Towards a Sociology of Mercy: A Mixed-Methods Analysis of Commutation Release in the United States

Abstract

Using a multimethod approach, this dissertation explores a legal mechanism of mercy in the United States: commutation of sentence. Commutation is a form of executive clemency that allows the president (at the federal level) or the governor (at the state level) to shorten a prisoner's sentence. While this release mechanism exists throughout the country, it has been neglected in empirical scholarship, particularly in state systems where state governors typically have wide discretion, with power to alter sentencing decisions made by prosecutors, judges, statutes, and sentencing guidelines. The unprecedented size and cost of imprisonment has led to increased bipartisan support for reducing the prison population. Commutation, if utilized widely, has the potential to reduce incarceration rates. This study is the first mixed-methods investigation of state-level commutation releases in the United States. The primary aim underlying this project is to identify how commutation release could be expanded. With so little known about the current use of commutation, I investigate four specific research questions: 1) who is being released through commutation? 2) in what contexts do these releases occur? 3) how does commutation occur? 4) why are individual commutation cases successful? I address questions 1 and 2 through quantitative methods, including statistical techniques such as logistic regression, mixed-effects, and multinomial models on individual, state, and national prison data. I find that this type of mercy is rarely used, with whites, women, and those with greater educational attainment favored in commutation releases. This holds true both when comparing commutation releasees to the entire prison population, and when comparing the determinants of release through commutation to release through other mechanisms such as discretionary and mandatory parole. Regarding the macro-level contexts in which commutation releases occur, I find Republican governorship and higher rates of violent crime predict more commutation releases. Cross-level interactions reveal that while for white women, governor political affiliation has no impact on the probability of commutation release, yet black men, black women, and white men all benefit under Republican governorship. I address questions 3 and 4 qualitatively, through an analysis of commutation policies, individual commutation applications, transcripts from past commutation hearings, and hearing observations in four sites (Iowa, Louisiana, Pennsylvania, and Washington). I find that despite wide variation in the bureaucracy of commutation, each of these sites has structural features that severely curtail the large-scale use of commutation as a mechanism of prison release. For individual applicants, I conclude that while under present practices there seems to be nothing a person can do to ensure success in their request for mercy, many conditions must be met for commutation release to be a reasonable possibility. I term these Necessary But Insufficient Conditions (NBICs), grouping these conditions into four categories: changeable, subjectively determined, external, and unchangeable. Theoretically, I argue that despite its infrequency, commutation serves an important and overlooked function in modern penal practice. Commutation plays a key role in the reification of state power and maintenance of social control, both within prisons and in society more broadly.