监狱三分工作

世界各国刑罚无不以自由刑为主，我国也不例外。刑法规定的五种主刑，四种为自由刑，还有一种虽为死刑，实际多为监禁执行的死缓刑。与财产刑、资格刑、生命刑等不同，以剥夺或限制人身自由为内容的自由刑的执行，由于具有可分割性和时空的连续性，因此不仅能够满足罪刑相适应需要，更好实现量刑公正，而且可以将罪犯的“刑期”变“学期”，追求改造罪犯为守法公民的行刑目标。然而，自由刑的惩罚本质与行刑改造目标之间由于存在固有矛盾，如何缓解这一矛盾也就成为自由刑执行必须破解的难题。自由刑执行权由惩罚、监管和改造等不同权能构成，且具有刑事司法、特别行政、职务权利等不同法律属性。因此，突破行刑悖论，自由刑执行需要根据不同的权能，遵循不同的运行程序、规范要求和监督路径。

自由刑执行坚持惩罚和改造相结合，以改造人为宗旨。然而，以平均分配有限行刑资源为特征的传统自由刑执行模式，不仅加剧了剥夺或限制自由的惩罚与促进罪犯融入社会的改造之间的固有矛盾，而且引起了改造手段与目的之间的背离，带来了行刑不公与效率不高的消极影响。随着国家法治进程加快，特别是刑事司法人工智能的不断发展，统一刑罚执行改革必须借助人工智能技术推进自由刑执行的一体化。为此，针对传统行刑存在的供给与需求不匹配难题，要依托再犯风险评估智能化的发展，通过准确评估、即时预警、精准推荐、循证监管和改造实现犯因与应因的动态匹配，并以此实现自由刑执行标准统一、主体协同、关系协调、程序规范、前后衔接等一体化目标。

Abstract In December 2018, the First Step Act of 2018 was signed into law, which required the Federal Bureau of Prisons (BOP) to begin using risk and needs assessment to assign programs and provide inmates with required treatment. The Attorney General was tasked to develop an assessment to assess and assign inmates to four possible risk levels. In response, BOP and National Institute of Justice (NIJ) researchers created the Prisoner Assessment Tool Targeting Estimated Risk and Needs (PATTERN). Developed and validated on a large sample (N = 222,970) of inmates released between 2009 and 2015, the PATTERN is a gender-specific instrument and uses static and dynamic items to assess general and violent recidivism. Using a boosted regression procedure, the PATTERN achieved a high level of predictive validity. Tests for bias revealed the PATTERN further reduced racial/ethnic disproportionality.

ABSTRACT The Pareto principle is based on the concept that roughly 80% of outcomes are generated by 20% of inputs, efforts, or contributors within a group. Using a national sample of U.S. prison inmates, we examined various percentile rankings of self‐reported institutional misconduct to determine how much disorder is created behind bars by the most prolific offenders. Findings revealed that, regardless of sex, the top 20% of inmates were responsible for approximately 90% of all rule violations and write‐ups received. These general patterns remained similar even after adjusting infractions for time served in prison. Further analyses indicated that membership within these high‐rate groups was often significantly predicted by those who were younger, black, had more extensive criminal histories, committed violent crimes, resided in state facilities, anticipated being released, used drugs prior to their arrest, were diagnosed with a personality disorder or ADHD, and exhibited worse negative affect. Some sex‐specific effects were also observed. The disproportionate impact these chronic offenders have on the prison environment is detrimental to all individuals who live and work around them. Future research should investigate specific types of misconduct, distinct time intervals of incarceration, and facility effects such as management style, security levels, or offender composition.

This study develops an ideal model of penitentiary law for Indonesia's correctional system through a comparative analysis with other legal frameworks. The study examines systems in South Africa, Thailand, the Netherlands, Germany, France, Japan, and Brazil using doctrinal normative legal research and comparative legal analysis. Primary sources include legal documents, statutes, relevant legislation, and scholarly literature. The findings reveal key components of an ideal penitentiary system: a rehabilitative legal framework that aligns punishment with rehabilitation; provisions for individualised assessment and treatment; a disciplined, transparent system for grievance handling; legislative support for educational and vocational programs; and policies for community reintegration. Notable contributions from other countries include Japan's rehabilitation system, the Netherlands’ alternative sentencing approach, Germany’s vocational training programs, France’s parole systems, South Africa’s restorative justice initiatives, and Thailand’s family connection efforts. Brazil's flexible sentencing and rehabilitation strategies to address overcrowding further inform the model. This research proposes a comprehensive penitentiary law framework emphasising legal pluralism, situational flexibility, robust regulatory mandates, and a balance between security and humane treatment. The model is designed not only for Indonesia but also offers a valuable reference for other nations, grounded in empirical research on corrections and recidivism, and adhering to international human rights standards for prisoner treatment.

… In the latter part of the twentieth century, classification and treatment assessment tools were … in community corrections. The classification systems now used in both correctional facilities …

… In this chapter we again discuss assessments, but here we focus on assessment and … clients within a given correctional unit. To understand the need for such assessments, we must first …

… The two decided to build a risk assessment score for the corrections industry. Brennan wanted to improve on a leading risk assessment score, the LSI, or Level of Service Inventory, …