Relationism and Empiricist Accounts of Delusion

Ema Sullivan-Bissett

Currently, I am a PhD student at the University of York, and will soon be a Research Fellow at the University of Birmingham, working on Lisa Bortolotti’s Epistemic Innocence project. Paul Noordhof is Professor of Philosophy at the University of York. We are currently thinking about relationist accounts of perceptual experience, and what proponents of such accounts might have to say about delusional belief formation.

Paul Noordhof

Relationists about perceptual experience hold that perception is a relation of brute non-representational awareness of items in the world. This account can be contrasted with representationalist accounts which hold that perceivers represent the world to be a certain way. On the relationist account, the phenomenal character of perception—what it feels like to the subject to have a perceptual experience—is constituted by the items of which the subject is aware. In the case of hallucination, where intuitively, such objects and properties are missing, relationists are inclined to say two things. The first is that a hallucination of x is merely subjectively indiscriminable from a perception of x. More crucially though, for the relationist, hallucinations do not have any phenomenal character. The subjective indiscriminability of hallucinatory experience is just understood in terms of a common response—for example, in the judgements or beliefs of the subject—to something which does have phenomenal character on the one hand, perception, and something which does not on the other, hallucination.


We distinguish between what we call positive delusions and negative delusions, where the anomalous experiences involved in the formation of the former are hallucinatory, that is, objects and properties are perceived as in the environment and those objects and properties are missing. Examples include perceptual delusional bichephaly (the delusion that one has two heads), and other visual or auditory hallucinations which are interpreted in a delusional way. Negative cases, characterized by a lack of affect, are more numerous; examples include delusions of alien control which might arise in a subject who does not ‘experience volitional control of her own bodily movements’ (Gerrans 2002: 47), the Cotard delusion, characterized by phenomenological disembodiment, and the Capgras delusion, characterized by phenomenological estrangement or derealization (Gerrans 2002: 49).

We are attracted to empiricist accounts of delusion. An empiricist account is one which ‘grounds the delusion in an abnormal experience of some kind’ (Bayne and Pacherie 2004: 81). More specifically, we defend Brendan Maher’s account of delusion according to which the development of delusional beliefs does not differ significantly from the development of ordinary beliefs. It is not necessary to appeal to a deficit in belief-fixation processes, we need only appeal to the anomalous experiences of the subject. Such experiences are hypothesized to have the ‘irreducible primary quality that sensory experiences generally have’ (Maher 2003: 18). They are also ‘intense and prolonged’ (Maher 1999: 566), and are ‘repeated or continue over an extended period’, as well as being ‘vivid and intense enough to preoccupy the consciousness of the individual while [they are] happening’ (Maher 2006: 182).

The relationist need say nothing about negative cases of delusion as such cases do not involve experiences which present to the subject objects and properties in the environment which are not actually present. However, if we run a case of a positive delusion—perceptual delusional bichephaly—through Maher’s account, we get something like the following story: the patient visually hallucinates a second head, this causes him distress and anxiety, he searches out an explanation, and comes to form the delusional belief that he has a second head.

Because the relationist denies that hallucinatory experiences have phenomenal character, empiricist approaches which give anomalous experiences a role in the explanation of delusional belief formation, are not available to her. The relationist cannot adopt any approach which grounds the formation of delusional beliefs in the strange phenomenology of anomalous experiences.

The relationist then can accept empiricist accounts of the formation of what we identified as negative delusions, but they must offer a different story for cases of positive delusion. In denying phenomenal character to hallucinatory experience, the relationist in turn denies a unity of explanation which might otherwise be available if hallucinatory experiences were taken to have phenomenal character.