Tuesday, 18 March 2014

Delusions as Acceptances

Richard Dub
My name is Richard Dub. I'm currently a postdoctoral fellow at the Swiss Centre for the Affective Sciences, and I recently received my Ph.D. in Philosophy from Rutgers University. In my dissertation, I offered a model of delusions that attempted to answer two questions: What are delusions? How are they formed? Delusions, I argue, are pathological acceptances formed on the basis of pathological cognitive feelings.

Neither 'acceptance' nor 'cognitive feeling' is an entirely mainstream concept.  A concern that motivates a lot of my work is that it is procrustean to try to explain all mental phenomena in terms of a select few propositional attitudes.  There is little reason to insist that belief and desire must take their traditional place of prominence.  The mind is lush, not sparse.  The ordinary concept of belief is likely what Ned Block calls a "mongrel concept": a concept that imperfectly picks out various dissimilar cognitive states.

Psychopathologies present us with cases that ordinary folk psychology, with lumpen terms like 'belief', is not well-equipped to describe.  They often reveal that mental states and processes which usually coincide can be dissociated from one another.  We might need to be revisionary with our ontology of mental states in order to capture these dissociations.

Delusions are a case in point.  Others on this blog have discussed how, because delusions act like such strange beliefs, they might well not be beliefs at all.  This is non-doxasticism.  But then, what would delusions be, if not beliefs?  Luckily, we don't need to introduce an ad hoc category.  An alternative notion---"acceptance"---can on occasion be spotted lurking throughout philosophy, used to explain phenomena as diverse as voluntary change of mind, self-deception, akrasia, pedagogy, and our understanding of myth and legend.  Frankish and Velleman have both suggested that acceptances can help explain delusions.  Delusions involve acceptance without belief.

It is easiest to get an grip on the type of acceptance at play in delusions (an imperfect grip, mind you) by considering acceptance as supposition without introspective insight.  Over the course of a dinner, we can suppose something for the sake of argument, and then have a discussion as if we believed it.  We usually have good introspective insight about when we are merely supposing something rather than believing it.  However, what would it look like if we lost this insight?  I claim that such a state would look an awful lot like a delusion (though there are differences---delusions are not usually formed voluntarily, for example).  In fact, it turns out that many more mundane "beliefs", such as personal, ideological, or religious convictions, also share these features.  Lisa Bortolotti has argued in her excellent book on delusions that non-doxasticism is threatened by the fact that delusions do not look very different from more mundane irrational beliefs.  But it is not threatened if these other irrational "beliefs" turn out not to be beliefs either.

In a future post, I'll discuss how these pathological acceptances are formed.  You can read more about my research interests on my webpage, and feel free to e-mail me with questions if interested.  Also be sure to take a look at my iconographic representations of various delusion types!


  1. Hi Richard,

    Thanks for this really interesting post (and apologies in taking so long to get to it).

    As I think you know from our chat at the workshop, I prefer to take a doxastic view towards delusions.I think that when we compare the features of acceptance with the features of belief, delusions look to more easily fit with the doxastic view.

    Michael Bratman (1992) gives us four differences between beliefs and what he calls 'acceptance in a context' (which I don’t think is radically different from what you mean by 'acceptance' here):

    1. Belief is context-independent, acceptance is not.

    2. One cannot be consciously influenced by non-epistemic considerations in belief, one can be influenced consciously by non-epistemic considerations in acceptance.

    3. Belief is not something which we can control, acceptance is.

    4. Belief is ‘subject to an ideal of agglomeration’ across contexts, acceptance is not.

    I suspect that if we thought about these four features in the context of a delusion, the verdict would be nearer to the claim of the doxasticist, rather than the claim of your preferred view. So how do you understand delusions with respect to these four? Are you happy to say that delusions really are like acceptances with respect to these four features (so, context-dependent, can be influenced consciously by non-epistemic considerations, can be controlled, and are not subject to an ideal of agglomeration)?

  2. Very interesting. I'm not sure I agree with the counter-claim though, Ema.

    In terms of clinical presentation:

    1. Belief is context-independent, acceptance is not.
    - Delusions are often evoked and afforded by some contexts more than others. People whose delusions are causing them (or others) distress may express those beliefs across many contexts - but it is still common to find a degree of modulation.

    2. One cannot be consciously influenced by non-epistemic considerations in belief, one can be influenced consciously by non-epistemic considerations in acceptance.

    I think that makes a very strong claim about what a 'belief' is, and I'm not persuaded that it is psychologically-supportable.

    3. Belief is not something which we can control, acceptance is.

    - Seems a rather arbitrary, rhetorical move, but if we accept it then, as with [1] above, delusions fall into the acceptance category. There are times when people may be able to control them 'more' or 'less,' and one would associated feeling less able to control such thoughts and feelings with increased distress. But I think it's a spectrum, rather than two categories. However, I don't think it makes sense to say that we can't 'control' belief in the first place.

    4. Belief is ‘subject to an ideal of agglomeration’ across contexts, acceptance is not

    Very few beliefs, and very few delusions are like this, if by 'agglomeration' we mean a coherent and consistent set of thoughts, feelings and actions across contexts.

  3. Hi, Ema,

    I only just discovered that you responded to my post. Sorry to take so long to get back to you!

    You're right that Bratman's concept of acceptance is not radically different than mine. Still, it's narrower in important respects. Bratman was interested in using acceptances to explain planning, volitional reasoning, and joint action. He focussed on a type of acceptance that is relevant in explaining those. I agree that it's unlikely that delusions are acceptances in this sense. But I think that type is only one species of a broader genus. There's a broader class out there: one including acceptances that are context-independent, that are not under volitional control, and that cannot be consciously influenced.

    I guess I should mention that I am OK with what I take to be the most natural reading of the claim "unlike beliefs, acceptances can be context-independent, can be consciously influenced, and can be controlled." I read this as a claim about the sorts of possible acceptances that can exist, not as a claim about the essential modal properties of every token acceptance. (There's an ambiguity in the scope of the modal operators, and I accept one of the disambiguations. Compare: "Unlike fried liver, fried kidneys can be delicious." This shouldn't be read to imply that every token fried kidney has the power to become delicious.)

    I'm inclined to say that Bratman's feature #4 is true of the sorts of acceptances that delusions are, and that's because I'm skeptical that acceptances are subject to any norms other than instrumental or prudential norms. (This is debated.) I'm not sure that this necessarily speaks in favor of the doxasticist about delusions. Delusional individuals are clearly violating all sorts of norms, but it's not obvious that one of the norms they are violating is an epistemic ideal of agglomeration.


Comments are moderated.