It is widely acknowledged that the approach taken in the development of a classification of mental disorders is guided by various values and assumptions. The author, who played a central role in the development of DSM-III (American Psychiatric Association  Diagnostic and statistical manual of mental disorders, 3rd ed. Washington, DC:Author) and DSM-III-R (American Psychiatric Association  Diagnostic and statistical manual of mental disorders, 3rd ed, rev. Washington, DC:Author) will explicate the basic values and assumptions that guided the development of these two diagnostic manuals. In so doing, the author will respond to the critique of DSM-III and DSM-III-R made by Sadler et al. in their 1994 paper (Sadler JZ, Hulgus YF, Agich GJ  On values in recent American psychiatric classification. J Med Phil 19:261-277
). The author will attempt to demonstrate that the stated goals of DSM-III and DSM-III-R are not inherently in conflict and are easily explicated by appealing to widely held values and assumptions, most of which appeared in the literature during the development of the manuals. Furthermore, we will demonstrate that it is not true that DSM-III places greater emphasis on reliability over validity and is covertly committed to a biological approach to explaining psychiatric disturbance.