Book Review: Bell, S. Crime and Circumstance: Investigating the History of Forensic Science. Westport, CT: Praeger Publishers, 2008. 241 pp. ISBN 978-0-313-35386-4

• Published date: 01 March 2010
• DOI: 10.1177/0734016809349175
• Date: 01 March 2010
• Subject Matter: Articles

116
Criminal Justice Review 35(1)
Many of the efforts in these chapters are impressive and convincing, but the scale of the problem
is still daunting. In principle, one would need to consider all these features and adjustments jointly,
and there are yet many unknowns. For instance, Planty proposes plausible adjustments for under-
counting of repeat victimizations in the calculation of NCVS crime rates. He notes, though, that
because the UCR may well undercount series victimization too, one is not quite sure what to make
of changes in the apparent divergence between the two data sets’ trends when the NCVS is adjusted
for series victimization. Also, along with design issues, the psychology of recall by survey respon-
dents and organizational factors that influence the police classification and reporting of crimes need
to be considered. (These areas are noted but are not the main concern of these chapters.) Work is
therefore far from done, and of course some of the analyses here seem more satisfying than others,
but these articles definitely will be important building blocks for future literature on this topic.
The editors’ main goal is to understand the two systems’ complementarity and how each should
be used, not to judge one as correct (or more correct) under some external criterion. A focus on com-
plementarity is appealing because it recognizes the inherent complexity of the phenomenon of crime
and ambiguity in its definition, and directs attention to the substantive utility of each system rather
than the pursuit of convergence. McDowall and Loftin argue that because actual convergence of
crime trends under the two systems is probably possible in only a rather weak or technical sense,
the most reasonable conclusion is simply to use each data set for the purposes for which it is best
suited. Certainly this conclusion seems sensible to researchers, as criminologists will naturally seek
out the particular data that can most convincingly address their research questions. However it is
important to recognize that even if researchers accept this view, policymakers and the public will
likely continue to notice and be concerned with differences in the two systems’ measurements of
the overall level of crime. And, of course,...