Philosophy, Psychiatry, & Psychology

As Peter Zachar rightly points out in his comment, the assessment of mental disorders underwent new developments with the release of the Diagnostic and Statistical Manual of Mental Disorders-V in 2013 (American Psychiatric Association, 2013). Whereas in 1980, the manual had been thought of in a rigorously categorical way, on the basis of distinct and closed syndromic entities, this new version advocated the development of a radically dimensionalist approach, with the diagnosis aiming this time to restore the whole symptomatological picture of the patient on a spectrum of ratings are supposed to cover the entirety of psychiatric semiology (American Psychiatric Association, 2013).

For its part, my article aimed to show how standardized assessment, guided by the return of psychiatry’s authority and the assertion of its legitimacy in the diagnostic field, gave the ascendancy in the 1980s to the classificatory tradition and thus to the categorical model, to the detriment of personality approach and test psychology which offered a competing solution here. This perspective, historically dated, may have led the reader to think that there is an antagonism that cannot be overcome here. Rather, it should be seen as a form of structuring opposition, or even a “complementarity inscribed in rivalry,” between psychiatry and psychology, which is naturally susceptible to change over time.

Indeed, the reversal operated by the Diagnostic and Statistical Manual of Mental Disorders-V is not total because it cannot be total. The defense of the dimensionalist approach, dear to the psychological testing tradition, was intended here to allow the identification of co-morbidities, present in most patients but poorly apprehended by the categorical approach, and to open up diagnosis to prevention by screening for minor manifestations of psychopathology when they suggest the [End Page 259] development of a manifest episode of the disease in the more or less long term. The risk here is that we end up with overdiagnosis and thus an over-representation of false positives (Wakefield, 2015). In other words, this type of approach does not escape the obligation to determine threshold scores, likely to indicate the lower limit of the disorder, and thus to resort to a categorical approach. There is a first affinity between the two perspectives which obeys, beyond the divergences in principle, a technical requirement.

But the two approaches have even more in common when we look at their epistemological background: they share the idea that the representation and the intervention on mental disorder, to use Ian Hacking’s expression (1983), must be based on the model proposed by the physical and natural sciences. And as Greta Kaluzeviciute aptly points out in her comment, this orientation leads to the idea of measuring mental disorder in particular to conduct elaborate quantification operations to determine, for example, whether two entities are statistically distinct or whether a drug has a significant effect on the evolution of a pathology.

This is in fact an old question that can be formulated as follows, to use the expression of the psychometrician Joel Michell (1999): are psychological dimensions metric? In other words, if we want to apply the language of numbers to mental disorders, they must share its properties, two of which seem essential: it must be possible to establish that each mental disorder, in the same way as qualities attributable to length for example, has a zero and can be divided by means of gradients (or degrees of intensity) whose interval is constant (Le Moigne, 2018; Michell, 1999). For lack of space, let us only address the second condition. This one has already been denied by many, starting with Bergson at the end of the nineteenth century (1889). He does not believe in the quantification of psychological expressions, starting with sensations. What is the common point, he wrote, between the feeling of a mild pain, as one feels a stiffness, and the sensation of a needle being hammered into your flesh? For him, one and the other of these intensities are incomparable, so that they do not belong to a similar register of sensations that one could stage from a 0 toward a...