Tuesday, 4 March 2014

Thought Insertion and the Minimal Self

Caitrin Donovan
I am a student in the Unit for the History and Philosophy of Science at the University of Sydney. In my recently completed honours dissertation, I argued that the delusional phenomenon of thought insertion problematises certain aspects of the 'minimal model' of self.

Many philosophers now believe that the self is in some way constructed by narrative; through socio-linguistically mediated story-telling, we achieve diachronic unity by taking a reflective stance on our experiences. According to the strong formulation of this thesis, conscious beings only develop selves once they acquire the higher-order linguistic and reflective capacities required for autobiographical self-understanding. 

Against this claim, Shaun Gallagher and Dan Zahavi argue that our capacity for narrative construction is prefigured and underpinned by a more basic form of self-awareness. Our experiences do not initially manifest as a sequence of anonymous events, only for us to subsequently reconcile them as our own via autobiographical thematisation. While autobiography may scaffold experiences and connect them over time, our stream of consciousness possesses an in-built, pre-linguistic and pre-reflective unity that the authors call the ‘minimal self’.

Gallagher and Zahavi’s formulation of the minimal model is premised upon a phenomenological account of self-consciousness, which involves a commitment to what we might call the ‘reflexivity thesis’. According to this thesis, subjectivity is not a dissociable aspect of our mental lives, but a transcendental pre-condition for all conscious experience. Consciousness always, and without exception, implies a tacit form of self-consciousness.

An implication of the reflexivity thesis is that it is not possible for a person to be aware of a given thought without at the same time being tacitly aware that it is their thought. All our occurent experiences are furnished with what Gallagher and Zahavi call a pre-reflective sense of ownership or mineness in virtue of which we are able to immediately recognise them as our own.

If the reflexivity thesis is true, and all our thoughts are self-intimating in this manner, then how do we make sense of the phenomenon of thought insertion? Patients experiencing the symptom report that they are introspectively aware of a thought ‘inside their mind’ that is not their own, most commonly asserting that it has been put there by another agent or external force (Rachel Gunn provides some excellent examples of patient reports in her blog post here).

Gallagher and Zahavi’s answer to this question is that subjects do not, in fact, deny ownership for their thought. What sufferers really mean by ‘they are not my thoughts’ is ‘I am not the agent of the thought’ or ‘I did not intentionally bring this thought about’. By striking this distinction between agency and ownership, Gallagher and Zahavi are able to defend the phenomenological theory of self-consciousness at the heart of the minimal model.

I side with theorists such as Bortolotti and Broome, Billon and Martin and Pacherie in arguing that the agency-based approach suffers from some fairly damning inadequacies. Perhaps the most significant problem is that citing an abnormal sense of agency does not help us distinguish thought insertion from the non-pathological and mundane experience of having an unbidden thought.

But prior to explanatory issues, it is not clear why we should interpret patients as merely denying agency for thoughts, as opposed to taking them at face value as denying ownership. The reasons Gallagher and Zahavi provide for rejecting an ownership-based interpretation are unsatisfactory and ultimately question-begging; without first assuming that awareness and self-awareness are inseparable, we have little reason to suppose, as proponents of the agency-based account do, that subjects with thought insertion are equivocating.

Insofar as subjects with thought insertion lack a sense of ownership for some of their thoughts, the reflexivity thesis is too strong a characterisation of the relationship between awareness and self-awareness. While Gallagher and Zahavi’s notion of a pre-reflective sense of self is salient, I contend that in order to properly characterise the anomalous experience of those afflicted with thought insertion, the robustness of the minimal self must not be over-exaggerated.


4 comments:

  1. Hi Caitrin, glad to see you on here!

    Couple of questions about your post. In reference to analysis of delusions of thought insertion in terms of a sense of agency deficit you say: "Perhaps the most significant problem is that citing an abnormal sense of agency does not help us distinguish thought insertion from the non-pathological and mundane experience of having an unbidden thought."

    But there are two replies to this sort of reasoning. First that it's not obvious that unbidden thoughts lack a sense of agency, especially when we consider the distinction between a thick and thin sense of agency. Whilst unbidden thoughts don't seem to be thought for a reason (e.g. I don't seem to have an intention to have my shopping list pop into my head) -- they lack a thick sense of agency-- is it really right to say they don't feel like they come from me-- a thin sense of agency? e.g. I don't feel like no one or someone else thinks about my shopping list.

    Second even if the 'no sense of agency' charge could be made to stick wouldn't the difference between unbidden and inserted thoughts just be about whether or not a sense of agency where expected? I.e. for inserted thoughts a sense of agency is predicted but not apparent, but for unbidden thoughts it's not expected and so it's absence isn't important.

    So, in either case, it seems like the sense of agency is still useful for distinguishing inserted thoughts from unbidden thoughts.

    Beyond that if a patients says “But when I am thinking in this way, without being able to stop it… I have no mastery over the course of these ideas… it seems to me as if it is not me who generates them…” isn't that best understood as a denial of agency?

    let me know what you think!
    Glenn

    ReplyDelete
  2. Hi Glenn,
    Thank you for your response. Your questions are very good ones, and I’ll try to answer them to the best of my ability (though not sequentially, as it happens!).

    To answer your last point first:
    “Beyond that if a patients says “But when I am thinking in this way, without being able to stop it… I have no mastery over the course of these ideas… it seems to me as if it is not me who generates them…” isn't that best understood as a denial of agency?”.

    Yes, I do think statements like this reveal an abnormal sense of agency. I do in fact think that the sense of agency is disrupted in cases of thought insertion, a point I should have made clear in my post above. What I contest is that thought insertion can be fully explained by an abnormal SoA, or accurately characterised as a misattribution of agency. Subjects who experience thought control also report the belief or feeling of being controlled by an external agent/force. What they do not do, however, is deny that the thought is their own. There is a misattribution of agency, but not ownership. In thought insertion there is both.

    ReplyDelete
  3. In response to your second point:
    “…even if the 'no sense of agency' charge could be made to stick wouldn't the difference between unbidden and inserted thoughts just be about whether or not a sense of agency where expected? I.e. for inserted thoughts a sense of agency is predicted but not apparent, but for unbidden thoughts it's not expected and so it's absence isn't important.
    So, in either case, it seems like the sense of agency is still useful for distinguishing inserted thoughts from unbidden thoughts.”

    If I understand you correctly, I think this kind of reply works well for disorders of somatic passivity, but I don’t think it works for thought insertion. If my limb starts performing intentional actions on its own and I lack a SoA for the movements, this is going to strike me as pretty strange. I may even arrive at the delusional hypothesis that someone else is controlling my movements. It is difficult to see, however, why the experience of not having predicted a thought would be an experiential anomaly, when semantically coherent, fully-formed and unintended thoughts arrive in one’s stream of consciousness as a matter of course. That is to say, it is not clear why a malfunction at the subpersonal level (e.g. the absence of an efferent copy) would produce an aberrant phenomenology of passivity, as opposed to the normal sense of passivity characteristic of much of our mental lives.

    But I should probably bracket this, because perhaps it is a non-problem if we distinguish between a thick and thin sense of agency, as you suggest in your first point.

    ReplyDelete
  4. Finally, your distinction between thick and thin SoA strikes me as quite similar to, perhaps synonymous, with Sousa and Swiney's (2013) distinction between causal and intentional agency. The authors argue that experiencing oneself as the causal origin point of a thought is the ‘sense of causal agency’ and distinguish it from the ‘sense of intentional agency’, which is the experience of having willed a thought. Further, while the latter is absent in both unbidden thoughts and inserted thoughts, only inserted thoughts lack the former.

    Gallagher’s account (2005) actually addresses the problem of coarsegrainedness in a similar, though not identical, and certainly etiologically distinct way to Sousa and Swiney. Instead of pointing to a failure of causal agency, he argues that thought insertion involves a failure in time consciousness. While unbidden thoughts may involve surprising content, be of a distressing nature or even occur compulsively, they do not lack the quality of being self-generated. On Gallagher’s account, it is not merely the fact that inserted thoughts lack a phenomenology of wilful directedness which is the problem, but the fact that they show up in a person’s stream of consciousness without that person having a sense that they are their causal source.

    I think that both these accounts manage to 1) deal with the problem of distinguishing between unbidden and insert thoughts 2) neatly capture something in the phenomenology of thought insertion missing from standard agency-based accounts.

    I tend to think that the sense of being the causal origin one’s thoughts tells us more about the ‘mineness’ of a person’s thoughts than it does their SoA, but this claim needs more argument than I can provide here. There may be room, by drawing a distinction between thick and thin SoA, to hold firm to a revised version of an agency-based explanans.

    What I take issue with, regardless of the aforementioned issues, is the way in which agency-based accounts set up the explanandum. I argue that whether subjects experiencing thought insertion lack an experiential SoO or a causal SoA, they nonetheless fail to make a judgment of ownership for the inserted thought, and often attribute ownership (not just agency) to someone else. They do not merely arrive at the delusional hypothesis that they are not the agent of the thought, but also – in violation of the immunity principle- that the thought is not their own. Gallagher and Zahavi’s arguments do not establish that patient reports are equivocal.

    If subjects do not lack a SoO, but fail to recognise themselves as the owner of the thought, then this substantially undermines the role the SoO plays on Gallagher and Zahavi’s model; it is not clear that we’re entitled to say that the sense of ownership does anything but contribute to, as opposed to constitute, the self-awareness identifiable with the minimal self. In other words, even if this revised agency-based explanans is accurate, the minimal model is still open to the kind of challenge I introduced in my post.

    ReplyDelete

Comments are moderated.