DSM disorders and their criteria: how should they inter-relate?

While the changes in psychiatric diagnosis introduced by Diagnostic and Statistical Manual third edition (DSM-III) have had major benefits to the field of psychiatry, the reification of its diagnostic criteria and the widespread adoption of diagnostic literalism have been problematic. I argue that, at root, these developments can be best understood by contrasting two approaches to the relationship between DSM disorders and their criteria. In a constitutive relationship, criteria definitively define the disorder. Having a disorder is nothing more than meeting the criteria. In an indexical relationship, the criteria are fallible indices of a disorder understood as a hypothetical, tentative diagnostic construct. I trace the origins of the constitutive model to the philosophical theory of operationalism. I then examine a range of historical and empirical results that favor the indexical over the constitutive position including (i) evidence that individual criteria for DSM-III were selected from a broader pool of possible symptoms/signs, (ii) revisions of DSM have implicitly assumed an indexical criteria-disorder relationship, (iii) the indexical position allows DSM criteria to be wrong and misdiagnose patients while such a result is incoherent for a constitutive model, an implausible position, (iv) we assume an indexical criteria-scale relationships for many personality and symptom measures commonly used in psychiatric practice and research, and (v) empirical studies suggesting similar performance for DSM and non-DSM symptoms for major depression. I then review four reasons for the rise of the constitutive position: (i) the 'official' nature of the DSM criteria, (ii) the strong investment psychiatry has had in the DSM manual and its widespread use and success, iii) lack of a clear pathophysiology for our disorders, and (iv) the absence of informative diagnostic signs of minimal clinical importance. I conclude that the constitutive position is premature and reflects a conceptual error. It assumes a definitiveness and a literalism about the nature of our criteria that is far beyond our current knowledge. The indexical position with its tentativeness and modesty accurately reflects the current state of our field.