BACK FROM THE DEAD: TRACKING THE
PROGRESS OF KENTUCKY’S FURMAN-
COMMUTED DEATH ROW
POPULATION*
GENNARO F. VITO
University of Louisville
DEBORAH G. WILSON
University of Louisville
After the U.S. Supreme Court decision in Furman v. Georgia (1972),
23 inmates at the Kentucky State Penitentiary in Eddyville had their capi-
tal sentences commuted. This study tracks 17 of these former death row
inmates, who have been paroled, and reports on their current legal status.
‘The analysis reveals that the recidivism rate of these parolees was not dra-
matically high and that none of the parolees committed another homicide.
In fact, those inmates who did recidivate had the most extensive prior
records and had committed homicides of police officers during an armed
robbery.
According to the NAACP Legal Defense Fund (1987), 1,874
prisoners nationwide were being held under a sentence of death as
* Paper presented at the “Moral Imperative: Issues Surrounding the Use of
the Death Penalty” panel of the 1986 meeting of the Southern Association of Crimi-
nal Justice Edueators.
This paper is drawn from a report published by the Kentucky Criminal Justice
Statistical Analysis Center (SAC). The SAC is funded by the Bureau of Justice
Statistics, U.S. Department of Justice, Grant No. 84-BJ-CX-0013. Points of view or
opinions stated in this document are those of the authors and do not necessarily
represent the official position or policies of the Bureau of Justice Statistics, the
Kentucky Office of the Attorney General, the Kentucky Corrections Cabinet, or
the University of Louisville.
‘The authors wish to express their appreciation to the following persons, who
offered important suggestions in the development of this paper: The NAACP Legal
Defense Fund, David Baldus, Kevin McNally of the Kentucky Department of Pub-
lic Advocacy, Leslie Bienin of the New Jersey Department of the Public Advocate,
Timothy J. Flanagan, Ken Ayers, Finn-Aage Esbesen, Francis T. Cullen, editor, and
anonymous reviewers for Justice Quarterly. We also express our heartfelt thanks
to an anonymous corrections professional who suggested this research topic, John
Sternberg (who collected data diligently and carefully), Betty Lou Vaughn and her
staff in the Kentucky Corrections Cabinet, Robert Crouse of the Southern Police
Institute, Major Edward Mercer of the Louisville Police Department, and the fol-
lowing Urban Studies Center staff members who provided consultation and support
services to the project: Knowlton Johnson, Jack Ellis, Jerry Hutchins, Vernon
Smith, Nancy Roseberry, Lisa Johnson, Monica Bowles, and Elizabeth Jones.
JUSTICE QUARTERLY, Vol. 5 No. 1, March 1988
© 1988 Academy of Criminal Justice Sciences
102 BACK FROM THE DEAD
of March 1, 1987. Since 1976, when the U.S. Supreme Court af-
firmed the death penalty in the Gregg decision, 69 prisoners have
been executed. Yet the growing use of capital punishment and the
method by which it is applied are central concerns of policy re-
searchers (see Cheatwood 1985). Advocates of capital punishment
often justify its application on the grounds that it effects crime
control both as a general deterrent and as a sanction which inca-
pacitates offenders and prevents their future criminality. Of
course, death represents the ultimate incapacitative sanction. As
van den Haag (1983:55) writes, “Death incapacitates totally and
permanently, imprisonment partially and temporarily.”
Although research on the deterrent effect of the death penalty
is widespread (Bowers and Pierce 1980; Ehrlich 1975; Klein, Forst,
and Filatov 1982), similar research on the extent to which capital
punishment reduces victimization (particularly future homicides)
by preventing the release of murderers into society has been lim-
ited to incarcerated murderers who commit additional homicides
in prison and to analyses of homicide offenders (not sentenced to
death) who eventually were released on parole. Sellin (1982), for
example, examined homicides in U.S. prisons in 1964-65 and con-
cluded that inmates convicted of homicide were responsible for
only 16 percent of these offenses; other studies demonstrated that
capital punishment states had the highest rates of prison homi-
cides (see also Wolfson 1982). Sellin (1982:120) also reviewed sev-
eral studies of murderers on parole from 1930-73 and concluded
that “in general capital murderers rarely commit a homicide while
on parole.”
The present study attempts to extend these analyses into un-
charted ground by examining the recidivism rates of paroled Ken-
tucky death row inmates whose sentences were commuted as a
result of Furman v. Georgia. Our central research question is this:
“Does the recidivism rate of this Furman group justify the use of
capital punishment to achieve societal protection?”
BACKGROUND: FURMAN v. GEORGIA
On June 29, 1972, the U.S. Supreme Court declared that the
imposition of the death penalty was unconstitutional. In a 5 to 4
decision in which each of the justices in the majority wrote his
own opinion (“an unusually splintered majority”: Bedau 1982:254),
the manner in which the death penalty had been administered be-
came the central issue. Alternatively, the justices referred to exe-
cution as “little more than a lottery system” and drew a chilling
VITO AND WILSON 103
analogy between capital punishment and being “struck by light-
ning.” After the Furman decision, the sentences of 23 Kentucky
death row inmates were commuted to life imprisonment.
Kentucky was not the only state in which such commutations
took place. Greenberg (1982:915) wrote that as a result of the
Furman decision, the capital sentences of 629 persons on death row
were commuted. According to a list provided by the NAACP
Legal Defense Fund (1987), the death sentences of 613 offenders
nationwide were commuted as a result of the Furman decision. A
complete list of these cases by state (ranked by number of offend-
ers with commuted sentences) is provided in Table 1.
On the NAACP list the Kentucky death row ranks ninth with
21 cases, if we allow for ties. It is necessary here to clarify some
points about these figures and those produced in the current study.
Our analysis of files maintained by the Kentucky Corrections Cab-
inet revealed that the Furman death row consisted of 23 inmates.
Not all of these offenders, however, were included in the current
study; five members of the cohort were still incarcerated at the
conclusion of the analysis. One member of this group “refused”
the commutation of his sentence on the grounds that he was not
guilty; therefore, in his mind, the acceptance of the commutation
would constitute an admission of guilty (Lexington Herald-Leader
1987:B1). He has been housed continuously on death row since
1960 and still resides there. A second offender had committed rape
during the original homicide. At the time of his offense, the Ken-
tucky criminal code (KRS 435.090) provided for a sentence of
death, a sentence of life without privilege of parole, or a sentence
of ten to 20 years for rape. In this case the offender was sentenced
to death for the murder and to life without parole for the rape.
Although his death sentence was commuted as a result of the
Furman decision, the sentence of life without parole stood, and he
is still incarcerated in Kentucky. This sentencing procedure was
determined to be constitutionally acceptable (see Moore v. Cowan
and Green v. Commonwealth), but it was excluded from Kentucky
criminal law when the revised statutes were adopted on January 1,
1975. A third member of the group presently is incarcerated in an-
other state for a homicide that he committed in that state only
weeks before his Kentucky offense. Two additional inmates, who
had killed police officers, had yet to be released on parole. A sixth
inmate was progressing toward parole (his release date had been
set) when he was killed in.an accident. He had been making re-
pairs in the boiler room of the prison when.one of the boilers rup-
tured, and he died of severe burns shortly thereafter.
104 BACK FROM THE DEAD
Table 1. Number of Death Row Inmates in the United States
Whose Sentences Were Commuted as a Result of
Furman v. Georgia.
Number of Number of Total
State Murder Cases Rape Cases Number
Florida 73 29 102
Ohio 63 0 63
Texas 42 10 52
Louisiana 36 12 48
Georgia 29 14 43
Alabama 26 5 31
Illinois 31 0 31
Pennsylvania 25 0 25
Maryland 18 5 23
Massachusetts 23 0 23
Kentucky 21 0 21
Missouri 16 0 16
Tennessee 12 4 16
Oklahoma 15 0 15
Virginia 12 0 2
North Carolina 9 2 i
South Carolina 10 1 11
Washington 10 0 10
Indiana 9 0 9
Mississippi 9 0 9
Nevada 8 0 8
Arkansas 6 0 6
New York 5 0 5
Utah 5 0 5
Connecticut 4 0 4
Delaware 3 0 3
District of Columbia 3 0 3
Colorado 2 0 2
Kansas 2 0 2
Nebraska 2 0 2
Grand Totals 528 85* 613
* The rape case total includes two offenders convicted of armed robbery.
METHODS
The data for the present study were obtained through a study
conducted under the auspices of the Kentucky Criminal Justice
Statistical Analysis Center (SAC). The Kentucky SAC is housed
in the office of the Attorney General and is operated by the Urban
Studies Center of the College of Urban and Public Affairs at the
University of Louisville in cooperation with the U.S. Bureau of
VITO AND WILSON 105
Justice Statistics. All information concerning the 23 Furman-com-
muted death row offenders in Kentucky was obtained from institu-
tional files. Information concerning the offense itself and
background data on the offenders were collected from the
presentence investigation report.
One of the most obvious limitations of this analysis is the rela-
tively small size of the paroled death row group (n = 17). Yet, as
Travis (1983) noted, the case study approach in criminal justice re-
search is capable of generating great quantities of descriptive infor-
mation which can be used by policy makers. It is especially
valuable in times of rapid change because it allows an immediate
response to an historical event or a natural experiment—the effect
of a judicial order on the operations of a prison or a change in sen-
tencing policy. This type of research design is exemplified well by
the impact of the Furman v. Georgia decision on the death row
population in this country and by the subsequent performance of
these offenders after they were paroled.
RESEARCH FINDINGS
The central research question of this study is: “How did the
paroled death row inmates fare in terms of recidivism?” Seven-
teen members of the initial group were released on parole after
their death sentences were commuted. The mean time served be-
tween commutation and parole was 77 months; the mean total
time served by this group was 123 months; the average length of
the follow-up period was 42 months. All the paroled offenders left
the state either immediately or eventually.
Of this group, one of the offenders has obtained final release
from supervision with a full restoration of civil rights. This of-
fender also received commendations from officials in his adopted
state, including one from the attorney general, praising his partici-
pation as an expert witness in criminal prosecutions. Eleven other
parolees, including two who are described more fully below, even-
tually were placed on “inactive supervision”: although their period
of parole supervision technically had not lapsed, these offenders
had performed well enough on parole that continued close supervi-
sion was not warranted. Technically, however, these offenders are
still defined legally as under parole supervision.
Three of the paroled offenders committed technical violations
of the conditions of their parole supervision period. Two of these
offenders were arrested, and one of these arrests resulted in a con-
viction. All three offenders received revocation hearings; two of-
fenders returned to prison and one was granted leniency and
another chance to adhere to the conditions of parole supervision.
106 BACK FROM THE DEAD
Table 2. Recidivism Patterns among Kentucky’s Furman-
Commuted Death Row Population.
Parole Outcome Total Number Percent
Total paroled 17 —
Obtained final release 1 6
Placed on inactive supervision 11* 59
Technical violators 18
Arrested.
New conviction
Hearing held:
Returned to prison
Granted leniency
New crime:
Arrested
Convicted
Jailed
Imprisoned
Type of crime:
Robbery
Burglary
Drug possession
Total arrest rate
Total conviction rate
Total jail reincarceration rate
Total prison reincarceration rate
OR RRR HN WH Dw
BRD
Ba Bk
* The recidivist who was jailed for a drug possession offense and one of the
parole violators eventually performed well enough to be placed on inactive
supervision.
One of the parole violators and one of the recidivists eventually
performed well enough to be placed on inactive supervision, bring-
ing the total on inactive supervision to 11. (See Table 2.)
Finally, four offenders committed new crimes after their re-
lease from prison. All four were arrested and convicted: one was
sentenced to jail and three were returned to prison on a new con-
viction. None of the offenders committed murder again, however;
armed robbery was the most serious offense listed as a new crime.
As a group, the Furman-commuted death row parolees had an
arrest rate of 35 percent, a conviction rate of 29 percent, a jail in-
carceration rate of six percent, and a prison incarceration rate of
29 percent. These results are summarized in Table 2.
Of course, these data do not present a complete picture of the
on-parole performance of these inmates. In addition to the parolee
VITO AND WILSON 107
whose civil rights were restored, several others also enjoyed suc-
cess. Parole supervision progress reports revealed that these pa-
rolees evidenced positive adjustment to society by “kicking” their
drug/alcohol abuse problems (two), obtaining an education (one),
reestablishing family ties (three), and finding jobs (the entire
group). In short, they became productive citizens.
Yet we must also keep in mind the background to the parolees
reformed performance, no matter how commendable that per-
formance may be. These parolees originally were responsible for
the deaths of 14 people, including six police officers. In fact, the
on-parole performance of the “cop killers” (see Allen, Mannle,
Lewis, and Carlson 1979) sheds further light on the recidivism
rates for the total group of paroled offenders. Nine of the Furman
death row offenders had murdered or had participated in the mur-
der of six police officers during an armed robbery.1 After their
sentences were commuted, seven members of this subgroup even-
tually were paroled. The two “cop killers” who were still incarcer-
ated when this research ended had killed two officers. Therefore
the paroled subgroup of offenders was responsible for the deaths
of four police officers. Analysis of the parole performance data re-
vealed that two of the offenders were placed on inactive supervi-
sion and that none violated the conditions of their supervision
period. Three committed new crimes, however, and eventually
were convicted and reincarcerated.
Notably, these recidivists accounted for most of the total ar-
rests and convictions for the entire group of recidivists and for all
the offenders who were imprisoned or who committed robbery or
burglary. The offender who was incarcerated for burglary was
also convicted as an habitual offender. As this account indicates,
the offenders who had murdered police officers, and who eventu-
ally were paroled, were responsible for the bulk of the new crimes
committed by the entire group of Furman-commuted death row
parolees. Minus this group, the recidivism rate of the death row
group would be almost nil. On the other hand, we must also point
out that four of the seven paroled “cop killers” have not engaged
in any further officially known criminal activity. Therefore we
cannot draw a definitive conclusion concerning the “risk potential”
of this group.
1 It is assumed most often that the most dangerous call for a police officer to
answer is the domestic disturbance. Yet, research by Margarita (1980) on homicides
of police officers in New York City between 1844 and 1978 revealed that officers
were most likely to be slain in response to a robbery call (68/239, or 28.5%).
Although our data do not address this question directly, the presence of so many
killers of police on death row tends to support this conclusion.
108 BACK FROM THE DEAD
Prior record is another crucial variable which is related to risk
of recidivism. The inmates who recidivated had a more extensive
prior record for previous arrests and convictions than did the
nonrecidivists. They average approximately two prior arrests, one
prior conviction, one prior incarceration, and two years’ previous
imprisonment, and demonstrated a more severe prior record of vi-
olent crime. The poor parole performance of these recidivists is
not surprising, in view of the severity of their prior records.
Table 3. Comparison of Parole Recidivism Rates for Murderers:
Kentucky’s Furman-Commuted Death Row and Other
Studies.
Outcome Failure
Study Definition Rate Kentucky
Donnelly & Bala (1984) Recommitted for 27.3% 29%
new offense or
technical
violation
Bedau (1982) Conviction:
New homicide 0.6 0
Other felony 3.3 18
Conviction:
New homicide 0.3 0
Other felony 15 18
Reincarcerated:
New offense 11 18
Violation 5.5 n
Wallerstedt (1984) Reincarcerated 22.6 29
for new offense
or violation
Table 3 compares the recidivism figures from a study of Ken-
tucky’s Furman-commuted paroled offenders to the figures from
similar studies conducted in other states during several periods.
Although it is important to compare the recidivism rates of the
group under consideration to those of similarly situated offenders,
this comparison is far from perfect. The Furman-commuted parol-
ees represent a unique group, which experienced the rigors of
death row incarceration. Although they may have been similar
before this experience, one can argue that they are not comparable
to convicted homicide offenders who received some other sort of
sentence (such as life) and who eventually were released on pa-
role. In addition, these studies consist of offenders released during
different periods with follow-up periods of varying length.
The Donnelly and Bala (1984) study was based on a five-year
follow-up of 66 murderers released on parole from New York
VITO AND WILSON 109
State prisons in 1977. The compilations of Bedau (1982) were
based on reports of recidivism rates of released convicted murder-
ers in 12 states during different years. The first group of figures
was based on the following states and periods: California (1945-
54), Connecticut (1947-60), Georgia (1943-65 and 1973-76), Mary-
land (1936-51), Massachusetts (1941-50 and 1957-66), Michigan
(1938-72), New Jersey (1907-60), New York (1957-61 and 1945-61),
Ohio (1945-60), Oregon (1939-64), Pennsylvania (1932-69), and
Rhode Island (1915-58). Bedau’s second compilation was based on
reports of nationwide recidivism rates during the first year of re-
lease for convicted murderers paroled in from 1965 to 75; the third
compilation focuses on male convicted murderers released from
1971 to 74. Finally, Wallerstedt (1984) reports the median rein-
carceration rate for homicide recidivists from five states: Michigan
(1978), Nebraska (1980), New York (1976), Oregon (1978), and
Rhode Island (1979).
Table 3 indicates that the Furman-commuted offenders did as
well as those offenders listed in Donnelly and Bala (1984) and in
the Wallerstedt Bureau of Justice Statistics bulletin (1984), and
that they fared somewhat worse than the offenders listed in
Bedau’s figures (1982). The time of the follow-up periods in the
New York State and Wallerstedt reports is related more closely to
the figures in the current study, while the figures cited by Bedau
are taken from a much earlier period and extend over a longer
time. Thus the two former reports may serve as the best points of
comparison. It appears that the Furman-commuted group did not
present a serious threat to public safety after their release on
parole.”
There is one other point of comparison: the type of offense
committed most often by homicide recidivists. The Bedau figures
indicate that homicide recidivists engaged most often in “other
felonies” after their release. The Oregon data reported by Waller-
stedt reveal that this group of offenders was most likely to commit
violations of their conditions of supervision. Neither pattern
matches exactly the experience of the Kentucky Furman group,
which was divided more evenly between violations (18%) and new
crimes of robbery, burglary, and drug possession (11%).
? At least one member of the nationwide Furman-commuted group, how-
ever, committed a second murder after parole. Urofsky (1984:569-570) reports that
Robert Lee Massie repeated his original offense of murder and robbery and again
was sentenced to death in 1979.
110 BACK FROM THE DEAD
CONCLUSIONS
In view of the small size of this cohort, it is difficult to draw
any incontrovertible conclusions about the Furman-commuted pa-
rolees. Yet it is clear that the on-parole performance of this group
was not especially poor. In fact, no group member committed a
second homicide. The significant recidivist activity came from a
single, identifiable, and perhaps unsurprising source: offenders
who originally had killed police officers during an armed robbery
and who had a more extensive criminal history of arrests and con-
victions than did the nonrecidivists.
In light of the fact that this group was identified by their
sentences as posing the ultimate threat to society and that they
“came back from the dead,” their on-parole performance seems
impressive. If not for the Furman decision, this group probably
would have been executed. Their recidivism rate is relatively low;
certainly it is no higher than that of paroled murderers elsewhere.
The on-parole performance of the Furman-commuted group
demonstrates that they are not a clear-cut risk to society. In view
of the nature of their original crimes, this performance is even
more surprising. The homicides committed by this group were not
“heat of passion” crimes but just the opposite; most of the murders
were committed against strangers during an armed robbery. As a
whole, the Furman-commuted group committed the type of homi-
cide that may be feared most by the public because the victim is a
stranger (often a police officer). The most reliable statement we
can make is that further research on the remaining 528 Furman-
commuted death row inmates (convicted of murder) is necessary.
We plan soon to undertake such a study nationwide.
REFERENCES
Allen, Harry E., Henry W. Mannle, Peter W. Lewis, and Eric W. Carlson (1979)
“Law Enforcement Officers Slain and ‘Cop Killers’ on Florida’s Death Row.”
Journal of Contemporary Criminal Justice 1(1):1-2,13.
Bedau, Hugo A. (1982) “Recidivism, Parole and Deterrence.” In Hugo A. Bedau
(ed.), The Death Penalty in America. New York: Oxford University Press, pp.
173-80.
Bowers, William J. and Glenn L. Pierce (1980) “Deterrence or Brutalization: What
Is the Effect of Executions?” Crime and Delinquency 26:453-84.
Cheatwood, Derral (1985) “Capital Punishment and Corrections: Is There an Im-
pending Crisis?” Crime and Delinquency 31:461-79.
Donnelly, Henry C. and Gerald Bala (1984) 1977 Releases: Five Year Post Release
Follow-up. Albany, NY: Department of Correctional Services.
Ehrlich, Isaac (1975) “The Deterrent Effect of Capital Punishment: A Question of
Life and Death.” American Economic Review 65:397-41T.
Greenberg, Jack (1982) “Capital Punishment as a System.” The Yale Law Journal
91:908-36.
Klein, Lawrence R., Brian Forst, and Victor Filatov (1982) “The Deterrent Effect of
Capital Punishment: An Assessment of the Evidence.” In Hugo A. Bedau
VITO AND WILSON pane
(ed.), The Death Penalty in America. New York: Oxford University Press, pp.
138-59.
“ ‘Legal Web’ Keeps Inmate on Death Row,” Lexington Herald-Leader, April 13,
1987, p. Bl.
Margarita, Mona (1980) “Killing the Police: Myths and Motives.” The Annals of
the American Academy of Political and Social Science 45(2):63-71.
NAACP Legal Defense Fund (1987) Death Row U.S.A. New York.
Sellin, Thorsten (1982) The Penalty of Death, Beverly Hills: Sage.
Travis, III, Lawrence F. (1983) “The Case Study in Criminal Justice Research: Ap-
plications to Policy Analysis.” Criminal Justice Review 8:46-51.
Urofsky, Melvin I. (1984) “A Right to Die: Termination of Appeal for Condemned
Prisoners.” Journal of Criminal Law and Criminology 15:553-82.
van den Haag, Ernest and John P, Conrad (1983) The Death Penalty: A Debate.
New York: Plenum,
Wallerstedt, John F. (1984) Bureau of Justice Statistics Special Report: Returning
to Prison. Washington, DC: U.S. Department of Justice.
Wolfson, Wendy Phillips (1982) “The Deterrent Effect of the Death Penalty upon
Prison Murder.” In Hugo A. Bedau (ed.), The Death Penalty in America.
New York: Oxford University Press, pp. 159-73.
CASES
Furman v. Georgia 408 U.S. 238 (1972)
Green v. Commonwealth Ky., 556 S.W.2d 684 1977)
Gregg v. Georgia 428 U.S. 153 (1976)
Moore v. Commonwealth 556 F.2d 1298 (1977)
