though they undoubtedly most often operate in terms of personal advantage, can rationalize their offense as contributing to the fiscal health of their employer. These may not be crucial variations, but it would seem preferable to examine offenses such as embezzling, tax evasion, corporate violations, and fee splitting as distinct forms of crime which may be related to each other in some ways and to other offenses in different ways. It would also appear reasonable to concentrate initially on the elements of the criminal act for purposes of grouping it rather than upon the social characteristics of the perpetrators of the acts, and to group behavior in terms of the latter item only for the most compelling pragmatic or interpretative reasons. The crimes of medical doctors, for instance, would appear to be susceptible to differentiation on more meaningful terms than the professional status of their perpetrators. The offenses of fee splitting and abortion, both committed by doctors, seem about as related in most essential respects as the offenses of infanticide and adultery, both of which may be committed by mothers.
Until this analytical impasse is more fully resolved, the concept of white-collar crime may stand indicted on Tappan's charge that it soars "into vacuity, wide and handsome," 56 and on Vold's allegation that it is at the moment "ambiguous, uncertain, and controversial." 57
It may be noted in conclusion that the need for white-collar crime to be studied in terms of more homogeneous units represents a requirement common to the field of criminology. "It should be abundantly clear that theories which treat crime as though it were a unitary concept are particularly prone to failure and that the search for something which explains crime in general is the blind spot in criminology," Gibbs has written, pointing out that "if our concern is with causal homogeneity, as it should be, we can ill afford to deal with broad categories of behavior; it is far better to look within these broad categories and delimit specific types of behavior for investigation."58