capital juries may consider non-statutory mitigating factors in Lockett v. Ohio, the state of North Carolina subsumed these non-statutory mitigators into its ‘catch-all’ mitigator category. The McKoy decision required that these non-statutory mitigators not only be considered, but also individually recorded.
17. Collinearity diagnostics revealed no evidence of problematic collinearity with these data.
18. Because our data constitute the population of capital trials in the State of North Carolina (1977–2009), we do not rely upon statistical significance when interpreting our findings. Furthermore, should the reader consider this population, instead, as a sample of the universe of capital jury decisions, their nonrandomized nature precludes probabilistic inferential interpretations, leading to a substantial likelihood of misinterpretation (Hirschi & Selvin, Citation1973). The authors focus explicitly on direction and magnitude of effects. Inferential statistics are, however, reported in Table for the readers to employ as they see fit.