Empirically Based Sentencing Guidelines and Ethical Considerations (From Research on Sentencing - The Search for Reform, P 184-193, 1983, Alfred Blumstein et al, ed. - See NCJ-91771)

• NCJ Number: 91775
• Author(s): F M Fisher; J B Kadane
• Date Published: 1983
• Length: 10 pages
• Sale Source: National Institute of Justice/; National Academies Press; NCJRS Photocopy Services
• Language: English
• Country: United States of America

Annotation

Empirically based sentencing guidelines are a species of computer-driven conservatism which do not avoid hard ethical decisions and mislead criminal justice practitioners by substituting statistical sophistication for ethical sophistication.

Abstract

The rationale for sentencing guidelines assumes that judges were correct in the past, but that the judges themselves and society would wish to reduce the extent of individual variation around those averages. It also reasonable to require judges explicitly to justify any large departures from the systematic collective wisdom, but the point where this justification is required is an ethical choice as well as dependent in part on past data. If variables are erroneously omitted from equations underlying the guidelines, the estimated effects will be wrong and the guidelines misleading. This will be particularly important if the omitted variables are ethically irrelevant, such as race. However, correct specifications are very difficult to achieve. Models illustrate approaches to sentencing formulas where race has influenced past decisions. These equations require ethical decisions, such as should blacks be treated as courts used to treat whites or should an average be used. These ethical choices challenge the view that guidelines can or should be based on past behavior rather than constructed directly from ethical or societal considerations. Formula and five footnotes are included.