Race, Labor, & Punishment in the New South by Martha A. Myers. Columbus: Ohio State University Press. 1998. 326 pp. Cloth $55.00. Paper $24.95. ISBN 0-8142-5001-7.

Reviewed by B. Dan Wood, Department of Political Science, Texas A&M University. Email: bdanwood@polisci.tamu.edu.

This is a quantitative historical study of incarceration patterns in the state of Georgia between 1870 and 1940. Focusing on disparities in incarceration patterns as a function of race, the author attempts to identify the social and economic forces that shaped punishment for both blacks and whites over this seventy year period. In particular, the author analyzes admission rates to the penitentiary, length of sentences, rates of release, and the evolution of lessee/hard labor patterns.

A core theme is that the state of Georgia and its system of punishment changed between 1870 and 1940. In the nineteenth century black men were more frequently caught within the system of formal punishment, and found release to be more difficult. Black men were also more likely to be "leased-out" for their labor, a sort of perpetuation of slavery in the New South. By the twentieth century, prisoner "lease-out" programs had been abolished to be replaced by county and state run labor programs. Still, it was black men who disproportionately comprised the convict population. And the rate at which black men were admitted to prison increased dramatically through this period, relative to that for white men.

The author also examines the effect of economic factors and war on incarceration patterns. She finds that incarceration and release rates for black men were more responsive to economic conditions. However, the direction of relationships was less predictable and systematic. The incarceration rate for black men was more responsive to the cotton economy. A severe depression in the 19^th century led to higher incarceration rates for both black and white men, but a shorter post WWI recession led to lower incarceration rates. Generally, economic crises had less impact on black men than it did on white men. The sentences imposed on black offenders were strongly dependent on the economy, including its general health, depressions and recessions, the price of cotton, and size of harvest. Economic crises had more uniform effects on white offenders, discouraging their release from prison. In contrast, economic crises often discouraged the admission of black offenders and encouraged their release.

The author tests a number of other hypothetical relationships including the effects for war, population growth, urbanization, and the New Deal on incarceration, punishment, and release. The study contains numerous statistical analyses, all of which cannot be discussed in this summary.

Methodology: The author used Box-Jenkins time series analysis of historical data from published reports of administrators responsible for managing Georgia's penitentiaries. Additionally, data were drawn from the Central Register of Convicts at the Georgia Department of Corrections.

At times the author relied on the cross-correlation function, a bivariate tool, to draw conclusions, as well as to guide specification of Box-Jenkins models. Several of the models were intervention type models, while others were multivariate models complete with pre-whitening and transfer functions. The analysis appears to have a good grasp of the requirements of Box-Jenkins analysis, and was particularly sensitive to issues of stationarity in both level and variance. The analysis seemed competently done.

One area of possible methodological improvement relates to the comparison of black and white incarceration variables. In most analyses, the author conducts separate estimations for blacks and whites, rather than a single analysis of covariance. The use of separate estimations precludes the ability of the study to conduct formal hypothesis tests of differences between blacks and whites. The author nevertheless drew strong conclusions that the two groups were different on a number of attributes. Given the centrality of these comparisons to the study, this seems to me a critical oversight. Still, no study is perfect, and I'd have to rate this book as among the top quantitative studies in this area.

Other Comments: This is credible scholarly work that should interest those doing work on social justice, criminal justice, or social welfare policy. The book might also be appropriate for courses in the politics of race or civil rights. The book is well organized and clearly written.

However, if readers are seeking answers to more general theoretical questions, the book misses the mark. The empirical analyses are poorly connected with overarching theories of sociology, law, or political science. It lacks the "big picture" perspective we would expect of theoretically interesting work. Indeed, the book often fits more into the category of raw empiricism. In my view, there should have been more discussion of the implications for historical treatment or developing social science theory.

Additionally, the book would be a complicated and difficult read for those not familiar with Box-Jenkins time series analysis. The first three and last chapters provide a nice summary of the findings, but the middle chapters are somewhat technical and at times cryptic. This would probably not be an appropriate book to assign for most undergraduate courses, and would perhaps be a bit difficult for graduate classes unless students had specialized training.

Copyright 1995