Pii: S0005-7967(98)00144-2

Nelson-Gray and Farmer argue that behavioral assessment and functional analysis may be bene®cially applied to personality disorders (PDs). While this is a reasonable response to the largely nonbehaviorally derived Diagnostic and Statistical Manual (DSM), it is not yet clear that grafting such theoretically incongruent elements will be viable. In essence, they argue that a syndromal classi®cation system could serve a nomothetic role of guiding a functional, idiographic analysis. This is possible, but it seems unlikely that this process would remain in equilibrium, with no interactive e€ect of the functional analysis on the syndromes themselves. Yet the DSM system has shown itself to be surprisingly closed to a more functional approach, so the relationship between the DSM and functional analysis is not open in both directions. What is needed is a nomothetic level of analysis that is also functionally derived. The primary bene®t of functional over syndromal categories is one of treatment utility, a concept that is itself surprisingly absent from the authors' otherwise comprehensive discussion of behavioral assessment. # 1999 Elsevier Science Ltd. All rights reserved.