514 10TH S TREET NW, S UITE 1000
WASHINGTON, DC 20004
TEL: 202.628.0871 • FAX: 202.628.1091
S TAFF@S ENTENCINGPROJECT.ORG
WWW.S ENTENCINGPROJECT.ORG

THE MEANING OF “LIFE”:
LONG PRISON SENTENCES IN CONTEXT

BY
MARC MAUER, RYAN S. KING, AND MALCOLM C. YOUNG
MAY 2004

This report was written by Marc Mauer, Ryan King, and Malcolm C. Young, Assistant Director,
Research Associate, and Executive Director respectively, of The Sentencing Project. Research
assistance was provided by Michael Coyle. The authors wish to thank the following for
providing information and insight into the issues covered in the report: John Anderson, William
DiMascio, Elizabeth Leonard, Barbara Levine, Sue Osthoff, David Saxner, and Mara Taub. And
particular appreciation to Douglas Dennis and staff of The Angolite, who have produced strong
journalism on these issues for many years.
The Sentencing Project is a national non-profit organization engaged in research and advocacy
on criminal justice policy issues. Funding for this project was supported by grants from the
JEHT Foundation, John D. and Catherine T. MacArthur Foundation, Open Society Institute, and
individual donors.
Copyright © 2004 by The Sentencing Project. Reproduction of this document in full or part in
print or electronic format only by permission of The Sentencing Project.
For further information:
Malcolm C. Young
The Sentencing Project
514 10th St. NW
Suite 1000
Washington, D.C. 20004
(202) 628-0871
www.sentencingproject.org

INTRODUCTION
There’s a popular misconception that life in prison doesn’t mean all of one’s
natural life. In just the last year, there are 21 Georgia lifers who are no longer
around to tell you otherwise. If they could, they’d let you know that parole for a
life sentence is a rare commodity.1
-- Georgia State Board of Pardons and Parole
A crime prevention policy which accepts keeping a prisoner for life even if he is
no longer a danger to society would be compatible neither with modern principles
on the treatment of prisoners during the execution of their sentence nor with the
idea of the reintegration of offenders into society.2
-- Council of Europe
Over the past three decades the political climate in the United States has increasingly embraced
“get tough” policies as the primary focus of a crime control strategy. These policies have been
wide-ranging in their impact, and include such features as high levels of drug law enforcement,
greater reliance on determinate sentencing, and most significantly, a vastly expanded use of
imprisonment. Along with these changes has come a virtual abandonment of the principles of
rehabilitation that had been central to the nation’s correctional philosophy, even if not always
implemented to a significant degree, as recently as the 1960s.
In the past several years there has been a significant shift both in public discussion and policy
attention to the use of imprisonment. At the state level, many policymakers are now advocating
changes in sentencing policy and practice that reflect new thinking and options for lower-level
drug offenders in particular. These proposals emphasize shorter prison terms and/or diversion to
treatment programs for cases in which substance abuse is an underlying contributor to crime, and
for which it is believed that these approaches will provide greater public safety benefits.
In contrast to these developments, a variety of policy changes beginning in the 1970s and
increasing in recent decades have set in motion a movement to extend considerably the length of
time that other offenders spend in prison. These changes include such policies as mandatory
sentencing, “truth in sentencing,” and cutbacks in parole release. While many of these initiatives
apply primarily to persons convicted of a violent offense, in some cases they mandate long-term
incarceration even for persons convicted of property or drug offenses.
Foremost among the changes affecting the prison population in recent years are laws affecting
“lifers,” those persons receiving a sentence that may result in an offender never being released
from prison. 3 Policy considerations for persons sentenced to life are very different than for
offenders who appear far less threatening, such as low-level drug offenders.
For violent
1

News release, Georgia State Board of Pardons and Paroles, “More Violent-Crime Lifers Die in Prison Than Are
Paroled,” June 1, 1998.
2
“Life Imprisonment,” United Nations, 1995, p.5.
3
The term “life sentence” has been used in a variety of ways over time and consequently there is much public
confusion regarding its meaning. While the intuitive definition of a life sentence is a prison term for the remainder
of one’s natural life, in fact the term also includes various indeterminate sentences, or sentences whose length can be
reduced by commutation, parole, or pardon.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

2

_____________________________________________________________________________________________

offenders who have taken lives or who pose a serious threat to public safety, incapacitation as a
means of assuring public safety is a legitimate and compelling concern at sentencing. In
addition, under current laws and sentencing philosophies, lifetime incarceration is deemed an
appropriate punishment for the harm done by serious violent offenders.
However, the issue of life sentences is far more complex and cannot be regarded as merely strict
sentencing for a deserving population of serious violent offenders. A closer examination of the
number of people serving life sentences, their offense characteristics, and the judicial process by
which their sentences were imposed, challenges many of the assumptions about the composition
of the lifer population. Among those serving life are persons who themselves have not
committed violent acts and others whose life circumstances suggest they are more vulnerable
than violent. A selection of such cases is highlighted throughout this report.
Most notable in this regard is California’s “three strikes and you’re out” law, under which any
felony following two previous “strikes” can result in a life term in prison. A recent high-profile
example is the case of Leandro Andrade, whose third strike involved thefts of children’s
videotapes worth $153 and intended as Christmas gifts for his nieces. In affirming the
conviction, the U.S. Supreme Court upheld the California law in 2003 and he is now serving a
sentence of 50 years to Life.4
Santos Reyes
In 2003 a federal appeals court upheld a “three strikes” sentence of 26 years to life for Santos Reyes,
whose third strike in California involved trying to take the written portion of a driver’s license test for
his illiterate cousin. The court ruled that Reyes’ claim of cruel and unusual punishment had been
foreclosed by a previous ruling on the law by the U.S. Supreme Court. Reyes admitted his perjury in
filling out the license application, stating that his cousin needed the license in order to work as a
roofer. The conviction followed two previous offenses, one for a juvenile burglary conviction in 1981
and another for an adult robbery conviction in 1987. Reyes had been offered a four-year prison term
if he pled guilty, but chose to go to trial, believing he could demonstrate that he had not understood
what constituted perjury when he took the exam. At the time of his sentencing in 1998, Reyes was
married and had children ages 1 and 3.

California’s “three strikes” law is not an isolated example. Other policy changes in recent years
have led to life sentences being imposed on a significant number of people for whom such
sentences may not have been intended or for whom they are not necessarily appropriate. Further,
opportunities which previously existed to review and modify life sentences which proved longer
than necessary have been reduced or eliminated in many states. These changes have come about
at a time of increased concern about the possibility of unjust convictions and sentencing, as
dramatically evidenced in the imposition of the death penalty.

4

Lockyer v. Andrade, 538 U.S. 63 (2003), decided with Ewing v. California, 538 U.S. 11 (2003).

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

3

_____________________________________________________________________________________________

In this report we assess the dramatic increase in the imposition of life sentences in the context of
incapacitation and public safety, fiscal costs, and the sentencing goal of punishment, including
the implications for both victims and offenders. Major findings of the report include:
•
•
•
•
•
•
•
•
•

One of every 11 (9.4%) offenders in state/federal prison – 127,677 persons – is now serving a
life sentence.
Of the lifers in prison, one in four (26.3%) is serving a sentence of life without parole, having
increased from one in six (17.8%) in 1992.
The number of lifers in prison rose by 83% from 69,845 in 1992 to 127,677 in 2003.
Time to be served for lifers admitted to prison increased by 37% from 1991 to 1997, rising
from 21.2 years to 29 years.
In six states -- Illinois, Iowa, Louisiana, Maine, Pennsylvania, and South Dakota -- all life
sentences are imposed without the possibility of parole.
Seven states -- Alabama, California, Florida, Illinois, Louisiana, Michigan, and Pennsylvania
-- have more than 1,000 prisoners each serving sentences of life without parole.
The increase in prison time for lifers is a result of changes in state policy and not continuous
increases in violent crime.
Four of every five (79.4%) lifers released in 1994 had no arrests for a new crime in the three
years after their release. This compares to an arrest-free rate of just one-third (32.5%) for all
offenders released from prison.
Imposing a life sentence carries with it a potential cost to taxpayers of $1 million.

Life sentences are of great consequence to the individuals who receive these sentences and to the
society that imposes them. The findings in this report demonstrate that there are today a
dramatically rising number of lifers and that they are serving increasingly longer terms of
incarceration. This growth is linked to policy changes, not increases in crime rates. The analysis
and discussion in this report raise serious questions about the fairness and reliability of the
judicial process that leads to the imposition of life sentences. As a result, lifers include those for
whom the length of sentence is either unjust or inappropriate. Life sentences in many cases
represent a misuse of limited correctional resources. This report challenges the supposition that
all of these sentences are necessary or effective in advancing public safety. We conclude with a
set of recommendations for change in law, policy and practice which would, if adopted, address
the principle deficiencies in the sentencing of lifers to prison.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

4

_____________________________________________________________________________________________

AN OVERVIEW OF LIFE S ENTENCES
The life sentence in the United States was initially created at the time of the founding of the
nation, and was advanced in large part as a reaction to the frequent imposition of the death
penalty in the Colonial period. The life sentence was developed as an indeterminate sentence;
that is, as a term of imprisonment without a prescribed duration at the time of sentencing.
Sentences such as “15 years to life” denote a minimum length of time to be served, but the actual
length of the sentence to be served is undetermined and is decided by a parole board or other
administrative body based on factors such as the inmate’s behavior while incarcerated.
Indeterminate sentencing is based on the premise that in the face of good conduct and evidence
of rehabilitative efforts while incarcerated (participation in counseling or drug programming,
obtaining education or work skills), offenders can and should be released from prison.
Historically, life sentences have also been used to impose a sentence that is intended to last the
offender’s natural life, or “life without the possibility of parole.” However, even this seemingly
clear distinction can sometimes be blurred, since on rare occasions an offender serving life
without parole may be granted a pardon or commutation for unusual circumstances.
The changes in sentencing policy and practice that will be documented in this report mirror those
of the criminal justice system generally over the past several decades. Much of this change has
involved a shift in decisionmaking at various stages of the system. Prior to the 1970s, a system
of indeterminate sentencing had been the norm nationally for nearly a century. This structure
was characterized by broad sentencing ranges established by legislative bodies, with great
discretion given to judges to establish a sentence length within those ranges. In an ideal
situation, such a system would allow for careful distinctions to be made among various
combinations of offenders and offenses, informed by a consideration of individual
circumstances. In practice, judicial discretion sometimes resulted in vastly different sentences
being imposed on similarly situated offenders.
Similar dynamics and problems pertained to decisionmaking regarding release from prison as
well. Parole boards generally maintained broad discretion, and were often critiqued for decisions
that were viewed as either too lenient or too harsh in particular cases.
The critique of indeterminate sentencing led to policy changes in sentencing and parole that now
impose a much greater level of restriction on judges and parole officials. At an extreme, such
policies as mandatory sentencing take away virtually all discretion from judges at the time of
sentencing. In other cases, legislative and executive mandated changes have resulted in
increased numbers of offenders being sentenced to prison and/or increases in length of time
served in prison.
For lifers, the changes in public policy have been most significant in terms of the length of time
to be served in prison. Given the serious nature of offenses for which most lifers have been
convicted, virtually all would be sentenced to prison under any sentencing system. But, as will
be seen in this report, the determination of how much time should be served in prison is a
challenging policy issue. This raises considerations of public safety, concern for victims, fiscal
costs, and prospects for rehabilitation. The findings of this report suggest that the movement
toward reduced discretion in such cases has resulted in lengthier periods of incarceration than are
necessary to achieve public safety goals.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

5

_____________________________________________________________________________________________

STATE P OLICY CHANGES
Policymaker action intended to increase the duration of incarceration by lengthening sentences
and reducing the use of discretionary parole has played a large role in the growing lifer
population. This has come about in varying ways over the past twenty years.
Life without parole has always been a sentencing option, but the frequency with which this has
been used has increased in recent decades. In many instances, this has been a reflection of the
use of more punitive sentencing policies in general, but in some cases it also results from the
increased use of life without parole as an alternative to the death penalty.
Many states have moved toward more determinate sentencing structures that have significantly
affected consideration for lifers’ release. This has often resulted in much lengthier periods of
imprisonment before parole eligibility. And in some states, while parole remains a possibility,
executive restrictions have diminished considerably the number of persons actually gaining
release.
The growing number of lifers in prison is not primarily a by-product of increasing crime rates.
In looking at the growth of the lifer population documented in this report, first doubling from
1984 to 1992 and then increasing by 83% from 1992 to 2003, we see two distinct crime trends.
The rate of violent crime (more relevant for lifers than total crime) rose by 40% from 1984 to
1992, but declined by 35% from 1992 to 2002 (2003 figures not yet available). While the rate of
violent crime is not unrelated to the number of lifers, clearly factors other than crime rates have
contributed to those trends.
These changes can be seen in the experience of a number of states in recent years:
California
In California, the parole board is obligated to enter the cases of lifers entering the system on a
matrix (noting the minimum and maximum years to be served), and to periodically review each
prisoner’s progress. However, this practice changed under Governor Gray Davis, who refused to
release lifers convicted of murder.
In one California case, the “no parole” policy for murderers led to a constitutional challenge.
Robert Rosenkrantz had been recommended for parole by the state parole board, and was further
supported for parole by the judge who sentenced him, members of the victim’s family, and both
the Los Angeles Superior Court and the Court of Appeals, which ruled that the offense was not
sufficient to deny him parole.5 Rosenkrantz was considered a model prisoner, had completed
therapy and become a computer expert since his imprisonment, and had received several job
offers. Yet his suit was denied by the California Supreme Court, which ruled that the governor
had the right to reinterpret a case beyond the findings of a jury. 6

5

“ACLU and Religious Groups Bring Constitutional Challenge to CA Governor’s ‘No Parole’ Policy.”
http://www.aclu.org/CriminalJustice.cfm?ID=7222&c=52&Type=s&insearch=lifers, April 18, 2001.
6
Mentor, Marvin. “California Governor has Carte Blanche in Denying Lifer Paroles.” Prison Legal News , 14 (7),
30-1, July 2003.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

6

_____________________________________________________________________________________________

Overall, Governor Davis turned down all but 8 of the 294 paroles in murder cases approved by
the parole board (not all of whom received a life sentence). During his first two months in
office, new Governor Arnold Schwarzenegger approved parole release for eight persons
convicted of murder.
Illinois
In Illinois, in 1978, indeterminate sentencing for lifers was changed so that all life sentences did
not offer the possibility of parole; whereas previously a life sentence meant parole eligibility
after eleven years, today there is no parole possibility for any life sentence. Over the past
decade, there are no known cases in which a person with a life sentence has received clemency.
Louisiana
In Louisiana, until the 1970s, a life sentence commonly translated into a ten-year prison term.
By the 1990s a life sentence came to mean life in prison, inspiring the popular saying “in
Louisiana life means life.” By law, a life sentence in Louisiana is now life-long unless the parole
board commutes the sentence; like other states, such parole is given rarely. As of February 2003
Louisiana had 3,822 lifers, of whom 3,392 were housed at Angola Prison, a prison scheduled to
soon house only lifers.
Leotha Brown
While working at a bar to earn money for college in 1963, Leotha Brown shot and killed a man in a
dispute tinged with racial overtones. Brown received a life sentence and at age 62 is now incarcerated
at Louisiana’s Angola Penitentiary. While Brown’s early years in prison resulted in a number of
disciplinary reports, not unlike many young offenders, he has for many years been engaged in
leadership programs in the prison. He was one of seven prisoners selected for a negotiating team
when the Department of Justice came to the prison in the 1970s to resolve conflicts between prisoners
and staff. He now devotes his time to college correspondence courses and the prison’s Bible College.

The procedure by which Louisiana lifers can gain clemency has been made considerably more
restrictive in recent years. At the 1973 state Constitutional Convention, a provision was added to
the Constitution requiring that the Governor could only grant clemency after receiving a
recommendation from a newly-created Pardon Board. In 1995, the legislature added a provision
that “at least” one of the five members of the Pardon Board be appointed from a list submitted by
the membership of Victims and Citizens Against Crime, Inc., a New Orleans-based victims
rights organization. Then in 1997, a new statute was adopted requiring that votes of four of the
five members of the Pardon Board be gained for a favorable recommendation, as opposed to the
previous requirement of a majority of the Board (or three votes). 7 Lifers in Louisiana are now
ineligible to apply for clemency until 15 years after being sentenced; if denied, they cannot apply
for six more years.

7

Berrigan, Helen Ginger, “Executive Clemency, First-Offender Pardons; Automatic Restoration of Rights,”
Louisiana Law Review, Vol. 62, No. 1, 2001, pp. 49-57.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

7

_____________________________________________________________________________________________

Michigan
Since 1942 Michigan law had required that parolable lifers would become eligible for parole
after serving 10 years. In practice most such lifers were released within a few years of becoming
eligible. In 1992, though, a change in state law required that lifers serve at least 15 years before
parole eligibility. In addition, an increasingly conservative political climate in regard to lifers led
to a position of the parole board that “life means life,” and that in order to be released,
“something exceptional must occur.” This policy was in contrast to the understanding of judges
when they imposed life sentences. In a 2002 survey of judges conducted by the Michigan State
Bar, a majority responded that the possibility of parole was a factor in their sentencing decisions
and that they had assumed that parolable lifers would serve 20 years or less. Data from the
Department of Corrections show that while 124 lifers were paroled in the 1960s, only 31 were
paroled in the 1990s (with a far larger prison population than in the 1960s).8
Charles Lohn
Charles Lohn was given a parolable life sentence in Michigan for armed robbery in 1968. After
serving 20 years, he was released on parole in 1988. Unable to manage a family crisis, Lohn failed to
report to his parole agent for several months. Despite having remained employed and crime-free, he
was returned to prison on a technical violation of parole, with a recommendation from the parole agent
that he get “another try at parole in the not too distant future.” Upon being returned to prison, the
Parole Board member who interviewed Lohn concurred with the recommendation, adding that the
Board “is ordering a short continuance to impress upon Mr. Lohn that the rules, regulations and
conditions of parole must be adhered to.” With a new parole board policy that “life means life,” Lohn
was given five-year setbacks on parole consideration twice in the 1990s. Charles Lohn has just turned
60.

New Mexico
In New Mexico, a life sentence had long translated into ten years time served before parole
eligibility. But two state statutes were contradictory in prescribing the number of years to be
served by lifers. In response to this, a 1986 opinion by the state Attorney General recommended
that lifers serve at least 30 years before parole consideration and that they receive no good time
credit. 9
Pennsylvania
Since 1941 Pennsylvania law has required that all life sentences are imposed as life without
parole. But for many years, governors frequently exercised their commutation powers so that
considerable numbers of lifers were released after serving about 20 years. In the administration
of Governor Raymond Shafer (R), for example, 95 lifers received a commutation from 1967-71.
Following that, Governor Milton Shapp (D) issued 317 commutations in his two terms from
1971-79. Those figures dropped dramatically in the 1980s, with Governor Richard Thornburgh
(R) commuting only seven cases from 1979-87. Since 1995 only one lifer has received a

8

Information from The High Cost of Denying Parole: An Analysis of Prisoners Eligible for Release, Citizens
Alliance on Prisons and Public Spending, November 2003.
9
Op Atty. Gen. No. 86-01 (New Mexico, 23 Apr 1986).

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

8

_____________________________________________________________________________________________

commutation, a man who had served 20 years for lending a gun to a friend who then committed a
robbery. 10
The discretion inherent in these processes can be seen in examining parole actions for all
prisoners, not just lifers. In 1992, 77.4% of eligible inmates were paroled; this proportion was
cut in half, to 38.8%, by 1996 under a new governor. Another change in administration then led
to an increase in paroles granted to 50.9% by 2002. 11
Tyrone Werts
In 1975, Tyrone Werts received a mandatory life sentence for second-degree murder resulting from
his participation in a robbery with four other people. Werts had been in a car parked two blocks from
the scene of the robbery, but because of a killing that resulted from the robbery was convicted due to
his knowledge of the crime. In the 28 years he has since served at Graterford Prison in Pennsylvania,
he went from being a high school dropout to earning a college degree. He also initiated a range of
community service projects both within and outside the prison, including tutoring illiterate inmates,
leading restorative justice initiatives, and organizing an Anti-Crime Summit at the prison. Among his
many awards is a commendation for preventing serious injury to a guard who was being attacked by a
prisoner.

Tennessee
In Tennessee a life sentence with the possibility of parole now requires that a minimum of 51
years be served before meeting the parole board. Tony Baldwin had been convicted of murder in
1979 and sentenced to a life term. At the time of his sentencing, the state’s policy was that
offenders had to wait 30 years before becoming eligible for parole. But Baldwin’s good
institutional record had earned him sentence-reduction credits that moved up his hearing to 2001,
after serving 22 years. His request for release was denied by the parole board, which also told
him that he could not apply for release again for 20 more years, until 2021. In a challenge to the
decision, the state Court of Appeals ruled in 2003 that the decision represented “an arbitrary
exercise of the Parole Board’s authority,” and failed to recognize that “over time people can
change, and that even a convicted felon may be able to live in accordance with the law.” The
court concluded “the essential effect of the Board’s action is to change Mr. Baldwin’s sentence
to life without parole, contrary to what the Legislature intended.”12
10

Information from Life Sentenced Prisoners, Pennsylvania Prison Society, as obtained from the Pennsylvania
Department of Corrections.
11
Annual Statistical Report, Pennsylvania Department of Corrections, 2002, p. 45.
12
Johnson, Rob. “Court strikes down long parole wait,” The Tennessean, August 20, 2003, p. 2B.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

9

_____________________________________________________________________________________________

LIFERS IN PRISON
The number of persons serving life sentences has grown along with the increase in the overall
prison population of recent decades. Our national survey obtained data on lifers in prison from
state and federal corrections systems as of 2002/2003. Data were obtained either from the
website or direct contact with the departments of corrections. As seen in the table on the
following page, there are now a record 127,677 persons serving a life sentence in state or federal
prison. Overall, one of every eleven (9.4%) inmates is now serving a life term.
In recent years policy changes have also increased the number of persons serving a sentence of
life without parole. Our survey finds that one of every four lifers – 33,633 persons – is now
serving a sentence of life without parole.
Imprisonment of lifers has grown nationwide, but is particularly significant in a number of states:
•
•
•

In 12 states more than 10% of the prison population is serving a life term; in California and
New York the proportions are approaching one of every five inmates.
The use of life without parole statutes is dramatic as well. In Louisiana and Pennsylvania
one of every ten prisoners is serving a life sentence, all of which are sentences of life without
the possibility of parole.
In four other states -- Illinois, Iowa, Maine, and South Dakota -- and the federal system, all
life sentences are life without parole. While mechanisms for release exist in each of these
states, the presumptive sentence is that offenders will never be released.

There are also significant variations in the degree to which states employ both life sentences and
life without parole. New York, for example, maintains the highest proportion (19.4%) of lifers
in its prison population, yet uses sentences of life without parole very sparingly (0.1%). Overall,
the lifer proportion of the prison population varies from 0.9% in Indiana to 19.4% in New York.
And in terms of the use of life without parole, four states – Alaska, Kansas, New Mexico, and
Texas – have no such inmates, while Louisiana leads the nation with 10.6%. (Note, though, that
Texas holds the second largest number of people on death row in the nation.)
Table 1 - State Variations in Lifer Populations
% Lifers

States

% Life
without
Parole*

0-5%

Alaska, Arizona, Connecticut, Federal,
Illinois, Indiana, Kentucky, Maine**,
Montana, New Jersey, New Mexico, North
Dakota, South Dakota, Vermont

0-1%

5.1-10%

Arkansas, Colorado, Delaware, Hawaii,
Idaho, Iowa, Kansas, Maryland, Michigan,
Minnesota, Mississippi, Missouri, Nebraska,
New Hampshire, North Carolina, Oklahoma,
Oregon, Pennsylvania, Rhode Island, South
Carolina, Texas, Virginia, Wisconsin
Florida, Georgia, Louisiana, Ohio,
Tennessee, Wyoming
Alabama, California, Massachusetts, Nevada,
New York, West Virginia

1.1-3%

10.1-15%
15.1-20%

3.1-5%

States
Alaska***, Arizona, Connecticut, Georgia, Indiana,
Kansas***, Kentucky, Maryland, Minnesota, New
Jersey, New Mexico***, New York, North Dakota,
Ohio, Oregon, Montana, Rhode Island, Texas***,
Vermont, Wyoming
Arkansas, California, Colorado, Idaho, Illinois,
Maine, Mississippi, Missouri, New Hampshire, North
Carolina, Oklahoma, South Carolina, Tennessee,
Virginia, Washington

5.1-7%

Alabama, Delaware, Federal, Nebraska, Nevada,
South Dakota
Florida, Iowa, Michigan, West Virginia

7.1-11%

Louisiana, Massachusetts, Pennsylvania

* = No information for Hawaii, Utah and Wisconsin
*** = State has no Life without Parole statute
**= 36 of the 45 persons serving a life sentence in Maine are serving life without parole. The remaining nine

persons were sentenced to life prior to a law that mandates that all life sentences be imposed without parole.

Table 2 - Lifers in Prison, 2002-03
Life

% Prison
Population

Life without
Parole

% Prison
Population

Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Federal System

4,837
193
1,149
1,094
28,807
1,151
417
391
8,643
6,517
391
418
1,291
233
553
809
748
3,822
45
2,200
1,618
4,572
392
2,003
2,452
61
257
1,908
149
928
303
12,985
3,307
32
4,729
1,880
623
3,865
162
1,907
142
1,918
7,698
N/A
55
1,894
1,017
556
1,322
171
5,062

17.3
4.5
3.7
8.1
18.1
6.2
2.3
7.2
11.2
13.6
6.9
7.2
3
0.9
6.6
8.9
4.6
10.6
2.4
9.4
16.9
9.2
5.6
8.9
8
1.9
6.4
18.6
5.9
3.1
5.0
19.4
9.8
2.8
10.5
8.3
5.1
9.7
5.7
8.3
4.6
11
6.1
-3.1
6.1
6.2
16
6.1
10.1
3.4

1,334
0
117
408
2,984
290
189
178
4,478
332
N/A*
79
1,291
66
553
0
30
3,822
36
211
756
2,629
26
285
712
25
184
417
55
11
0
110
479
10
105
460
105
3,865
20
439
142
204
0
N/A
5
425
478
222
N/A
4
5,062

4.8
0
0.3
3
1.9
1.6
1
3.3
5.8
0.7
-1.4
3
0.2
6.6
0
0.1
10.6
1.9
0.9
7.9
5.3
0.3
1.3
2.3
0.8
4.6
4.1
2.2
0
0
0.1
1.4
0.9
0.2
2
0.9
9.7
0.7
1.9
4.6
1.2
0
-0.3
1.4
2.9
6.4
-0.2
3.4

Total

127,677

9.4

33,633

2.5

Life without
Parole as %
of all Lifers

26.3

State

* Staff in Department of Public Safety unable to provide breakout for those serving
life without parole; estimated to be less than 1% of the state prison population.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

11

_____________________________________________________________________________________________

The 127,677 lifers in prison represent an increase of 83% from the number of lifers nationally in
1992,13 which in turn had doubled since 1984. 14 During the 1990s the growth of persons serving
life without parole has been even more precipitous, an increase of 170%, between 1992 and
2003. Overall, one of every six lifers in 1992 was serving a sentence of life without parole. By
2003, that proportion had increased to one in four.
Table 3 - Growth of Lifer Prison Population, 1984-2003
Year

Lifers

Life Without Parole

% of Total Lifers

1984
1992
2003

34,000
69,845
127,677

N/A
12,453
33,633

--17.8%
26.3%

In addition, the number of long-term prisoners is considerably greater than just the total of lifers,
and contributes to the population of what can be considered “virtual lifers.” These are persons
serving very long sentences, or consecutive sentences, that will often outlast the person’s natural
life. One 2000 study estimated that more than one of every four (27.5%) adult prisoners was
serving a sentence of 20 years or more.15 And data from the Department of Justice show that as
of 2002, state and federal prisons held 121,000 persons age 50 or over, more than double the
figure of a decade earlier.16

13

1992 figures except for Alaska, Indiana, and Maryland (not available) from Kathleen Maguire, Ann L. Pastore &
Timothy J. Flanagan. (Eds.). (1993). Sourcebook of Criminal Justice Statistics 1992. U.S. Department of Justice,
Bureau of Justice Statistics. Washington, DC: USGPO. Data for Alaska, Indiana and Maryland as of 1990 and
adapted from Kathleen Maguire & Timothy J. Flanagan, (Eds.). (1991). Sourcebook of Criminal Justice Statistics
1990. U.S. Department of Justice, Bureau of Justice Statistics. Washington, DC: USGPO.
14
Corrections Compendium, Vol. 3, No. 9, 1984. Lincoln, NE: 1984. This count included 47 states, Guam and
Puerto Rico.
15
Camp, C.G. and Camp, G.M. (2000). The 2000 Corrections Yearbook: Adult Corrections. Middletown, CT:
Criminal Justice Institute (p. 54).
16
Ron Word, “Elderly Inmates Swell Nation’s Prisons,” Kansas City Star, February 27, 2004.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

12

_____________________________________________________________________________________________

INCREASING TIME IN PRISON
In addition to the growing numbers of lifers in prison, policy changes of recent years have
resulted in lifers spending significantly greater time in prison. As we have seen, this has been
the result of several factors:
•
•
•

Increasing number/proportion of sentences to life without parole;
Gubernatorial policy changes that have limited or eliminated consideration of commutation
for lifers;
Legislative changes such as “truth in sentencing” that have increased the amount of time
served before parole consideration.

The impact of these changes can be seen in states such as Georgia, where the Board of Pardons
and Paroles reported in 1998 that more life-sentenced inmates had died in prison that year than
been paroled. According to Parole Board Chair Walter Ray, “There’s a popular misconception
that life in prison doesn’t mean all of one’s natural life. In just the last year, there are 21 Georgia
lifers who are no longer around to tell you otherwise. If they could, they’d let you know that
parole for a life sentence is a rare commodity.” 17
Using data on prison populations and releases from the Bureau of Justice Statistics we have
developed estimates for the amount of time that lifers will serve in prison. 18 (See Appendix for a
description of this methodology.) Our analysis indicates that from 1991 to 1997 there was a 37%
increase in time to be served by lifers prior to release. Persons admitted in 1991 could expect to
serve an average of 21.2 years, a figure which rose to 29 years by 1997 (most recent figures
available). Thus, in contrast to popular imagery which sometimes portrays lifers as serving short
prison terms, the average life sentence today results in nearly three decades of incarceration.

Table 4 - Estimated Time to be Served by Lifers
Year of Admission
1991
1997

17

Time to Be Served
21.2 years
29

News release, Georgia State Board of Pardons and Paroles, “More Violent-Crime Lifers Die in Prison Than Are
Paroled,” June 1, 1998.
18
U.S. Department of Justice, Bureau of Justice Statistics. National Corrections Reporting Program, 1997: [United
States] {Computer file}. Conducted by U.S. Department of Commerce, Bureau of the Census. 2nd ICPSR ed. Ann
Arbor, MI: Inter-university Consortium for Political and Social Research [producer and distributor], 2000 AND U.S.
Department of Justice, Bureau of Justice Statistics, and U.S. Department of Justice, Federal Bureau of Prisons.
Survey Of Inmates In State And Federal Correctional Facilities, 1997 {Computer file}. Compiled by U.S.
Department of Commerce, Bureau of the Census. ICPSR ed. Ann Arbor, MI: Inter-university Consortium for
Political and Social Research [producer and distributor], 2000.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

13

_____________________________________________________________________________________________

A PORTRAIT OF LIFERS
The overwhelming majority of lifers in prison have been convicted of a violent offense. As seen
in Table 5 below, as of 1997, 90% of lifers were incarcerated for a violent offense, including
68.9% for homicide. Thus, judging by the offense of conviction, most lifers fall in the category
of the kinds of offenders for whom policymakers and the public assume a long sentence is
appropriate and desirable. And among this population are persons whose crimes clearly warrant
an extended period of incarceration due to the severity of the crime or because their release
presents a potential danger to the public.
Table 5 – Sentenced Offense of Lifers
Offense Type
Homicide/Murder
Rape/Sexual Assault
Armed Robbery
Other Violent
Drug Offense
Property
Other

Percent of Population
68.9
8.9
7.5
4.8
4.0
3.9
2.0

But a closer examination of the population of lifers in prison, and the process by which they were
sentenced, reveals that the profile of persons serving life sentences is far more complex than is
often assumed:
•

Lifers include those convicted for violent offenses but for whom many persons would
recognize that a life sentence is overly harsh and inappropriate. These include people
sentenced as violent offenders but for whom there are serious questions regarding culpability
or persons who have not committed violent acts themselves and who pose little threat of
injury to others in the future.

•

The composition of the lifer population is changing due to the increasing frequency of
imposing life sentences for drug crimes and “three strikes” cases that do not involve violent
offenses.

•

The criminal justice system as presently constituted does not perform well in making the
distinction between serious offenders for whom a life sentence is warranted and those for
whom a less severe punishment is appropriate and fair. Further, increasingly compelling
evidence suggests that, as presently constituted, the criminal justice system’s reliability in
determining guilt and the appropriate length of sentencing is far from certain.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

14

_____________________________________________________________________________________________

Examining some of the categories of persons sentenced to life imprisonment illuminates the
complexity of the profile of lifers. Among these categories are the following:
Battered Women
One study of battered women estimated that there were 800-2,000 women serving prison terms
for killing their abusers, although some scholars view these estimates as conservative.19 A
review of women in prison for homicide in Georgia found that more than half had been the
victim of domestic abuse, and in nearly two-thirds of the cases in which a woman killed a
partner, abuse was claimed to have occurred at the time of the crime. 20 These women must often
face the question from law enforcement and legal professionals of, “Why didn’t you just leave?”
In addition to an insensitivity to the complexity that many women face in trying to make a
decision to leave, this assertion ignores the fact that the risk of additional injury or death at the
hands of an abuser often increases after fleeing. 21
Many battered women have received little help from the legal system, since many criminal
justice professionals, including defense attorneys, have never understood or appreciated the
significance of abuse. Sociologist Elizabeth Leonard writes that:
“… many lawyers and investigators fail to adjust their approach to better fit the
experiences of women. As a result, potentially exculpatory information is not
investigated, crucial evidence remains unused, and women are left with no real
defense…. Some accused women take the offered plea bargains to protect their
children, to spare their families, to avoid the death penalty threatened by prosecutors,
or to speed up what they see as the inevitable guilty verdict.”22
The law itself works against women who have killed their abusers. The strategy of self-defense,
which is the approach likely to be asserted by an abuse victim who has killed, is not written for
these circumstances. To assert self-defense, a defendant must show that she was reasonably in
fear of her life (or for the life of another) at the time that she killed, and that she had taken every
reasonable means to avoid the threat. Both conditions are difficult to prove. The abuser is
frequently sleeping or otherwise harmless at the time, and the abuse victim is often physically
able to leave rather than commit a violent act.

19

Elizabeth Leonard, Gendered Injustice in the Lives of Convicted Battered Women, paper presented at University
of Limerick, May 28, 2002.
20
Judith Haley, A Study of Women Imprisoned for Homicide, Georgia Department of Corrections, June 1992.
21
Clemency for Battered Women in Michigan: A Manual for Attorneys, Law Students and Social Workers, available
online at http://www.umich.edu/~clemency/clemency_manual/manual_intro.html
22
Leonard, op. cit.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

15

_____________________________________________________________________________________________

Marie La Pinta
Marie La Pinta was convicted along with her brother of second degree murder in the killing of her
husband. La Pinta had grown up in Sicily, where she witnessed her mother live a life of brutal abuse
by her father. She came to the U.S. for an arranged marriage at the age of 19, speaking little English.
Her marriage to Anthony La Pinta was marked by serious physical abuse. The current district attorney
of Suffolk County, New York, where the killing took place, describes the crime as involving “a
history of marital discord within the family of long duration which reached flash point.” He further
describes La Pinta as having been a model prisoner who “poses no threat either to herself, the
institution, or, by extension, society in general,” and states that her debt for the crime “has been more
than expurgated by the length and conditions of her confinement.” Marie La Pinta has applied for
gubernatorial clemency and is currently serving a sentence of 25 years to Life in Bedford Hills prison.

Among such women in prison are those who have no prior criminal history, yet are charged with
murder or manslaughter and often receive harsh sentences. A study of 42 convicted survivors in
California, none of whom had a previous conviction for violence, found that only two women
received determinate sentences; the other sentences ranged from 7-to-life to 35-to-life, with six
women receiving life without parole.23
Current “one size fits all” sentencing structures blur the complexities of cases in which victims
of physical and sexual abuse are convicted of murder. Sentencing schemes which do not permit
adequate consideration of the range of relevant factors in making sentencing decisions hold the
victims of abuse to the same penalties that apply to more culpable offenders.
Mentally Ill Offenders
Research conducted by the Bureau of Justice Statistics documents that one of every six people in
prison has a history of mental illness.24 These figures are even more pronounced among the
population of lifers, with nearly one in five lifers (18.4%) suffering from a mental illness,
totaling an estimated 23,523 life-sentenced prisoners.25 These numbers are of particular concern
considering that correctional facilities are ill-equipped to provide safety and appropriate
treatment for mentally ill persons.26 Prisons are dangerous and debilitating for inmates with
mental illnesses, who often are victimized by other prisoners. Moreover, symptoms of mental
illness are often misinterpreted by correctional staff as disruptive behavior, which can lead to
additional punishment and disciplinary actions that may extend the length of an individual’s
sentence.

23

Ibid.
At midyear 1998, an estimated 283,800 prison and jail inmates were mentally ill; Bureau of Justice Statistics,
Mental Health and Treatment of Inmates and Probationers, July 1999.
25
Our analysis of the lifer population, using the 1997 Inmate Survey and employing the criteria used by BJS in its
report, Mental Health and the Treatment of Inmates and Probationers, ibid., indicated that 18.4% of all lifers
(18.8% of lifers in state prison and 9.3% of lifers in federal prison) were estimated to be mentally ill. This
proportion was applied to the totals from this report’s survey in order to produce the estimate of 23,523.
26
For more information on the conditions of mentally ill persons in prison, see Ill Equipped: U.S. Prisons and
Offenders with Mental Illness, Human Rights Watch, October 2003.
24

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

16

_____________________________________________________________________________________________

The high incidence of mentally ill lifers reflects troubling aspects of the complex relationship
between the criminal justice system and mental illness. That relationship is on one hand
adversarial, with the criminal justice system’s interest in accountability, punishment and
incapacitation opposing mental health’s emphasis on treatment and restoration. On the other
hand, the relationship is necessarily dependent, as criminal law struggles to define the intent,
mens rea, and the culpability required to establish the appropriate determination of guilt,
punishment, and responsibility for persons suffering from a mental illness. Both disciplines
struggle to establish the scope of punishment for persons who have committed criminal acts but
are incompetent to defend themselves or unable to appreciate the consequences of their behavior.
Sentencing a mentally ill person to life imprisonment represents a legal decision to punish within
a correctional setting instead of providing treatment according to mental health principles. The
contemporary manner in which the legal system makes this decision has been driven historically
by public reaction to notorious cases in which a defense of insanity has been successfully
asserted. In 1843, the British public’s and the Crown’s dissatisfaction with a judge’s finding that
Daniel M’Naghten was insane when he attempted to assassinate Prime Minister Robert Peel
prompted creation of the “M’Naghten Rule,” which allowed acquittal only for a person unable to
distinguish between right and wrong. This rule survived, with modifications, until recently, but
increasing interest in psychology and the prospect of treatment for mental illness led to new
judicially-created tests to determine legal insanity in the 1950s. 27 Subsequently, the American
Law Institute created a Model Penal Code, intended to be a broad-based and more liberal
restatement of criminal responsibility. By 1984, the test set forth in the Model Penal Code 28 was
law in the majority of states and federal circuits. However, that same year, a jury acquitted John
Hinckley by reason of insanity following his attempted assassination of President Ronald
Reagan. Public outrage and a backlash resulted in restrictive changes in the law in the federal
system and in over half the states. Three states abolished the defense altogether. Popular
criticism of the use of competing experts in the Hinckley trial led to additional rules restricting
the admission of expert opinion.
Reliance upon punishment and the criminal justice system for persons with mental illness may be
counterproductive to public safety. Current research shows an attenuated causal relationship
between mental illness and violent crimes. It is true, as many perceive, that people with mental
illness are at a higher than average risk of offending, particularly for violent crimes. But the
offending is more closely related to the increased susceptibility of mentally ill persons to other
factors, such as homelessness, substance abuse, and non-compliance with medication regimens.
Thus, research suggests pro-active health care coordinated between mental health and substance
abuse services, compliance with medication regimens, and application of risk assessment
measures are the best means of assuring maximum public safety. 29

27

The Durham Test provided that a person was not criminally responsible if his or her unlawful act was a “product
of mental disease or defect.” Durham v. United States 214 D 2d 862 (1954).
28
The Model Penal Code permitted an acquittal by reason of insanity if a defendant proved a “substantial
incapacity” to appreciate the difference between right and wrong, and focused on the defendant’s ability to control
his or her actions.
29
Henderson, Susan, Mental Illness and the Criminal Justice System, Mental Health Coordinating Council, May
2003. See also, Mentally Ill Offenders in the Criminal Justice System: An Analysis and Prescription, The
Sentencing Project 2002, pp. 7-9.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

17

_____________________________________________________________________________________________

In response to these issues, the Council of State Governments convened a broad advisory board
of policymakers and practitioners to develop a comprehensive framework for addressing issues
of mental illness in the criminal justice system. 30 The group’s report included a call for law
enforcement collaboration with mental health partners to increase the involvement of treatment
professionals, train defense attorneys to recognize mental health conditions, and institutionalize
screening and identification of mental illness. Strengthening community response through
augmenting mental health services plays a significant role in promoting public safety. Given that
the offense that triggers a life sentence is unlikely to be a person’s initial contact with the
criminal justice system, it is crucial that resources are dedicated to training criminal justice
professionals to identify persons in need of intervention by mental health agencies prior to their
involvement in more serious offenses.
Juveniles
Over the past decade, increasing numbers of states have adopted laws making it easier to try
juveniles in adult court. As a result, crimes committed by young people can now result in much
harsher punishments, including life sentences. While there are presently no national data on the
number of such lifer cases, they are clearly increasing in many states. A recent report documents
that just in the state of Michigan alone, there are at least 146 persons serving life without parole
for crimes committed when they were 14-16 years old.31
It has always been possible to prosecute in adult court juveniles whose age or criminal acts made
them inappropriate for juvenile court and corrections, with their rehabilitative goals and an
inflexible mandatory release date, usually at the age of 18 or 21. After the Supreme Court’s
decision in Kent v. United States32 in 1966, to do so required a hearing before a judge, with due
process protections including standards such as consideration of the maturity of the juvenile.
From the late 1980s through the early 1990s, the murder rate for juveniles increased
dramatically. Subsequent research determined that this rise was not a result of an increasingly
violent cohort of young people, but instead, easier access to firearms. Despite the fact that
juvenile homicide rates began to decline substantially after this rise, policymaker and media
attention focused heavily on punishments for these offenders. Legislators responded by
rewriting laws to allow, at a prosecutor’s request, or to require, based on charge alone, increased
prosecution of children as adults. From 1992 through 1995, 40 states and the District of
Columbia passed laws making it easier to try children as adults in criminal court. Many of these
laws by-passed judicial review completely, allowing for the “automatic transfer” of children into
adult court. 33 In many instances, judges no longer decide whether a child should be tried as an
adult or a juvenile; that determination is left to the prosecutor.
Few of the new laws modified the procedural rules, definitions of offenses, or the sentences that
exist in adult criminal court. A life sentence mandated for any adult defendant who committed a
particular crime applied in full force to juveniles convicted in adult court for that crime.
30

Council of State Government, Criminal Justice/Mental Health Consensus Project, June 2002.
Jeff Kresnak, “ACLU to study inmates’ cases, Detroit Free Press, April 1, 2004.
32
Kent v. United States, 383 U. S. 541 (1966).
33
Patrick Griffin, Patricia Torbert & Linda Szymanski Trying Juveniles as Adults in Criminal Court: An Analysis of
State Transfer Provisions (Office of Juvenile Justice and Delinquency Prevention, 1998)

31

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

18

_____________________________________________________________________________________________

The case of 12-year-old Lionel Tate in Florida illustrates the risks inherent in charging a juvenile
in adult court, a system ill-prepared to address the needs and challenges of juvenile defendants.
Tate had received a life term after killing a 6-year-old playmate. His defense asserted a claim
that her death was a result of overly vigorous play wrestling. After Tate and his mother rejected
a plea offer of three years in prison and went to trial, he was convicted and sentenced to a
mandatory life term in prison. Ultimately, his conviction was overturned because a judicial
hearing had not been held to assess his mental competency, and in 2004 Tate agreed to a
renewed three-year prison term plea, which permitted his immediate release from prison.
The case of Tate and others across the country demonstrate a profound shift in the treatment of
juveniles, prompting the judge in a high-profile juvenile case in Michigan to question the shift
from the principles of rehabilitation and prevention that characterized the founding of a separate
juvenile system in the 19th Century to the punishment and retributive-oriented approach of the
adult system. In refusing to sentence Nathaniel Abraham, convicted of a shooting that occurred
when he was 11 years old, to life, Judge Eugene Arthur Moore voiced a critique of life sentences
for juveniles: “Don't ask the Judge to look into a crystal ball today and predict 5 years down the
road. Give the Juvenile system a chance to rehabilitate. Don't predict today, at sentencing,
whether the child will or will not be rehabilitated, but keep the options open.” 34
While the objectives of both incapacitation and punishment are appropriate considerations in the
sentencing of juveniles who commit serious violent offenses, life sentences assume incorrectly
that teenagers are as able as adults to assist in their defense and that they possess the maturity to
understand the consequences of their actions in the same manner as adults. Further, life
sentences for juveniles ignore the capacity of teenagers to change over time.
“Accountability” Offenses
Lifers sentenced for violent offenses also include persons sentenced under “accountability”
doctrines. Under these provisions, a participant in a crime, such as a store robbery, who serves
as a lookout or driver and may not even be aware of the presence of a gun, may be sentenced to
prison for life as the result of a murder committed in the course of the robbery. The case of
Gladys Wilson, below, illustrates the severe consequences of such prosecutions.
Gladys Wilson
Gladys Wilson pled guilty to aiding and abetting an armed robbery in Michigan, in 1978, her first
offense. She was a 31-year old mother of an 11-year-old girl and had worked at the same company
for ten years. Wilson’s crime had involved driving and then picking up her husband at a grocery store
that he planned to rob. She was not aware he would kill someone, for which he is currently serving a
mandatory life sentence. Gladys Wilson has earned 67 college credits while in prison, has obtained a
paralegal certificate, and has worked as an aide for mentally and physically impaired inmates. In
1986, in anticipation of parole release after serving 10 years, Wilson was moved to a minimum
custody prison. Parole board action was delayed until 1992, by which time the “life means life”
policy precluded strong consideration of her case. In 2003, despite a letter from the successor
sentencing judge urging her release, she was again denied parole consideration. Under current
Michigan sentencing guidelines, her offense today would result in a minimum sentence of 9-15 years.
34

Sentencing Opinion for Nathaniel Abraham, Judge Eugene Arthur Moore, State of Michigan in the 6th Circuit
Court for the County of Oakland Family Division, File No. 97 63787 FC, January 13, 2000.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

19

_____________________________________________________________________________________________

Drug Offenders
While the lifer population overwhelmingly consists of persons convicted of a violent offense,
4%, or about 5,000 lifers, have been convicted of a drug offense. In the federal system alone,
approximately 2,000 life sentences are for drug offenses, representing about 39% of all life
terms. Some drug offenders have been convicted of high-level offenses, such as drug “kingpins”
who have imported substantial quantities of narcotics into the country. But others are serving
life terms only because of particularly harsh state laws that would not result in such severe
sanctions in neighboring states.
In California, for example, increasing numbers of offenders are being sentenced to life in prison
under the “three strikes” law, which requires a sentence of 25-Life for a third strike conviction
following two previous qualifying strike offenses. To date, 1,281 persons (17.5%) of the state’s
7,335 three strikes cases have had a life sentence triggered by a drug offense; of these, 53.5%
were for drug possession offenses.
In Michigan, more than 200 persons are serving life terms for drug offenses under the state’s
“650 Lifer” law. Under this law, anyone convicted of selling 650 grams of cocaine or heroin,
even for a first offense, is sentenced to life in prison. As originally adopted in 1978, the law
required that the sentence be life without parole, but this was changed by the legislature to life
with the possibility of parole in 1998. Despite this, only a handful of persons sentenced under
this provision have since been released on parole.
Property Offenders
As is true of drug offenders, many of the estimated 5,000 persons 35 serving life for a property
offense have committed a serious offense and/or have a lengthy criminal record. But in this
regard as well, the total includes growing numbers sentenced under California’s three strikes
law, currently numbering 2,303, or 31%, of all third strikers in the state. These include such
cases as an offender sentenced for the theft of three golf clubs and another for videotape theft
with a value of $153. Overall, 57.5% (4,220) of California’s three strikes cases involve a nonviolent offense as the third strike.

35

This estimate is likely to be quite conservative since it is based on the percentage of property offenders serving
life sentences in 1997, well before the effects of California’s “three strikes” law had been fully experienced.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

20

_____________________________________________________________________________________________

THE IMPACT OF DEFICIENCIES IN THE CRIMINAL COURTS
Our analysis has indicated that the lifer population includes persons for whom a term of lifetime
incarceration is excessive considering factors such as the severity of the offense and the
circumstances of the offender. In addition to individual level factors that bring to light questions
regarding the appropriateness of lifetime incarceration, there are also structural considerations,
including the reliability and fairness of the judicial decision process. In the American system of
justice, there should be no reasonable doubt that defendants convicted and sentenced for the most
serious of crimes were in fact guilty of the offense, particularly where the punishment is so
severe and the financial cost so great. Unfortunately, the evidence from a wide range of sources
raises doubts about both the reliability and the fairness of the judicial system as it decides for
whom a life sentence is appropriate.
In recent years, the application of new DNA technology and the work of organizations such as
the Innocence Project have resulted in findings that a number of people sentenced to death were
innocent of the crime of which they were convicted. 36 In addition, the work of the Innocence
Project has led to the exoneration of 38 lifers, and a comprehensive analysis at the University of
Michigan suggests that the number of innocent persons in prison for all offenses is likely to
number in the thousands. 37
The documented frequency of erroneous convictions and unfair sentences in capital cases raises
serious questions about the reliability of convictions and sentences in lifer cases since the
procedural safeguards are greater in capital cases. Attorneys are supposed to be more
experienced, fairly demanding standards are supposed to govern their work, and their access to
defense investigatory and forensic resources generally more assured. Sentencing is by a jury,
which is given guiding standards intended to eliminate arbitrary sentencing, and a death sentence
can only be imposed by a unanimous verdict in most states. A direct appeal is automatic and
collateral appeals are more frequently applied.
These same protections are not applied to defendants when the sentence may “only” be life in
prison. For example, unlike defendants in capital cases, persons sentenced to life have no right
to post-conviction counsel in most states. The consequences of this practice can be seen in the
case of Walter McMillian, whose jury recommended life without parole after he was convicted
of murder in Alabama in 1988. The sentencing judge overrode the jury’s recommendation and
imposed a death sentence. Because McMillian was under a sentence of death, he received legal
assistance from the Equal Justice Initiative of Alabama, which proved his innocence after several
years. Ironically, had the judge accepted the life without parole sentence, McMillian would still
be in prison today without access to any legal review.
According to legal experts, bar associations, and the opinions of appellate judges, the criminal
court process, particularly for indigent defendants, is routinely failing to fairly and reliably sort
out the guilty or determine sentencing issues. While most states have established public
defender programs of some sort, and while defense counsel are theoretically available in most
jurisdictions, the American Bar Association found that many public defender programs are
36
37

See The Innocence Project, online @ www.innocenceproject.org
Adam Liptak, “Study Suspects Thousands of False Convictions,” New York Times, April 19, 2004.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

21

_____________________________________________________________________________________________

overloaded, rely on inexperienced and underpaid attorneys, and do not provide a vigorous
defense. As one national expert summarized a lifetime of observation:
As of 2003, virtually everywhere in this country defense services are deficient,
and in some places they are woefully inadequate. The goal in providing lawyers,
as Gideon38 said, was to assure fairness in the criminal justice system and to
prevent the wrongful convictions of innocent persons.
Gideon, as well as the Supreme Court’s other right to counsel decisions, have
constituted an enormous unfunded mandate imposed upon the states and local
jurisdictions for the past forty years. Moreover we have been in virtually every
jurisdiction in a state of crisis, and recently the problem seemingly has gotten
worse as state and local governments have wrestled with very difficult budgetary
problems. And I have reluctantly concluded that unless there’s a major change in
policy and in funding of indigent defense these problems we talk about are likely
to continue indefinitely. 39
The implication of the inadequacy of criminal defense services establishes a constant risk, indeed
an ongoing reality, that truly innocent persons are convicted in the criminal justice system. 40
These cases represent systematic failures and “assembly-line” justice that virtually assures
erroneous convictions. A recent study examining the indigent defense system in Virginia
concluded that substandard defense practice is an “accepted norm,” lawyers for indigent
defendants make little use of services such as expert witnesses who are often essential to proper
representation, and that appointed defense lawyers are underpaid compared to lawyers providing
services for other government functions. In Virginia, the state will only pay a maximum of
$1,096 in defense fees for felonies that may result in a life sentence. 41
While ABA and state defender standards require a full and adequate defense for all defendants,
including sentencing advocacy, 42 in 1984 the high court established in Strickland v. Washington
a very low standard by which the services of defense counsel would be found constitutionally
sound. Essentially, a convicted defendant must show on appeal that, first, trial counsel's services
were deficient as judged by practice standards, and second, that were it not for the deficiency, a
jury would almost certainly have returned a favorable verdict. 43 It is often relatively easy to
establish the first Strickland test; however, it has proven difficult to satisfy the second, for judges
on appeal are free to speculate about the potential impact of evidence on a jury. For example,
judges often differ among themselves dramatically, so that the history of many cases includes
38

Gideon v. Wainwright, 372 U.S. 335 (1963).
Norman Lefstein, Professor of Law and Dean Emeritus, Indiana University School of Law, at the National Legal
Aid and Defender Association conference, November 13, 2003.
40
Ibid. For additional examples of wrongful convictions, see the Department of Justice study documenting 28 such
cases. Edward Connors, Thomas Lundregan, Neal Miller, and Tom McEwen, Convicted by Juries, Exonerated by
Science: Case Studies in the Use of DNA Evidence to Establish Innocence After Trial, June 1996, Washington, DC:
U.S. Department of Justice.
41
“Justice Denied in Virginia,” Washington Post, February 10, 2004.
42
American Bar Association, Standards for Criminal Justice, 3 rd Edition. Standard 4-8.1 Sentencing; Standard 185.17 Sentencing hearing. State Standards are compiled in “Compendium of Standards for Indigent Defense
Systems,” (Dec 2000) available on line from the National Criminal Justice Reference Service.
43
Strickland v. Washington, 466 U.S. 668, 687, 694 (1984)
39

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

22

_____________________________________________________________________________________________

vigorous disagreement among judges who have heard the case in different courts, one side
expressing certainty that the impact of counsel could not have made a difference, the other
certain that it played a crucial role.
Strickland's legacy is that inadequate attorney representation is seldom reviewed and almost
never reversed. This systemic failure in the judicial process directly impacts the quality of the
decisions courts make to impose life and other lengthy sentences. As noted, among those who
are categorized as violent offenders are battered women who killed their partners, children, and
mentally ill persons. For each of these, the skill of the defense team, its collective knowledge
about abuse, adolescent development and behavior, or mental health, and its ability to present
that information to a prosecutor, a judge, and a jury are essential to a fairly adjudicated decision
and sentence. At present, the judicial system is often inadequate to sustain confidence in the
accuracy or fairness of a verdict and sentence, and this unreliability is exacerbated in cases in
which issues are complex and nuanced.

Johnnie Brown
Johnnie Brown was found guilty of armed robbery and sentenced to 30 years in Illinois in 1991. At
the trial, Brown testified that he believed that the victim was going to attack him, and as he became
alarmed for his safety, he grabbed the victim and demanded to know why he was being followed.
According to Brown, it was only at the end of this confrontation that Brown asked for money from the
victim, a take of fifty cents and an adult bookstore token. Brown had a long and well-documented
history of mental illness, including treatment and hospitalizations; however, the court was never
informed of his psychiatric background. Despite the fact that Brown’s illness led to inappropriate
behavior and outbursts in court and greatly complicated his ability to communicate with his public
defender and participate in his own defense, his attorney never followed up with a psychiatric
examination before sentencing and never obtained relevant medical records. During his trial, there
was no reference to the fact that at the time of the offense Brown was a diagnosed schizophrenic who
had stopped necessary medications.
The US Court of Appeals for the Seventh Circuit reversed Brown’s conviction because of his trial
lawyers’ “half-hearted,” “lackadaisical,” and “less than lawyerlike” efforts, which were compounded
by the failure of court psychiatrists and the probation department to obtain Brown’s history or to speak
to his family. The result was that the sentencing judge was “less than well informed of critical
information.” To reach this appeal, Brown had to fight through a succession of state court and then
federal court proceedings. Brown’s case was the subject of eight different decisions by the trial court
hearing post-conviction matters; each of these courts purportedly applying virtually the same legal
standards as did the Seventh Circuit, and each denying Brown’s request for relief. It was only through
the defense work of the Bluhm Legal Clinic of the Northwestern University School of Law that
Brown’s case was able to be carried to the Seventh Circuit, a fortunate resource for Brown that is not
available to the majority of offenders.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

23

_____________________________________________________________________________________________

RECIDIVISM AND PUBLIC SAFETY CONCERNS
Overview and Previous Research
For lifers, as with all offenders, the rationale for imposing a prison term may involve several
goals of sentencing. A first is clearly punishment, reflecting the seriousness of the offense. The
determination of how much punishment is imposed for an offense is subject to a broad range of
considerations – the degree of culpability of the offender, historical norms, community attitudes,
and other factors. In general, “get tough” policies of recent decades have dramatically increased
the severity of punishments imposed on all offenders, in terms of increasing both the number of
prison sentences and the length of time to be served in prison.
A second major consideration in imposing a life sentence relates to the goal of incapacitation;
essentially, providing public safety through incarceration. In this regard, the question for public
policy regards the extent to which long-term incarceration provides crime control benefits. More
specifically, to the extent that prison terms have increased in recent years, what are the additional
benefits, and costs, of these policies?
In the past a number of studies have examined the recidivism rate of released lifers, particularly
those imprisoned for homicide. Generally, these have found that lifers have very low rates of
recidivism, including for violent crimes. For example, in Michigan, 175 persons convicted of
murder were paroled between 1937 and 1961; none committed another homicide and only four
were returned to prison for other offenses. In Canada, between 1920 and 1967, 119 persons
originally sentenced to death for murder had their sentences commuted to life and were
eventually released on parole; one was convicted of another homicide. From 1959 to 1967, an
additional 32 persons were released and by 1967 only one had been convicted of a new offense
(not a murder).44
Estimating Lifer Recidivism
These previous studies are instructive, but rely on data that is now several decades old. While
there have been no recent analyses of lifers and recidivism, we can examine recidivism data from
the Bureau of Justice Statistics to develop estimates of the magnitude of these dynamics.45 Using
these data, we can determine the following:
Low rates of rearrest for released lifers
As seen in Table 6 below, lifers are rearrested at much lower rates than for the overall prison
population. In particular:

44

Sheppard, Colin. “Towards a Better Understanding of the Violent Offender,” Canadian Journal of Criminology
and Corrections, 1971, 13(1): 60-67.
45
U.S Department of Justice, Bureau of Justice Statistics. Recidivism of Prisoners Released in 1994: [UNITED
STATES] {Computer File}. 3 rd ICPSR version. Washington, DC: U.S. Department of Justice, Bureau of Justice
Statistics, and Regional Justice Information Services Commission [producers], 2002. Ann Arbor, MI: Interuniversity Consortium for Political and Social Research [distributor], 2003.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

24

_____________________________________________________________________________________________

•
•

Lifers are less than one-third as likely as all released offenders to be rearrested within three
years of release from prison.
Four of every five lifers are not rearrested. Of the lifers released in 1994, 20.6% were
rearrested, compared to an overall rearrest rate of 67.5%.46 (Note that these data all refer to
rearrest rates, and that not all arrests result in a conviction on a new offense.)
Table 6 – Rearrest Rates (%) of Prisoners Released in 1994 47
Rearrest Offense

Release
Status/Offense
All
Lifers
Violent
Property
Drug
Public Order

All Offenses

Violent

Property

Drug

Public Order

67.5

21.6

31.9

30.3

28.3

20.6
61.7
73.8
66.7
62.2

18
27.5
21.9
18.4
18.5

19.9
25.5
46.3
24
22.9

2.9
22.6
27.2
41.2
22.1

8.3
27.4
29.2
27.7
31.2

Lifers not more likely to be rearrested for a violent offense
Lifers – 90% of whom are incarcerated for a violent offense – are no more likely to be rearrested
for a violent offense (18%) than property (21.9%) or drug offenders (18.4%). Even for murder,
lifers are only marginally more likely to be rearrested than property or drug offenders (1.1% vs.
0.8% and 0.7%). Overall, released lifers are slightly more likely to be charged with a property
offense (19.9%) than a violent offense (18%).48
These trends illustrate the importance and potential benefit of sound risk-based analysis for
release, and provide parole officials an opportunity to assess the public safety risks for individual
lifer releases and to plan reentry and supervision plans to respond appropriately. This is a more
effective policy than the current system which often blindly locks up persons for long periods of
incarceration, releases people without protocol in place for risk assessment or reentry planning
for the transition back into society, and, therefore, poses significant potential negative
consequences for public safety.

46

Since the size of the sample of lifers released from prison in 1994 is small, this analysis is not intended to be
immediately generalizable to the entire population of released lifers; rather, this information should be considered a
sample of released lifers from 15 states, recognizing standard error issues inherent in working with small sample
sizes.
47
The data on lifers is taken from analysis by The Sentencing Project. All other data is from Table 10 in Langan,
P.A. & Levin, D.J. (2002). Recidivism of Prisoners Released in 1994. Washington, DC: U.S. Department of
Justice, Bureau of Justice Statistics. Numbers do not add to totals due to multiple arrest charges for some offenders.
48
It should be noted that there is significant overlap in the rearrest offense figures; many persons released from
prison in 1994 were rearrested for more than one charge. Of the 1,228 released lifers in our analysis, 253 (20.6%)
were rearrested, accounting for over 600 offenses, or nearly 2.5 offenses per person. Thus, four of every five
released lifers were not rearrested, but those who were averaged nearly 2.5 charges per person. Among the lifer
population, it is likely that many of these multiple offenses arose out of the same event (e.g., a person arrested for
both drug selling and possession of a weapon) since lifers arrested for even a single offense are likely to have their
parole revoked and be sent back to prison.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

25

_____________________________________________________________________________________________

Implications for Public Policy
These data on recidivism have several implications for public policy in regard to lifer
imprisonment:
Fiscal Cost of Lifer Imprisonment
The fiscal cost of maintaining a large lifer population is significant and growing. A conservative
annual cost of incarceration is $20,000. At this rate the annual national cost for life
imprisonment is an estimated $2.5 billion. However, this figure understates the actual costs of
life imprisonment. Since the cost of imprisonment rises significantly as the level of security
increases, life imprisonment, often in maximum-security facilities, results in considerably higher
costs. Moreover, life sentencing creates a geriatric prison population with far more expensive
costs for health care and other services. The cost of maintaining an aging prisoner is estimated to
be as much as $69,000 a year.
As a result of these dynamics, we can estimate that an average life sentence imposed by a judge
costs taxpayers about $1 million. Assuming that a typical lifer is sentenced at the age of 30 and
will live until 70, we can estimate conservatively that incarceration costs of $20,000 a year from
age 30 to age 60 will total $600,000. From age 60 to 70, costs are conservatively at least
$40,000 a year, yielding a total lifetime cost of $1 million.
In Texas alone, approximately 5% (6,364) of the prison population is 55 years and older, and
estimates are that by 2008, there will be over 10,000 persons in state institutions over the age of
55.49 In the federal system, elderly prisoners are the fastest growing population, with nearly 12%
of federal inmates over the age of 50. 50 This is due, in large part, to longer prison sentences
generally, and specifically, to an increase in the use of life sentences. In Texas, where there are
slightly under 8,000 persons serving life sentences, the growth of the aging population is a
significant financial burden, and any policy responses are affected by legislative constraints on
the early release of certain types of offenders.
For example, two convicted elderly sex offenders, a population for whom early releases are
forbidden, cost the state $1 million per month for treatment of heart and lung ailments. Both are
in “near vegetative states” and pose no risk to public safety; however, the prison system is
prohibited from releasing them to a nursing-home facility where their care would be far less
expensive. Texas has budgeted nearly $300 million for health care for prisoners in 2005;
however, cuts to the budget and staff due to the financial crisis come at an inopportune time, as
elderly prisoners continue to accumulate in prisons across the country as a result of life
sentences.
Lifers who are eventually released will obviously result in lower costs of incarceration, but even
at an average of 29 years documented in this report, these costs will approach $600,000 or more.
As a matter of public policy, therefore, the issue becomes whether to invest up to $1 million in

49
50

John Moritz, “With Sick Inmates Costing Millions, State Sees Crisis,” The Houston Chronicle, March 27, 2004.
Vanessa Blum, “Gray Area,” Legal Times, March 29, 2004.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

26

_____________________________________________________________________________________________

public funds to imprison an offender for life, or to impose a substantial sentence, but less than
life, at lower cost.
Length of prison term’s impact on recidivism
The relatively low rate of recidivism for lifers in comparison to other offenders is no doubt due
in significant part to the length of time they have spent in prison. In general, older offenders
have lower rates of recidivism than younger offenders. From a public safety perspective, the
policy issue concerns how to maximize the efficiency of the criminal justice system. To take an
extreme example, locking up all shoplifters for life would reduce any recidivism on their part,
but would obviously represent an inefficient use of public resources.
In a review of the limited studies that have been conducted to assess the effect of time served on
recidivism, researchers at the Washington State Institute for Public Policy concluded that an
individualized assessment of these dynamics was suggested by the experience to date:
… the effects of … the length of time served on recidivism are perhaps offenderspecific…. For some offenders, incarceration and longer confinement increase the
risk of recidivism. For other offenders, recidivism rates will either be unaffected
or reduced by longer terms of incarceration. It is possible that for some types of
offenders, there is an optimum length of sentence which minimizes recidivism. 51
To the extent that lifers are considered for release, policymakers need to assess relative rates of
recidivism and costs for varying periods of incarceration. For example, releasing lifers on
average after 20 years would save substantial funds but might increase recidivism, whereas
incarcerating lifers for 40 years would be very costly but might result in less recidivism. In any
of these scenarios, the challenge is to view corrections spending in broad terms. That is, to the
extent that shorter prison terms yield cost savings, how can those funds be invested in public
safety approaches, and what are the relative benefits that would be achieved? And conversely, if
longer prison terms result in higher costs, from where will those funds be drawn and what will
that impact be?
Reducing recidivism
With more than 600,000 persons coming out of prison each year, including about 3,000 lifers,
recidivism is clearly a key concern for crime policy. As is true for all offenders, a focus on
reducing recidivism for lifers involves a variety of strategies – programming in prison, risk
assessment, and transitional services upon release.
Growing interest in recent years in the reentry of prisoners to the community has focused
attention on factors that contribute to reducing recidivism among released prisoners. Generally,
scholars have concluded that some rehabilitation programs in prison are effective in reducing
recidivism. These include academic and vocational skills training, cognitive skills programs
focusing on goal-setting and problem-solving, and drug treatment. A 2001 study by the
51

Lin Song and Roxanne Lieb, “Recidivism: The Effect of Incarceration and Length of Time Served,” Washington
State Institute for Public Policy, September 1993.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

27

_____________________________________________________________________________________________

Washington State Institute for Public Policy found that a number of these programs lowered
recidivism by 5-10%, and some by as much as 20-30%. A fiscal analysis of interventions both
within and outside the criminal justice system demonstrated crime reduction savings of as much
as $7 for every dollar spent. 52 Such findings suggest that preparing lifers for release through
targeted programming can result in significant public safety benefits.
Research has also documented ways in which the post-prison environment affects recidivism.
Paul Gendreau and colleagues have used meta-analysis techniques to assess the various factors
related to recidivism. 53 They categorize these as either static or dynamic factors. The former are
factors that cannot be changed, such as adult criminal history, race, age, and family rearing
practices. Dynamic factors are circumstances for which interventions can be developed,
including such ones as companions, substance abuse, and social achievement. Many of these
factors have a statistically significant, though modest, impact on recidivism, but the researchers
find that as a group, the dynamic factors are more predictive. This finding suggests that targeted
interventions with released prisoners can be effective in reducing future criminal activity.
This analysis does not suggest that all lifers necessarily be released after serving a certain
number of years. Such decisions ultimately are the responsibility of a parole board or governor,
depending on state policies. But given the very substantial fiscal costs of housing aging
prisoners, it does imply that increasingly longer incarceration of lifers is not necessarily the most
efficient use of public safety funds. This is particularly the case in regard to the increasing
frequency of life without parole sentences. The recommendations section of this report provides
a series of policy and practice strategies that can be employed to provide appropriate
consideration of sentencing and release issues that address public safety concerns.

52

Steve Aos, Polly Phipps, Robert Barnoski, and Roxanne Lieb, The Comparative Costs and Benefits of Programs
to Reduce Crime , Washington State Institute for Public Policy, May 2001.
53
See discussion of these findings in Joan Petersilia, When Prisoners Come Home , Oxford University Press, 2003.
pp. 150-153.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

28

_____________________________________________________________________________________________

LIFERS IN INTERNATIONAL CONTEXT
As is true of most criminal justice policy in the U.S., lifer sentencing is generally far more
excessive than in most industrialized nations. While there are limited comparative data, general
international norms are moving away from large-scale use of life imprisonment.
A comparative study of the U.S., England and Wales, and Germany by law professor Dirk van
Zyl Smit found that the U.S. employed life sentences more frequently, and with more
restrictions, than the two other nations.54 Examining prison data for 1999, the author found that
10.7% of U.S. prisoners were serving life sentences, in comparison to 8.4% in England and
Wales, and 3.1% in Germany. (These figures only report on absolute numbers, and do not
account for differences in crime rates.)
Among European nations, there are mixed trends in recent years. While the French Parliament
recently enacted a law providing for life without parole, Norway and Portugal have abolished life
imprisonment entirely. In Slovenia, the maximum term of imprisonment is twenty years, and a
German Constitutional Court decision has held that life imprisonment must generally include the
possibility of release.55
While the European Court of Human Rights has not opposed the use of life sentences for adults,
it has held that discretionary life terms imposed by a judge must include the possibility of release
from prison. 56 And for juveniles, the United Nations’ Convention on the Rights of the Child
(1989) expressly opposes life sentences for anyone less than 18 years of age.57
A key distinction among industrialized nations often lies in regard to the use of sentences to life
without parole. As we have seen, one quarter of the U.S. lifer population is now serving such a
sentence. Even in European nations, which permit such a sentence, it is generally used only
rarely. In England, for example, all but 23 of the 4,206 life prisoners had a minimum term set
after which they would be parole eligible.
One factor that likely provides part of the explanation for the greater use of life imprisonment in
the U.S. is the application of the death penalty in the justice system, in contrast to virtually all
other industrialized nations. Since sentencing systems are generally established on a
proportional basis – more serious crimes are punished more severely – the presence of the death
penalty in the sentencing structure serves to exert upward pressure on the severity of the
penalties imposed for all offenses.

54

van Zyl Smit, Dirk, Taking Life Imprisonment Seriously, Kluwer Law International, 2002.
Albrecht, Hans-Jorg, “Post-Adjudication Decisions in Comparative Perspective,” in Michael Tonry and Richard
Frase, editors, Sentencing and Sanctions in Western Countries, Oxford University Press, 2001, p. 303.
56
Kurki, Leena, “International Standards for Sentencing and Punishment,” in Tonry and Frase, p. 355.
57
“Life Imprisonment,” ibid., p.6.
55

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

29

_____________________________________________________________________________________________

RECOMMENDATIONS
As this report has documented, the use of life sentences, and life without parole, has increased
dramatically over the past two decades. At the same time, the use of discretionary release
mechanisms has been substantially curtailed. The consequences of unnecessary or wrongful life
sentences are great, highlighting the importance of making appropriate decisions to incarcerate
for life from both a humanitarian and fiscal perspective. In calling attention to these
developments in a 2003 address to the American Bar Association, Supreme Court Justice
Anthony Kennedy noted that pardons have now become infrequent and the pardon process has
been “drained of its moral force.” Kennedy called upon the ABA to advocate to “reinvigorate
the pardon process at the state and federal levels.”
Life sentencing policies should incorporate a range of perspectives. These include the varied
goals of sentencing in such cases, the harm to and needs of victims, public safety objectives, and
the impact on costs and management of corrections facilities. The following represent key
policy changes that can address these issues:58
Defense Representation and Sentencing
•

Provide adequate defense representation

In recent years documentation of inadequate defense representation in capital cases has become
widespread. In Illinois and other states, this has contributed to innocent persons being sentenced
to death, while in 2003 the Supreme Court ruled in the Wiggins case that defense failure to
prepare mitigation evidence where such evidence exists and would likely have had an impact
constituted ineffective assistance of counsel. Problematic as these cases are, they suggest that
representation in lifer cases may be even more challenging, given that the levels of legal
representation and defense resources are generally lower than in capital cases. In order to reduce
the risk of error and to allow for full consideration of the appropriate sentence in serious cases,
states should provide adequate funding to assure adherence to standards of representation
promulgated by the American Bar Association and other professional bodies.
Such standards include sufficient funding for uniform statewide representation, investigation,
experts, and sentencing specialists. Caseloads must be reasonable so as to permit the rendering
of quality representation, “vertical representation” should be mandated to insure continuous
representation of the client through all trial stages, and counsel should be assigned with
appropriate training and experience to match the complexity of the case.59 Moreover, a full presentence report should be required by both a court service as well as the defense, which will
include a recent investigation into the defendant’s history. Finally, standards recommended by
the American Bar Association for indigent defense representation and for the conduct of the pretrial, trial, and sentencing stage should be followed in practice. In order to permit redress of a
wrongful or dubious conviction or sentence, state legislatures and/or high courts acting in their

58

Recommendations in this section have benefited greatly from those contained in Joan Petersilia, When Prisoners
Come Home, Oxford University Press, 2003.
59
American Bar Association, The Ten Principles of a Public Defense Delivery System, February 2002.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

30

_____________________________________________________________________________________________

supervisory roles should also reconsider Strickland’s restrictive limitations imposed on
meaningful review of defense counsel’s adequacy in any particular case.
•

Restore appropriate discretion to sentencing judges

The most significant change in sentencing policy over the past several decades has been the
limiting of judicial discretion. With the imposition of mandatory sentencing, “three strikes”
policies and similar measures, judges in many jurisdictions are now greatly restricted in the
sentences they can impose for many offenders. For lifers, the issue is generally not a question of
whether the offender should be sentenced to prison, since virtually all persons convicted of these
serious offenses would be imprisoned under any sentencing scheme. But the mandatory nature
of many penalties often unduly restricts the ability of judges to sentence based on the individual
circumstances of the offender and the offense. Restoring judicial discretion in such cases would
permit judges to evaluate the appropriate mix of sentencing objectives to be achieved in each
case while avoiding overly restrictive practices that do not contribute to public safety. In
addition, factors that permit a judge to sentence below a mandatory minimum or guideline
sentence should be expanded to include whether the defendant has a history of abuse at the hands
of the victim, a mental disability or illness, or was a juvenile at the time of the offense.
Prison Programming and Release
•

Restore professionalism to parole boards

In 1967, the President’s Crime Commission recommended that parole boards be staffed by
corrections professionals, as opposed to political appointees. Despite this, politically appointed
boards are the norm in almost all states, with two-thirds of states maintaining no standards for
professional qualifications for service. One exception is the state of Ohio, in which parole board
members are persons with strong backgrounds in criminal justice who are appointed by the
commissioner of corrections and serve in civil service capacities. The politicization of parole, as
noted earlier in the case of California, renders the prospect of a rational and empirical
consideration of each individual’s application for parole unlikely, with future electoral concerns
more likely to guide decisions. Instead, parole board members should have strong backgrounds
in corrections or social services so as to best assess risk and release decisionmaking.
•

Adopt risk-based policies for release

Persons sentenced to life represent a broad range of offenses and individual characteristics,
which in turn suggest varying prospects for public safety upon release from prison. While parole
guidelines were widely used in the 1970s and 1980s to promote a greater degree of equity in
release consideration, they have largely fallen into disuse as parole boards have lost much of
their discretionary power. Guidelines that incorporate risk assessments for lifers can be used as a
tool by professionals on parole boards to develop release decisions that restore accountability
and public safety as primary goals of the system.
The state of Texas adopted parole guidelines in 2001 that score inmates on a risk system based
on factors that include the severity of the offense, prior incarcerations, prison disciplinary record,

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

31

_____________________________________________________________________________________________

and age. By using such a system, state parole officials have available a guideline that builds
public safety prospects into the release decision.
Risk assessment instruments such as the Level of Service Inventory-Revised (LSI-R) are being
used by a number of states to identify offender needs and risks. The tool incorporates an
assessment of both static and dynamic factors of prisoners and parolees, including criminal
history, employment, and financial resources. While originally developed as a mechanism for
use by parole agents in the community, Washington state and others now use it to assess both
inmate needs in prison and service needs upon release.
•

Restore prison programming to prepare inmates for release

The growing interest in prisoner reentry has focused attention on the importance of providing
programming in prison that can better prepare inmates, including lifers, for release to the
community. In recent years, though, as the prison population has steadily increased,
programming and services within prisons have actually declined. Between 1991 and 1997, for
example, the proportion of state prison inmates nearing release who had participated in
vocational programs fell from 31% to 27%, along with a decline from 43% to 35% in education
programs. 60 Since research has demonstrated that participation in such programming can reduce
recidivism, public safety considerations suggest that restoration of program services would lead
to better preparation for release. This approach echoes the remarks by Supreme Court Justice
Anthony Kennedy in a 2003 address to the American Bar Association, in which he notes we
must “acknowledge that the more that two million inmates in the United States are human beings
whose minds and spirits we must try to reach.”
•

Expand release consideration for elderly, ill prisoners

The increasing use of life sentences will result in larger numbers of elderly and infirm prisoners.
Incarcerating such offenders serves no public safety objective and is very costly for corrections
systems. States should adopt policies that provide for careful review of such cases so that
societal interests in balancing competing objectives of punishment, public safety, compassion,
and cost control can best be achieved. While some states consider ways in which they can
release elderly and ill prisoners without risking public safety, strict legislative provisions in the
federal system make early release an unlikely option. The federal government should consider
adopting legislation to permit persons over the age of 60 who have served one-third of their term
to apply for early release.61

60
61

James P. Lynch and William J. Sabol, Prisoner Reentry in Perspective, Urban Institute, September 2001.
Blum, op.cit.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

32

_____________________________________________________________________________________________

Policy Changes
•

Eliminate life without parole in all but exceptional cases

Imposing a sentence of life without parole makes an assumption that an offender will never be
considered for release, no matter how old or changed in behavior. Experience suggests that
many persons sentenced to life in fact change substantially while in prison, both by expressing
genuine remorse for their actions and engaging in programming and changed attitudes. In states
that employ the death penalty, a sentence of life without parole is often viewed as a lesser
alternative, but the scale of such sentences -- 33,633 such persons in prison today – suggests that
these penalties are being imposed in a far broader range of cases. While policymakers may wish
to maintain the option of life without parole, its imposition should only be considered in cases
where either the circumstances of the offense or public safety considerations mandate its use.
•

Exempt juveniles from life sentencing

As the recent case of 12-year-old Lionel Tate has demonstrated, life sentences are particularly
inappropriate for juveniles. The Tate case demonstrates that juveniles should not be subject to
life prison terms because they do not have the maturity of an adult in aiding in their defense. In
addition, as noted by Judge Moore in the case of Nathaniel Abraham, life sentences represent an
entire rejection of the longstanding traditions of our treatment of juvenile offenders, which is that
rehabilitation should be considered as a primary objective when sentencing children.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

33

_____________________________________________________________________________________________

CONCLUSION
Victor Hassine
Born in Egypt and reared in New Jersey, Victor Hassine was convicted as an accomplice to murder
just months after graduating from law school in 1980, and sentenced to life without parole in
Pennsylvania. Despite suffering permanent physical disability as a result of prison violence, Hassine
has become a noted writer and creative artist. He is the author of Life Without Parole, a text used in
many universities, has written a play that has been performed in prison, and was named by Pen
America as one of the nation’s outstanding inmate authors. While in prison, Victor Hassine also cofounded the nation’s first accredited prison synagogue and three faith-based post-release transition
homes.

This report has documented the vast increase in the use of life sentences in the United States in
recent decades, resulting in one of every eleven persons in prison now sentenced in this manner.
These developments raise questions of public safety, use of tax dollars, and the efficacy of
criminal justice policies.
As we have documented, many lifers have been convicted of serious crimes and present an
immediate threat to public safety, but many others are housed in prison long after they are
dangerous due to overly restrictive parole and commutation policies. Additional numbers of
lifers would probably not have received such severe sentences had they had access to adequate
defense representation or had judges not been overly constrained by sentencing policies requiring
such prison terms.
The findings of this report do not suggest that long-term imprisonment is necessarily
inappropriate in any individual case, either for reasons of punishment or public safety. Nor do
they suggest that persons who have been victimized in many such cases have not suffered
greatly. But they do compel us to question whether the broad-scale imposition of such penalties
has resulted in the use of life imprisonment in ways that too often represent both ineffective and
inhumane public policy. As legislative bodies begin to reconsider the appropriate balance of
resources to produce public safety, the issue of long-term imprisonment should become an
important aspect of that discussion.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

34

_____________________________________________________________________________________________

APPENDIX – ESTIMATING TIME TO BE S ERVED IN PRISON

There are several methods available by which to develop estimates of time to be served in prison.
Each provides a different type of measurement for analyzing sentencing and release practices.
Estimates based on exit cohorts are analyses of the average amount of time served by persons
leaving prison in a given year. These figures generally underestimate prison time served for a
particular type of offense category since persons serving shorter sentences are overrepresented in
the cohort. Further, as noted by Sabol and Lynch, time served by persons exiting prison may
also differ from expected time to be served for recently admitted prisoners due to changes in
policy and practice.62
A second method is the expected time to be served/time served estimate, which is calculated
through inmate self-reports on how much time they have left to serve on their sentence. This can
be done with a cohort of prisoners upon admission, asking them about their sentence length. In
addition, it may be asked of the stock population in prison, in which prisoners report how much
time they have served and how much time they expect to serve until release. This approach
overcomes the bias in using the exit cohort method, but self-report estimates are fraught with
concerns of reliability. Even if an inmate does accurately recall the amount of time expected to
be served from a certain point in time, prison release practices change over time, often causing
this method to yield estimates that differ significantly from actual practice.
The methodology chosen for this report is the stock/flow analysis used by the U.S. Office of
Juvenile Justice and Delinquency Prevention in its projections of juvenile detention populations.
As developed by Butts and Adams, this method involves a computation of time served based on
a ratio between persons currently in prison and those being released.63 Although the stock/flow
method is subject to creating biased estimates if admissions or release patterns change
significantly over time, in the case of this project this approach is the least biased estimator and
the best indicator of length of stay over time. Since admissions have been increasing and
releases decreasing, both steadily and without any sudden deviations, the emergence of biased
estimates from this approach is not a serious concern.
Estimates for the stock population have been taken from the 1997 Survey of Inmates in State and
Federal Correctional Facilities for all persons in state prison serving a life sentence. Estimates
for the flow, or released, lifers for that year are estimated using data from the 1997 National
Corrections Reporting Program.
z = y / xrelease
In the above equation, z represents the years a person entering prison with a life sentence can
expect to serve prior to release. Expected time to serve is a function of the stock population (y),
62

Sabol, W.J. & Lynch, J.P. (2002). Sentencing and Time Served in the District of Columbia Prior to “Truth-inSentencing.” Washington, DC: U.S. Department of Justice. NCJ 191860.
63
Butts, J. & Adams, W. (2001). Anticipating Space Needs in Juvenile Detention and Correctional Facilities.
Washington, DC: Office of Juvenile Justice and Delinquency Prevention. NCJ 185234.

THE MEANING OF “LIFE”: LONG PRISON SENTENCES IN CONTEXT

35

_____________________________________________________________________________________________

which includes admissions, in a given year divided by the flow (xrelease), or releases, during that
year. The underlying principle of the stock/flow ratio is that in a state of equilibrium, with
steady entries and exits from prison, the length of time to be served is a function of the stock
population at time t and the releases (flow) during that same period. The size of the “exit door”
dictates at what rate lifers accumulate in prison.