The Ontological Background of The Relationalist Turn in Understanding Mental Disorders

Zachar's (2020) commentary can be read as a sophisticated theoretical extension of the target article, and I am very pleased to see support for a relationalist turn in understanding mental disorders. Indeed, underlying causal properties are only an ingredient of essentialism and a more elaborate discussion of this issue brings us readily to the discourse on kinds and ontology. Zachar (2002, 2015, 2020) is an advocate for thinking of psychiatric constructs as practical kinds. This logically leads him to think that the problem with categorical classification systems like the Diagnostic and Statistical Manual of Mental Disorders (DSM) and the International Classification of Diseases are practical problems (lack of training of clinicians in understanding the nature and limits of classifications) which ask for practical solutions (proper training of clinicians).

I am very much in agreement with Zachar's (2020) concerns about a lack of training of clinicians in understanding the nature and limits of the current categorical classification systems, yet I do not think that the "practical program" is able to bring the complete story. If clinicians understand that psychiatric classifications indeed do not represent natural kinds and if the reification effect will thereby wane, we are not there yet. Even if we consider psychiatric constructs as practical kinds, they still try to represent and embody reality and influence reality. Following Hacking (1999) I understand psychiatric classifications to be "interactive kinds." An interactive kind is a type of classification that interacts with the object it classifies (in casu: humans). The interactive kind assumes an interaction between the classification, the classified patient and the material, social, political and scientific environment (see also Köhne, 2019). People may experience themselves in a different way and may develop feelings and behavior in part because they are so classified. This may even lead to changes in the classification itself, which is what Hacking (1999) terms the classificatory looping effect.

A categorical classification system is thus not an innocent practical tool that is merely misunderstood in its reach. A construction upon reality may change reality: it may change patients, change the classification itself by looping effects, directs research efforts, determines our clinical practice (treatments), and may even engender stigma. The key point of critique is that the categorical models commit us to a false ontology. This false ontology fixes our view and hampers scientific progress. My point is that we need a better ontology that helps us to find more adequate scientific explanations and help us to develop better treatments.

Whereas the practical program seems to ignore this ontological complicity, the relationalist program incorporates and explicates this ontological complicity. Thereby it aims to discuss and warn for a priori, top-down, reductionist, brain-based accounts and welcomes nonreductionist, bottom-up accounts. Although I am in agreement with Zachar that there is no such thing as a view from nowhere, this ontological stance commits us to a "least construct-most neutral" position. A relationalist ontology however holds that we cannot separate pure object from pure subject given they constitute each other: the structure of the subjective and objective order is relational.

Relationalism objects to explaining and understanding things in isolation, but means to view things in their ontological interconnectedness. This stance may be best illustrated by an example from (meta)physics concerning the status of space–time. The debate on the ontological status of space–time is historical as well as actual and concerns the question whether we should view space–time as an independently existing entity or as an ontologically related entity of material objects. According to Newton's substantivalism and its modern and adjusted versions, space–time (and its parts) is a fundamental constituent of reality that exists in its own right. Substantivists believe space–time to be an entity that stands on its own and exists independently of material objects. In contrast, relationalism entails to view space–time as a derivative existing entity that is shaped by matter. Within (varieties of) the relationist approach, space–time and material entities (objects) are ontologically dependent constituents of reality. Stating or claiming something about space...