Nothing to Lose? An Examination of Prison Misconduct Among Life-Without-Parole Inmates

Abstract

Inmates serving life-without-parole (LWOP) have been described as having “nothing to lose” by virtue of their sentence, leading to an assumption that they are more prone to disciplinary violations. This study refutes such an argument and is consistent with research demonstrating that LWOP inmates do not pose a disproportional risk for disciplinary misconduct. Results from our study comparing LWOP with parole-eligible, life-sentenced (LWP) inmates revealed neither significant differences in the total violation count nor the time to commission of an act of disciplinary misconduct. Trajectories of misconduct showed slightly higher prevalence of misconduct among LWOP inmates, a pattern of declining prevalence during the first 18 months of confinement, and a convergence with LWP inmates thereafter. The failure of assumptions of high violence risk for LWOP inmates has important public policy and correctional implications.

Keywords

life-without-parole prison violence prison misconduct long-term inmates

Introduction

Throughout most of the 20th century, inmates sentenced to life imprisonment could be released on parole at some point in the future. The federal system had in place a policy of parole review after 15 years of incarceration dating back to 1913 (Nellis, 2013). Eligibility for parole among such lifers was widespread and consistent with indeterminate sentencing models in force in all states. A shift to structured sentencing and tough on-crime policies (e.g., three strikes laws, truth-in-sentencing) enacted since the 1980s has increased the number of persons subjected to life-sentences and the amount of time to be served prior to parole eligibility. Currently, 49 states utilize life-without-parole (LWOP) as a sanction in certain instances, while six states and the federal system have completely abolished parole for any life-sentenced prisoners (American Civil Liberties Union [ACLU], 2013; Death Penalty Information Center [DPIC], n.d.).The most recent figures available show that the number of inmates nationwide serving LWOP sentences was 53,290 as of 2016, a 328% increase in the LWOP population since 1992 (Nellis, 2017). As of January 2015, there were 4,436 inmates serving LWOP sentences in the Federal Bureau of Prisons (Schmitt & Konfrst, 2015).

Historically, correctional personnel espoused a belief that (parole-eligible) lifers (hereafter referred to as LWP inmates), along with other long-term inmates, were generally among the most manageable prisoners (Flanagan, 1980, 1995; Wardlaw & Biles, 1980; Zamble, 1992). Perhaps this was owed to an acceptance of their fate, the effect of aging, the process of adjusting to the prison environment that comes with a lengthy incarceration, growing accustomed to minor privileges accorded for good behavior, or the hope of eventual release (Johnson & Dobrzanska, 2005). With the advent of LWOP sentences, fears surfaced about an emerging group of inmates who, arguably having nothing to lose, would become a “new breed of super-inmates prone to violence and uncontrollable behavior” (Stewart & Lieberman, 1982, p. 16).

While some have suggested that LWOP inmates be housed under austere and extremely restrictive conditions of confinement, barring misbehavior, they are generally housed alongside parole-eligible inmates under similar conditions ranging from medium to maximum security confinement (Blecker, 2013). Given that so many more inmates are receiving sentences of LWOP and will likely serve the bulk of their sentences under conditions that allow them to interact routinely with other inmates and correctional staff, the question arises as to what level of difficulty these inmates pose for prison administrators, staff, and other inmates throughout their incarceration. The current study seeks to shed light on this question by examining conduct violations from a jurisdiction housing LWOP inmates alongside other inmates in maximum security prisons, wherein classification decisions are based on adjustment to confinement rather than an inmate’s eligibility for parole.

Theoretical Perspectives

Various models have been offered to explain prison maladjustment, broadly encompassing personal attributes and experiences of inmates interacting with institutional features. The importation model emphasizes prison misconduct derived from an inmate’s personality style, traits, and values brought into prison as an expression of disruptive attitudes, aggressive behavior, and aberrant subcultural values (Irwin & Cressey, 1962; Steiner, Butler, & Ellison, 2014). The deprivation model focuses on the influence of the oppressive physical and environmental features of prison life on inmate adjustment (Steiner et al., 2014), aptly described by Sykes (1958) as the “pains of imprisonment.” The unique environmental/situational factors and managerial styles of prison administrators and their policies have also been offered as contributing factors to prison maladjustment (Dilulio, 1987). The career criminal model hypothesizes that as a fundamental part of prison life, an individual continuously perpetuates a criminal mind-set through behavioral deviance and violence, although not necessarily in a linear fashion (DeLisi, 2005, 2016; DeLisi & Piquero, 2011).

General Strain Theory (GST), originally promulgated by Agnew (1985, 1992), has demonstrated broad relevance within the criminal justice system as a way to integrate criminological theories. A specific formulation has been applied to understand the progression or cessation of criminal behavior and violence during periods of incarceration, emphasizing the strain that individuals face upon entering prison and how they react to such experiences (Blevins, Listwan, Cullen, & Jonson, 2010; Morris, Carriaga, Diamond, Piquero, & Piquero, 2012). Prison incarceration produces a number of strains leading to negative emotional and behavioral reactions depending on an individual’s ability to cope with prison adversity. Blevins et al. (2010) characterized prison strain in relation to the elimination of positively valued goals and application of pernicious occurrences. Such burdens include loss of freedom, autonomy, and independence; application of restrictions/rules; absence of privacy; restricted privileges, safety concerns around victimization, or threat of harm; and other factors related to the staff or institution. Some inmates faced with chronic or cumulative strain are overwhelmed and may cope through deviant behavior, such as committing disciplinary violations and perpetrating violence on others.

Of particular interest in the current study is the stress engendered by lack of parole eligibility on the progression of prison infractions among murderers serving LWOP. The following section provides an examination of available studies on disciplinary infractions among inmates coping with the strains associated with facing a lifetime of confinement.

Prior Studies of the Behavior of LWOP Inmates

The first study to examine misconduct among LWOP inmates was completed in Louisiana, one of the earliest states to implement a “natural-life term” in 1977 (Williamson, 1985). Inmates sentenced to life prior to that date could establish parole eligibility after serving 10.5 years. The findings showed that a group of 50 LWOP inmates was more disruptive, in terms of their rate of rule violations, than a group of 50 LWP inmates, by a factor of 2. The comparison of rates, however, between LWOP and LWP inmates was confounded by a difference in observation periods. Since the law changed in 1977, the LWOP inmates were part of a more recent cohort that had served an average of only 34 months, compared with LWP inmates who were part of an older cohort that had served an average of 126 months. Differences between the groups could be expected given a generally declining pattern of rule violations due to cohort effects, aging of the samples, and adjustment over the course of time to the prison environment.

Utilizing a similar methodology, Morris, Longmire, Buffington-Vollum, and Vollum (2010) investigated differential parole eligibility on capital murderers’ propensity to commit conduct violations. During 1991 in Texas, parole eligibility changed from 15 years (noncapital life) to 35 years (capital life), and 2 years later, to 40 years for inmates sentenced to life imprisonment for capital murder. While technically allowing for parole eligibility, the newer statute certainly limited the chances of a capital life inmate being released from prison. The authors found that capital lifers were no more likely than noncapital lifers to commit rule infractions during their first 3 years of confinement. Survival analysis showed that hazard rates (commission of a violent act during a given time interval) were significantly lower for capital lifers than for noncapital inmates. Such findings highlight the adaptation of these life-sentenced inmates to the stresses associated with lengthy confinement. Their study, however, suffered from the same limitation as the Louisiana research, namely, potential confounds resulting from different follow-up periods by cohort which could have influenced inmates’ level of offending and/or the level of reporting during either period.

Research on LWOP inmates in other jurisdictions has supported findings from the Texas investigation described above. The largest study of LWOP inmates to date (n = 1,897, in Florida) revealed that hazard rates for these inmates (time to the commission of a violent act during 1998-2003) were not significantly different from those of other long-term, parole-eligible inmates (Cunningham & Sorensen, 2006). A study of the federal system showed that 145 LWOP inmates were found to commit various levels of rule violations at a rate similar to a comparison group of high-security inmates during 2001-2005 (Cunningham, Reidy, & Sorensen, 2008). The main drawback to both, however, is that the comparison groups included a cross section of high-security inmates convicted of various types of crimes serving sentences of varying lengths.

Three studies have included information about the misconduct of LWOP inmates in Missouri. The first compared rule infractions among 323 LWOP inmates convicted of capital murder with a sample of 232 LWP inmates convicted of second-degree murder during 1977 through 1992 (Sorensen & Wrinkle, 1996). A regression model found no significant difference between the groups in terms of their total number of rule violations. The second study broadened the comparison group to include 2,199 parole-eligible inmates, but was limited to one maximum security facility—Potosi Correctional Center (PCC; Cunningham, Reidy, & Sorensen, 2005). Controlling for other relevant variables in a logistic regression model, LWOP inmates were shown to be significantly less likely to commit a violent rule violation in comparison with term-sentenced inmates. In the third, a replication, it was reported that LWOP inmates (n = 702) convicted of first-degree murder engaged in equivalent or lower rates of serious and violent misconduct when contrasted with the behavior of 3,000 parole-eligible inmates serving time in PCC during 1991 through 2015 (Cunningham et al., 2016). Yet, the findings from both of the PCC studies were skewed by the selective migration of term-sentenced inmates to the high-security prison during those time periods for bad behavior elsewhere in the state prison system.

Such results continue to illustrate that neither a capital or first-degree murder conviction nor an associated sentence of LWOP is indicative of elevated rates of maladjustment warranting special classification or housing considerations. This research also supports a long-held belief that inmates facing lengthy prison terms have lower rates of disciplinary violations than inmates serving shorter sentences, but such a conclusion is mediated by age (Cunningham & Sorensen, 2006; Morris et al., 2010; Toman, Cochran, Cochran, & Bales, 2015). Younger inmates and those in the early phases of incarceration have been determined to commit more rule violations than inmates who have learned to adjust to the rigors of prison over time as they age (Cunningham et al., 2008; Flanagan, 1980, 1995; Toch & Adams, 1989). Higher levels of educational attainment and being married have been also been shown to lead to a more positive adjustment to incarceration, including lower levels of misconduct (Cunningham et al., 2005; Cunningham et al., 2008; Steiner et al., 2014). Alternately, prior criminal record and mental health issues are often related to adjustment problems, including higher levels of misconduct (Cunningham et al., 2016; Schenk & Fremouw, 2012).

A limited number of studies have investigated the trajectory of infractions for inmates throughout their incarceration, particularly among inmates serving LWOP or extended prison terms. Sorensen, Wrinkle, and Gutierrez (1998) analyzed patterns of rule violations in a sample of murderers sentenced to LWOP and LWP. Results revealed an overall low base rate of rule violations and a stable pattern of violations over time, consistent with prior research on long-term inmates (Flanagan, 1980). Other than slightly lower rates of rule violations during the first year, LWOP inmates’ trajectory of prison misconduct closely tracked that of LWP inmates. Cunningham and Sorensen (2006), relying on a similar strategy, did not find elevated rates of violence during initial incarceration for LWOP inmates, but rather a low rate of assaultive violations held constant across 6 years. Findings from each of these investigations ran contrary to other work examining the pattern of behavior among murderers, which found a declining rate of infractions early in their sentences and then a leveling off in the rate of infractions (Reidy, Cunningham, & Sorensen, 2001; Sorensen & Pilgrim, 2000).

The current study was designed to remedy limitations with earlier research and, in doing so, more clearly identify any differences in misconduct that could be tied to parole eligibility among lifers. This inquiry includes LWOP inmates and a comparison group restricted to LWP inmates serving time during the same observation period under conditions of maximum security confinement, thus enhancing equivalence among the groups and circumstances of confinement. Numerous important preprison control variables are included to further control extraneous variation in models estimating total infraction counts and time to commission of an initial act of misconduct. Trajectories of violations between LWOP and LWP inmates are also charted during their first 3 years of imprisonment.

Method

Sample

The sample includes LWOP inmates entering the Missouri Department of Corrections (MDOC) for the period August 2006 through February 2014. The starting point was chosen due to a change in the manner in which inmate rule violations were recorded. February 2014 was chosen as the cutoff date for sample selection to allow a full year follow-up period of observation; data were downloaded at the end of February 2015. During this time period, 189 LWOP inmates entered the prison system, with 93% (n = 176) having been convicted of first-degree murder and 7% (n = 13) convicted of sex offenses, either persistent or committed against a child. To best control for potential confounds, the analysis herein is limited to the 176 LWOP inmates convicted of murder.

For comparison purposes, all 319 LWP inmates were extracted from the larger cohort of inmates entering the MDOC during the same time period. Again, the final comparison sample was limited to inmates convicted of murder (n = 172), although sentences of LWP are generally issued to defendants convicted of second-degree murder. Despite this difference in degree, the group of LWP inmates is the closest equivalent to LWOP inmates for the purposes of the current comparison (Sorensen & Wrinkle, 1996). While LWOP inmates were not eligible for release on parole, those serving LWP sentences could become eligible for parole after serving 50% of their maximum sentence, calculated as 15 years on a 30-year life-sentence. This group of inmates was chosen to equalize the comparison group as much as possible in terms of their crimes, as well as the conditions of confinement under which they would serve their sentences. Both groups of inmates were classified to C-5 institutions, currently five facilities, upon entering the prison system.

Measures

Data were derived from computerized records of inmate misconduct provided by the MDOC and from which the prevalence and frequency of disciplinary misconduct were calculated. The Missouri Offender Rulebook assigns a specific definition and code number for each infraction. Outcomes in the current study were based on an omnibus measure that includes all substantiated incidents of disciplinary misconduct regardless of level or type. Depending on the specific analysis, disciplinary misconduct is coded in various ways including a raw count of conduct violations, number of months to commission of the initial act of misconduct after entering prison, and the prevalence of misconduct during 6-month intervals. Because time at risk varies, time served (in years) during August 2006 through February 2015 was also included as a control variable in the regression models. The main variable of interest was coded as a binary indicator, wherein LWP = 0 and LWOP = 1.

Several variables shown to be important correlates of inmate misconduct in the literature were selected as control variables for the analyses (Steiner et al., 2014). Most influential among these is age at intake, coded as a continuous variable. Two other dichotomous demographic variables were also included: marital status, married at intake = 1, else = 0; and race, White = 1, else = 0. Marital status was dichotomized in such a manner because marriage has been shown to be an indicator of social stability. Race was dichotomized in such a manner because there were very few inmates, less than 1%, of racial descent other than Black or White in the prison system. Ethnicity was not included because the variable was recorded as Hispanic in only 3.7% of the cases and unknown in 12.6% of the cases. The total number of prior felony convictions was selected as a measure of the extent of an inmate’s criminal history.

Three needs-assessment codes indicating the level of inmate need or the severity of a condition were included. A classification specialist assigns these codes based on inmate interviews and records reviews upon an inmate’s entry to prison. The codes range from 1, indicating the lowest level of need, to 5, indicating the highest level of need. Mental health codes at intake, as determined by screening assessments and mental health staff, include 1 = no treatment needs, 2 = mild level of treatment needs, 3 = moderate level of treatment needs, 4 = serious functional impairment due to mental disorder, and 5 = severe functional impairment due to mental health disorder. Educational needs codes are as follows: 1 = educationally prepared, 2 = minimal educational impairment, 3 = mild educational impairment, 4 = moderate educational impairment, and 5 = severe educational impairment. Institutional risk was based on prior institutional adjustment, misconduct, and escape history, and ranged from 1 = low risk to 5 = high risk. These needs-assessment measures are relied on by classification personnel in determining inmate security and custody levels, housing assignments, and programming selections.

Procedures

First, descriptive statistics are presented for LWOP inmates in comparison with the LWP inmates to determine the extent of similarity and differences between them. T tests were computed to test mean differences between the groups; chi-square was used to test percentage differences in categorical comparisons. Second, a negative binomial regression model was used to predict the total number of rule violations. Negative binomial regression is the most appropriate method of analysis when modeling over-dispersed count data (Walters, 2007), which is the case in the current situation (M = 10.5, SD = 13.5). Third, survival analysis was used to predict the time (in months) to the commission of a conduct violation. Used to model time to an event, survival analysis is employed on a sample, which includes some “survivors,” those for whom the outcome event had not yet occurred by the end of the observation period (Luke & Homan, 1998). Observations for the survivors are then “censored” at the end of the observation period to indicate survival up to the point that the observation period closed. Specifically, proportional hazard models are estimated using Cox regression to determine the extent to which LWOP and other predictor variables influenced the likelihood of committing a conduct violation across time intervals. Finally, the pattern of rule violations was plotted for LWOP and LWP inmates during their first 3 years of incarceration.

Results

Table 1 presents the sample characteristics for LWOP and LWP inmates. In terms of demographics, LWOP and LWP inmates were similar. Slight differences in age, race, and marital status were not significant. The groups averaged four to five prior convictions, also not a statistically significant difference. The only significant difference in intake codes showed that LWOP inmates had a lower level of educational need in comparison with LWP inmates. While time served ranged from 1 year to 8.6 years, both groups served an average of 5.0 years. Similarly, the prevalence of rule violations was above 90% for both groups, with an average total count of 10 to 11 violations, again not statistically significant.

Table 1.

Sample Characteristics for Life-Without-Parole and Life-With-Parole Inmates.

Sample characteristics	Life-without-parole(n = 176)			Life-with-parole(n = 172)
Sample characteristics	%	M	SD	%	M	SD
Age at intake		34.0	11.3		36.4	13.6
White	41.5			48.3
Married	17.6			15.7
Mental health code		1.7	0.9		1.8	0.9
Education code		2.5*	1.8		2.9	1.9
Institutional risk		1.3	1.0		1.2	0.8
Prior convictions		5.1	4.7		4.1	6.7
Time served		5.0	2.1		5.0	2.1
Disciplinary violations (Prevalence)	91.9			92.4
Disciplinary violations (Frequency)		11.0	12.9		10.0	14.1

Note. The t tests were computed to test mean differences, and χ² was used to test percentage differences.

p < .05.

Table 2 shows results from a negative binomial regression model predicting the total count of conduct violations. All of the demographic and predictor variables were related to the outcome in the expected direction, consistent with prior studies of misconduct (Steiner et al., 2014). Age, White, and married were all negatively related to the commission of conduct violations. Mental health problems, educational needs, and institutional risk were all positively related to the commission of conduct violations, as was the number of prior convictions. Of utmost interest, a sentence to LWOP was not significantly related to the total number of conduct violations after controlling for other demographic and predictor variables.

Table 2.

Negative Binomial Regression Model Predicting Disciplinary Conduct Violation Count.

Predictor variables^a	95% CI for Wald
Predictor variables^a	B	SE	Lower	Upper	Wald	Significance
Age at intake	−.037	.005	−.047	−.027	53.566	.000
White	−.533	.107	−.743	−.323	24.708	.000
Married	−.297	.179	−.648	.055	2.739	.098
Mental health code	.269	.057	.157	.381	22.052	.000
Education code	.097	.029	.041	.153	11.356	.001
Institutional risk	.204	.054	.099	.310	14.394	.000
Prior convictions	.016	.009	−.002	.034	3.178	.075
Life-without-parole	.102	.104	−.103	.306	.951	.330

Note. Model includes intercept and time at risk (not tabled).

Model χ² = 63.876; p < .001.

Results from the proportional hazard model are reported in Table 3. The relationship between predictors and the time to failure (months to commission of a conduct violation) mirrored the relationships found in the negative binomial model presented in Table 2. The exp b shows that White inmates commit an infraction during a given interval at a rate just over half that of African American inmates, holding other factors constant. Mental health need and institutional risk each increased the likelihood of a conduct violation by about one third for each incremental increase in scores. Most importantly, LWOP was not shown to influence time to commission of a conduct violation by an inmate.

Table 3.

Proportional Hazards (Cox Regression) Model Predicting Months to Failure (Commission of a Disciplinary Conduct Violation).

Predictor variables^a	B	SE	Wald	Significance	Exp(B)	95% CI for Wald
Predictor variables^a	B	SE	Wald	Significance	Exp(B)	Lower	Upper
Age at intake	−.030	.005	32.783	.000	0.970	0.960	0.980
White	−.554	.122	20.560	.000	0.575	0.452	0.730
Married	−.303	.162	3.485	.062	0.739	0.537	1.015
Mental health code	.276	.066	17.651	.000	1.318	1.159	1.499
Education code	.091	.031	8.304	.004	1.095	1.029	1.165
Institutional risk	.300	.067	20.034	.000	1.350	1.184	1.540
Prior convictions	.017	.009	4.167	.041	1.018	1.001	1.035
Life-without-parole	−.047	.115	0.170	.680	0.954	0.762	1.194

Model χ² = 116.269; p < .001.

The final exercise involved charting the pattern of rule violations for both groups during their first 3 years of incarceration. Figure 1 shows the prevalence of conduct violations occurring during each of the 6-month intervals. Both groups experienced their highest prevalence of violations during the first 6-month interval. Thereafter, both groups experienced a decline during the second and third 6-month intervals. An average difference of about 10 percentage points persisted between the groups through the first 18 months of confinement, with LWOP inmates having the higher prevalence. The gap continued to narrow thereafter, and the prevalence of conduct violations among the groups stabilized. By the last 6-month interval in the series, the prevalence of infractions between both groups was about 45%, after which the pattern (not charted) remained stable and unremarkable.

Figure 1.

Prevalence of disciplinary infractions committed by life-without-parole and life-with parole inmates during the first 3 years of incarceration.

Discussion

Inmates serving LWOP continue to be viewed by elected officials, prosecutors, jurors, and others as more prone to disciplinary violations, particularly violent actions toward other inmates and staff because they “have nothing to lose” (McInnis, 2003; Sorensen & Cunningham, 2007; U.S. v. Savage, 2013 ). This study refutes such an argument, and is consistent with research reviewed above demonstrating that LWOP inmates do not pose a disproportional risk for disciplinary misconduct. Results comparing LWOP with LWP inmates revealed neither significant differences in the total violation count nor the time to commission of an act of disciplinary misconduct. Trajectories of misconduct showed a slightly higher prevalence of misconduct among LWOP inmates, a pattern of declining prevalence during the first 18 months, and a convergence with LWP inmates thereafter. Similar results were obtained by Toman et al.’s (2015) investigation of sentence length. The trajectories found in our study and that of Toman demonstrate that inmates facing extended sentences, especially LWOP, initially anticipate the reality of decades, if not a lifetime of confinement, with all the attendant fears, concerns, and losses of freedom, family ties, and social relatedness with the outside world. Following these initial adjustment problems, most inmates likely reduced their proclivity for misconduct through changes in attitudes and expectations as they learned to cope and adapt to their environment (Leigey, 2010, 2015).

Initial adjustment problems may also be indicative of imported “street mentality” or factors such as alcohol/drug addiction (not included in the dataset) that require some mellowing period. Inmate interviews reflect that the deprivations and strains of lifetime incapacitation, such as loss of liberty and privacy, inability to make choices in their daily lives, monotonous daily routines, loneliness, and the overwhelming feeling that life has ended, are concerns for which these inmates are least equipped to manage (Leigey, 2010, 2015). How do LWOP inmates adjust to the challenges of an empty and meaningless existence within an oppressive and sometimes violent setting? Changes in adaptation by these long-term inmates to the rigors of incapacitation have been attributed to self-generated improvement in emotional reactivity and behavior, and finding a social niche (Flanagan, 1995; Leigey, 2015; Toch & Adams, 1989). Rather than deteriorating, research and inmate interviews illustrate that LWOP inmates demonstrate self-motivation to enhance their daily lives and make the most of the meager incentives offered in an austere, highly monitored, and restrictive setting (Crawley & Sparks, 2006; Leigey, 2015).The decreasing trend in disciplinary infractions by LWOP inmates demonstrated in our study supports findings from the qualitative studies described above.

Our findings are also consistent with the views of current prison administrators that inmates serving LWOP, like other long-term inmates, can be safely managed using modern classification principles, incentives, and disincentives. Flanagan (1980, 1995) described the view among correctional administrators that long-term inmates embraced a more compliant institutional deportment because the small incentives derived from cooperative adjustment become increasingly important as the duration of incarceration stretches into decades. Lifers, especially, savor the meager privileges available, which in turn are strongly reinforcing positive behavior to avoid losing these privileges. Sentence planning for LWOP inmates means creating more realistic and meaningful programming and incentives that not only benefit the inmate but can also foster better relations between inmates and staff. Small inducements such as access to a range of offender recreation programs, TV, visitation, decent paying prison jobs, commissary privileges, and novel incentives reinforce a common desire of most long-term inmates to “do the easiest time possible.” An example of a popular incentive in Missouri is the “Puppies for Parole” program that allows dogs from the local animal shelter to live with well-behaved inmate handlers behind prison walls to socialize the dogs for later adoption. Similarly, the prison system introduced quarterly inmate picnics with family and permitting homemade food, for those who maintain discipline-free behavior for extended periods of time. Such programming can reduce the “pains of imprisonment” and create an atmosphere of reduced stress and a lower likelihood of inmate misconduct.

While the policy implications are clear in that LWOP inmates can be housed among, and controlled in the same manner, as LWP inmates, it should not be read as support for the practice of sentencing defendants to LWOP. In fact, the United States is out of step with other Western countries in this practice. Recently, the European Court of Human Rights, the judicial body that adjudicates compliance with the European Convention for the Protection of Human Rights, ruled in a landmark decision that imposing life with no parole possibility constituted a violation of human rights ( Vinter and Others v. The United Kingdom, 2013 ). This decision is binding on all signatories of the European Convention on Human Rights, which requires member countries to have a mechanism in place for life-sentences to be reviewed within 25 years of their imposition. Many European countries go a step further and limit the number of years that may be served by a life-sentenced prisoner. For example, Anders Breivik, convicted of murdering 77 victims in an act of domestic terrorism, was sentenced to the maximum of 21 years in prison under Norwegian law (Blecker, 2013). However, the legal and political landscape in the United States that emphasizes retribution, law and order, and “tough on crime” policies means that the number of inmates facing a lifetime of incapacitation will continue to increase. Prison administrators will be faced with the need to find creative solutions to address the multitude of hurdles inherent in managing this population.

Limitations should be noted in regard to the current methodology. First, this study relied on official data, which is subject to underreporting and unreliable documentation of less serious incidents (Daggett & Camp, 2009; Wolff, Blitz, Shi, Siegel, & Bachman, 2007). Quantifying the extent of unreported misconduct is difficult to achieve and likely fluctuates with institutional controls, such as the discretionary power of correctional officers in reporting or defining infractions, unequal treatment of certain inmates or groups, budgetary restraints on staffing levels, and the degree of supervision in different facilities (Daggett & Camp, 2009; Drury & DeLisi, 2011). In particular, with the findings related to race, it is not possible to determine here whether White inmates committed fewer violations or were simply less likely to be officially written up by correctional officers, who are also more likely to be White (Steiner & Wooldredge, 2009). While important to note, the potential for differential reporting by race should not influence the key finding herein, that lack of eligibility for parole does not influence the level of misconduct for two reasons. First, there was a slightly higher percentage of Blacks (not significant) in the LWOP sample, so racial bias in sanctioning should be expected to impact the LWOP group more that the LWP group. Second, including an indicator for race in the regression models controls for the combined effects of behavior and reporting in estimating the net effect of parole eligibility.

A second limitation relates to the lack of control for institutional-level variables. While all of the inmates were housed in C-5 maximum security institutions, it is likely that managerial practices, programming, housing, incentives, prison-based treatments, medications, and other variables that have been shown to influence inmate misconduct varied across units and even within units over time (Steiner & Wooldredge, 2008). The inclusion of such measures in the current analysis was thwarted by the lack of available information. Even the potential inclusion of a control for each unit is complicated by the movement of inmates, typically the least well-behaved, across C-5 institutions during the observation period. Predictors achieving marginal significance should, therefore, be viewed cautiously due to an inability to factor in institutional-level variation in misconduct. The inability to consider institutional variation, however, is less likely to influence null results, namely, that the LWOP indicator is not a significant predictor of misconduct.

The comparison group presents additional concerns. If an inmate was sentenced to life and did not have an indicator in their record expressly prohibiting the possibility of parole, they could be included in the comparison group. No attempt was made to calculate what is often referred to as an “effective” life-sentence, meaning that the inmate would almost certainly die in prison (Henry, 2012; Nellis, 2013). The LWP group likely contained an undetermined number of inmates who were serving what is effectively a natural life-sentence either through a combination of terms to be served consecutively, habitual offender laws, or even a singular sentence of such duration that it extends beyond an inmate’s life expectancy. The data file received from the MDOC precluded our ability to sort out or estimate such sentences among the LWP comparison group with any degree of accuracy; therefore, some of the inmates in the LWP sample likely had little or no hope of parole. Alternately, it is possible that LWOP inmates, due to appeals in progress and/or hope of gubernatorial clemency, still held out hope of being paroled at some point in the future.

LWP inmates also differ somewhat from LWOP inmates in a manner that, by definition, cannot be controlled. A conviction for first-degree murder, a requisite of LWOP in Missouri, turns on proof of “deliberation,” an element not required for a second-degree murder conviction. While this difference adds some degree of uncontrolled extraneous variance into the analysis, this variation is mitigated by the fact that the crimes and state of mind among second-degree murderers sentenced to LWP often mimic those of first-degree murderers sentenced to LWOP. First, the fact patterns of the crimes are often identical. For instance, felony murders wherein deliberation cannot be proven are categorized as second-degree murder. Second, many of the second-degree murder convictions and subsequent sentences to LWP resulted from plea bargains.

Future research on the misconduct of LWOP inmates should take these limitations into consideration. Qualitative studies, such as interviews with LWOP inmates (Johnson & Dobrzanska, 2005; Johnson & McGunnigall-Smith, 2008; Leigey, 2015), provide a rich source of data, much like the firsthand biographies of those serving sentences of LWOP (George, 2015; Hassine, 2011; Rideau & Wikberg, 1992). Another consideration for empirical studies is a focus on variables that could be responsible for differences in offending among LWOP inmates, particularly institutional-level correlates of misconduct (Schenk & Fremouw, 2012). Included in such contextual variables is an assortment of institutional incentives potentially available to inmates as a means to foster acceptable behavior. The successful development and deployment of incentives like those used in Missouri should be explored more fully as management tools in other prison settings to reduce security threats. These are particularly critical to understanding the motivation of long-term inmates that, at first glance, may appear to have little to lose by disruptive behavior and disciplinary misconduct.

Finally, researchers are encouraged to tackle the larger issue concerning the effectiveness and efficiency of LWOP sentences vis-à-vis LWP sentences (Cheatwood, 1988; Kazemian & Travis, 2015; Leigey, 2010). Under what conditions, if any, are LWOP sentences preferred to LWP and vice versa? LWOP has become a reasonably safe alternative to the death penalty, and the sanction is being applied to take on a much larger share of the population of capital murderers (Olgetree & Sarat, 2012). LWOP is also being used increasingly for habitual offenders and sex offenders (Nellis, 2013). The use of LWOP for each of these subpopulations raises myriad issues and provides fertile ground for future research.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Jonathan R. Sorensen, PhD, is a professor in the Criminal Justice Department, East Carolina University. He also works as a trial consultant and expert witness. His research focuses on prison violence and capital punishment.

Thomas J. Reidy, PhD, ABPP, is a clinical and forensic psychologist in Monterey, California. He provides forensic psychological assessments and consultation for the military, law enforcement, and federal and state courts focusing on risk assessment, violence, and homicide. His professional publications are related to criminal behavior and risk assessment for prison violence, particularly related to capital homicide offenders.

References

Agnew

(1985). A revised strain theory of delinquency. Social Forces, 64, 151-167.

Agnew

(1992). Foundation for a general strain theory of crime and delinquency. Criminology, 30, 47-87.

American Civil Liberties Union. (November, 2013). A living death: Life without parole for nonviolent offenses. New York, NY: ACLU.

Blecker

(2013). The death of punishment: Searching for justice among the worst of the worst. New York, NY: Palgrave Macmillan.

Blevins

K. R.

Listwan

S. J.

Cullen

F. T.

Jonson

C. L.

(2010). A general strain theory of prison violence and misconduct: An integrated model of inmate behavior. Journal of Contemporary Criminal Justice, 26, 148-166. doi:10.1177/1043986209359369

Cheatwood

(1988). The life-without-parole sanction: Its current status and a research agenda. Crime & Delinquency, 34, 43-59. doi:10.1177/0011128788034001003

Crawley

Sparks

(2006). Is there life after imprisonment? How elderly men talk about imprisonment and release. Criminology & Criminal Justice, 6, 63-82. doi:10.1177/1748895806060667

Cunningham

M. D.

Reidy

T. J.

Sorensen

J. R.

(2005). Is death row obsolete? A decade of mainstreaming death-sentenced inmates in Missouri. Behavioral Sciences & the Law, 23, 307-320. doi:10.1002/bsl.608

Cunningham

M. D.

Reidy

T. J.

Sorensen

J. R.

(2008). Assertions of “future dangerousness” at federal capital sentencing: Rates and correlates of subsequent prison misconduct and violence. Law and Human Behavior, 32, 46-63. doi:10.1007/s10979-007-9107-7

10.

Cunningham

M. D.

Reidy

T. J.

Sorensen

J. R.

(2016). Wasted resources and gratuitous suffering: The failure of a security rationale for death row. Psychology, Public Policy, and Law, 22, 185-199. doi:10.1037/law0000072

11.

Cunningham

M. D.

Sorensen

J. R.

(2006). Nothing to lose? A comparative examination of prison misconduct rates among life-without-parole and other long-term high-security inmates. Criminal Justice and Behavior, 33, 638-705. doi:10.1177/0093854806288273

12.

Daggett

D. M.

Camp

S. D.

(2009). Do official misconduct data tell the same story as the individuals who live in prison? Criminal Justice Review, 34, 428-449. doi:10.1177/0734016808329291

13.

Death Penalty Information Center. (n.d.). Life without parole. Retrieved from https://deathpenaltyinfo.org/life-without-parole

14.

Delisi

(2005). Career criminals in society. Thousand Oaks, CA: Sage.

15.

DeLisi

(2016). Career criminals and the antisocial life course. Child Development Perspectives, 10, 53-58. doi:10.1111/cdep.12161

16.

DeLisi

Piquero

(2011). New frontiers in criminal careers research, 2000–2011: A state-of-the-art review. Journal of Criminal Justice, 39, 289-301. doi:10.1016/j.jcrimjus.2011.05.001

17.

Dilulio

(1987). Governing prisons: A comparative study of correctional management. New York, NY: The Free Press.

18.

Drury

A. J.

DeLisi

(2011). Gangkill: An exploratory empirical assessment of gang membership, homicide offending, and prison misconduct. Crime & Delinquency, 57, 130-146. doi:10:1177/0011128708325051

19.

Flanagan

T. J.

(1980). Time served and institutional misconduct: Patterns of involvement in disciplinary infractions among long-term and short-term inmates. Journal of Criminal Justice, 8, 357-367.

20.

Flanagan

T. J.

(Ed.). (1995). Long-term imprisonment: Policy, science, and correctional practice. Thousand Oaks, CA: Sage.

21.

George

(2015). A woman doing life: Notes from a prison for women ( Johnson

Martin

A. B.

, Eds., 2nd ed.). New York, NY: Oxford University Press.

22.

Hassine

(2011). Life without parole: Living and dying in prison today ( Johnson

Tabriz

, Eds., 5th ed.). New York, NY: Oxford University Press.

23.

Henry

J. S.

(2012). Death in prison sentences: Overutilized and underscrutinized. In Ogletree

C. J.

Sarat

(Eds.), Life without parole: America’s new death penalty (pp. 66-95). New York: New York University Press.

24.

Irwin

Cressey

D. R.

(1962). Thieves, convicts and the inmate culture. Social Problems, 10, 142-155.

25.

Johnson

Dobrzanska

(2005). Mature coping among life-sentenced inmates: An exploratory study of adjustment dynamics. Corrections Compendium, 30, 8-38.

26.

Johnson

McGunnigall-Smith

(2008). Life without parole, America’s other death penalty: Notes on life under sentence of death by incarceration. The Prison Journal, 88, 328-346. doi:10.1177/0032885508319256

27.

Kazemian

Travis

(2015). Imperative for inclusion of long termers and lifers in research and policy. Criminology & Public Policy, 14, 355-395. doi:10.1111/1745-9133.12126

28.

Leigey

M. E.

(2010). For the longest time: The adjustment of inmates to a sentence of life without parole. The Prison Journal, 90, 247-268.

29.

Leigey

M. E.

(2015). The forgotten men: Serving a life without parole sentence. New Brunswick, NJ: Rutgers University Press.

30.

Luke

D. A.

Homan

S. M.

(1998). Time and change: Using survival analysis in clinical assessment and treatment evaluation. Psychological Assessment, 10, 360-378.

31.

McInnis

(2003, April 2). Senate panel pushes no-parole sentencing option. Houston Chronicle, p. A-19.

32.

Morris

R. G.

Carriaga

M. L.

Diamond

Piquero

N. L.

Piquero

A. R.

(2012). Does prison strain lead to prison misbehavior? An application of general strain theory to inmate misconduct. Journal of Criminal Justice, 40, 194-201. doi:10.1016/j.jcrimjus.2011.12.001

33.

Morris

R. G.

Longmire

D. R.

Buffington-Vollum

Vollum

(2010). Institutional misconduct and differential parole eligibility among capital inmates. Criminal Justice and Behavior, 37, 417-438. doi:10.1177/0093854810361672

34.

Nellis

(2013, September). Life goes on: The historic rise in life sentences in America. Washington, DC: The Sentencing Project.

35.

Nellis

(2017, May). Still life: America’s increasing use of life and long-term sentences. Washington, DC: The Sentencing Project.

36.

Olgetree

C. J.

Sarat

(2012). Life without parole: America’s new death penalty? New York: New York University Press.

37.

Reidy

T. J.

Cunningham

M. D.

Sorensen

J. R.

(2001). From death to life: Prison behavior of former death row inmates in Indiana. Criminal Justice and Behavior, 28, 62-82.

38.

Rideau

Wikberg

(Eds.). (1992). Life sentences: Rage and survival behind bars. New York, NY: Three Rivers Press.

39.

Schenk

A. M.

Fremouw

W. J.

(2012). Individual characteristics related to prison violence: A critical review of the literature. Aggression and Violent Behavior, 17, 430-442. doi:10.1016/j.avb.2012.05.005

40.

Schmitt

G. R.

Konfrst

H. J.

(2015, February). Life sentences in the federal criminal justice system. Washington, DC: United States Sentencing Commission.

41.

Sorensen

J. R.

Cunningham

M. D.

(2007). Operationalizing risk: The influence of measurement choice on the prevalence and correlates of prison violence among incarcerated murderers. Journal of Criminal Justice, 35, 546-555. doi:10.1016/j.jcrimjus.2007.07.007

42.

Sorensen

J. R.

Pilgrim

R. L.

(2000). An actuarial risk assessment of violence posed by capital murder defendants. The Journal of Criminal Law and Criminology, 90, 1251-1270.

43.

Sorensen

J. R.

Wrinkle

R. D.

(1996). No hope for parole: Disciplinary infractions among death-sentenced and life-without-parole inmates. Criminal Justice and Behavior, 23, 542-552. doi:10.1177/0093854896023004002

44.

Sorensen

J. R.

Wrinkle

R. D.

Gutierrez

(1998). Patterns of rule-violating behaviors and adjustment to incarceration among murderers. The Prison Journal, 78, 222-231.

45.

Steiner

(2009). Assessing static and dynamic influences on inmate violence levels. Crime & Delinquency, 55, 134-161. doi:10.1177/0011128707307218

46.

Steiner

Butler

H. D.

Ellison

J. M.

(2014). Causes and correlates of prison inmate misconduct: A systematic review of the evidence. Journal of Criminal Justice, 42, 462-470. doi:10.1016/j.jcrimjus.2014.08.001

47.

Steiner

Wooldredge

(2008). Inmate versus environmental effects on prison rule violations. Criminal Justice and Behavior, 35, 438-456. doi:10.1177/0093854807312787

48.

Steiner

Wooldredge

(2009). The relevance of inmate race/ethnicity versus population composition for understanding prison rule violations. Punishment & Society, 11, 459-489. doi:10.1177/1462474509341143

49.

Stewart

Lieberman

(1982). What is this new sentence that takes away parole? Student Lawyer, 11, 14-17.

50.

Sykes

G. M.

(1958). The society of captives: A study of a maximum security prison. Princeton, NJ: Princeton University press.

51.

Toch

Adams

(1989). The disturbed violent offender. New Haven, CT: Yale University Press.

52.

Toman

E. L.

Cochran

J. C.

Cochran

J. K.

Bales

W. D.

(2015). The implications of sentence length for inmate adjustment to prison life. Journal of Criminal Justice, 43, 510-521. doi:10.1016/j.jcrimjus.2015.11.002

53.

U.S. v. Savage, E. Dist. PA, No. 07-00550. (2013).

54.

Vinter and Others v. The United Kingdom, Application Nos. 66069/09, 130/10, Eur. Ct. H.R. (July 9, 2013).

55.

Walters

G. D.

(2007). Using Poisson class regression to analyze count data in correctional and forensic psychology: A relatively old solution to a relatively new problem. Criminal Justice and Behavior, 34, 1659-1674. doi:10.1177/0093854807307030

56.

Wardlaw

Biles

(1980). The management of long-term prisoners in Australia. Canberra, Australia: Australian Institute of Criminology.

57.

Williamson

(1985). Rule violations: Will lifers play by the rules? Corrections Today, 47, 138-140.

58.

Wolff

Blitz

C. L.

Shi

Siegel

Bachman

(2007). Physical violence inside prisons: Rates of victimization. Criminal Justice and Behavior, 34, 588-599. doi:10.1177/0093854806296830

59.

Zamble

(1992). Behavior and adaptation in long-term prison inmates: Descriptive longitudinal results. Criminal Justice and Behavior, 19, 409-425.