UNTIL recent years, the personality disorders have been relatively une
xplored compared to other psychiatric diagnoses. Over 15 years ago, th
ere was little agreement on the diagnosis of borderline personality di
sorder (Ferry and Klerman 1978), but efforts to specify the constructs
and respective criteria for the borderline diagnosis spurred a pletho
ra of systematic research, The result is that, next to antisocial pers
onality disorder, borderline has become one of the best-documented and
validated personality disorders (Ferry and Vaillant 1989), One import
ant shift has been that good descriptive studies have gradually led to
studies of etiological factors, such as childhood physical and sexual
abuse, and severe neglect (Herman et al, 1989; Ferry and Herman 1992)
, which in turn have led to empirically based treatment approaches (He
rman 1992; Ferry et al. 1990), Despite inclusion in The Diagnostic and
Statistical Manual of Mental Disorders (DSM-III and DSM-III-R), narci
ssistic personality is still at the beginning of this process of descr
iption, empirical testing, and validation (Gunderson et al. 1991). Thi
s study empirically examines three descriptions of narcissistic person
ality in order to look for common underlying dimensions that may have
etiological and treatment significance.