Spinozan Doxasticism about Delusions

In today's post, Federico Bongiorno gives an overview of his paper "Spinozan Doxasticism about Delusions" which is forthcoming in Pacific Philosophical Quarterly. Federico is a postdoctoral researcher at the University of Oxford funded by an award from the Mind Association, working at the interface of philosophy of mind, cognitive science, and the philosophy of psychiatry.

Federico Bongiorno

There are normative standards that are widely held to be required for the practice of belief ascription. At a minimum, beliefs are to appropriately respond to the relevant evidence (epistemic rationality), to cohere with other beliefs (procedural rationality), and to drive consequential behaviour in the right conditions (agential rationality). We adult humans ascribe beliefs to ourselves, and to one another, in reliance of these standards, but belief ascription can be a tricky undertaking. It is especially tricky in cases of delusion, a clinical symptom observed across a variety of psychiatric disorders. One of the biggest issues in connection to delusions is whether they can be beliefs despite breaching all three standards listed above, that is, even if they are often (i) immune to evidence, (ii) inferentially encapsulated, and (iii) behaviourally inert.

There have been two major types of response to this issue. The most popular is what I call ‘standard doxasticism’ (Bortolotti, 2009, 2012), according to which if we deny belief status to delusions on the grounds of their being i, ii, iii, then we get the implausible result that we have very few beliefs, since many of the beliefs we routinely ascribe to each other are i, ii, iii. A second type of response is to insist that epistemic, procedural, and agential rationality are necessary for something’s being ascribed as belief, and that, since delusions tend to be i, ii, and iii, they are not beliefs, but, at best, non-doxastic, or semi-doxastic, attitudes in their vicinity.

What has been largely overlooked by both sides is the fact that ‘belief’ can refer either to a commonsense psychological category on which we rely when predicting behaviour, or to a mental type whose causal role is articulated by law-like generalisations uncovered by cognitive science. As far as I can tell, there has been scant attention paid to whether delusions are beliefs in this latter sense, and that reflects an important gap in the literature. Progress on this question can be made by asking what our belief-mechanisms are designed to do, so that we can then see how delusions stand with respect to their design specifications. If we are able to make cognitive scientific generalisations about the ways we normally fix and update beliefs, this may shed new light on the nature of delusions, for we can then ask whether or not these generalisations extend to delusions.

The sort of generalisation whose implications I explore in my paper is a theory of belief known as Spinozan Theory (see e.g., Gilbert, 1991; Mandelbaum, 2010, 2014; Asp et al., preprint). The main idea is that we acquire any proposition we entertain initially as a belief, and only after the initial acceptance, if enough cognitive resources are available, can the belief be dislodged. So, for example, even given an outlandish proposition like ‘clouds are made of cotton candy’, believing is default with the semantic comprehension of that proposition, and it takes an extra process of evaluation to either endorse it, or more importantly, disbelieve it.

What then is distinctive of the Spinozan Theory is that believing and disbelieving are performed by different cognitive processes, which also means that they are differentially affected by performance constraints. Because believing is reflexive, it is undemanding of cognitive resources, and so remains unaffected when cognitive resources are depleted by cognitive load. By contrast, evaluating (and possibly rejecting) a proposition is an effortful process which draws heavily on cognitive resources, and so can be disrupted by cognitive load.

Of course, the empirical evidence on the Spinozan Theory remains the subject of ongoing debate and investigation. But the question for me is supposing it was right, what would follow about whether delusions are beliefs? I argue that the Spinozan Theory supports a new version of doxasticism, what I call ‘Spinozan Doxasticism’, which has two advantages over the standard defence: it puts pressure on the very notion that one can be deluded that p without believing p, and it can accommodate i, ii, iii (the features of delusions that many see as most indicative of their not being beliefs) within an empirically viable theory of belief. In closing, I consider whether delusions can fit into the Spinozan picture of belief, and why this is something all doxasticists should want to accept.