As a philosopher specialized in cognitive and biological sciences, Dominic Murphy does not think that the narrative conceptualization of the self is accurate when it takes selfhood’s true nature to be narrative. Narrating and representing the self is, however, important for our being as humans – as social animals.
Your current work is on self-representations in the context of cognitive neurosciences and social psychology. What kind of philosophical implications does this have for our understanding of the self?
Philosophers in recent years have begun to take seriously the problem that traditional conceptions of the self appear to be incompatible with materialism. Concepts of a self that is sufficiently responsive to reasons to be the source of rational action were committed to a transcendent self, but with rare exceptions – who still cling to a transcendent sense – contemporary action theorists are materialists. The problem they face is to find a place in the contemporary cognitive and biological sciences for a version of the self that can serve as the foundation for a theory of autonomy: it is claimed that, in order to be a fully responsible agent, I must be able to introspectively access my beliefs and plans, so as to assure myself that they are really my own, and also possess an enduring self that can be the subject and object of the planning.
A self that can serve as the foundation of autonomous rational action is widely portrayed as narrative in nature. Many philosophers and psychologists have argued in recent years that the form of human self-representation is a narration in which episodes in one’s biography are woven together to form a coherent depiction of the self and to establish links between one’s past, present and future. This narrative conception of the self is widely advertised as a way to secure the foundations of a theory of rational agenthood without resorting to a metaphysically suspect conception of the self as transcendent.
I think this is largely undermined by contemporary cognitive science, which has established that we do not have the access to the sources of our behavior that it requires.
Introspection and self-representation depend on a set of loosely related capacities and do not form a psychological kind. Furthermore, the evidence for the narrative conception of the self is much weaker than is usually supposed. There is good reason to believe that the narrative, self-representational sense of introspection is a form of confabulation, since much of our mental life is unconscious, many of the sources of our conscious mental life are opaque, and many of our stories about why we act as we do are wrong.
Although a few philosophers, notably Gilbert Ryle and more recently Daniel Dennett, have defended such a conception of introspection, few have tied it to empirical work. (Dennett does, but he relies on a tendentious reading of multiple personality disorder1Daniel C. Dennett, Consciousness Explained. Little, Brown & Co., Boston 1991.). A number of psychologists have also defended the view more recently, but often in the light of psychodynamic theories that lack empirical support. But the psychologists and neuroscientists often argue that the unconscious consists largely of computational processes that cannot be brought to consciousness very reliably by self-examination, and conclude on this basis that introspection is a poor guide to our true nature. But that conclusion makes sense only on the assumption that the function of introspection is to uncover our true nature.
I aim to explore the possibility that the function of “self-knowledge”, construed as a story we tell about ourselves, is social: it has the function of mediating social interaction. Our representation of ourselves is chiefly for the consumption of others. In particular, I want to explore the idea that a fairly stable self-representation – a coherent narrative of oneself – is a way of reconciling two different bodies of research at the intersection of cognitive and social science: the first is the view that behavior is largely the product of situation rather than stable dispositions. The second is the view, common among experimental economists and psychologists influenced by evolutionary considerations, that one function of a number of mental states, especially affective ones, is to signal the existence of stable personal characteristics so as to develop a reputation that can influence repeated social interactions.
This theory is conceptually simple: reputation, operationally defined as the probability that one will take a certain course of action, influences the way others interact with you in the future – for example, by making them more likely to offer you a favorable exchange. But if our psychology is unstable, how can we present it in a way that makes it appear stable? Here may be one role for the products of introspection, by creating representations of ourselves – as possessing stable traits – that we can believe in and project outwards to influence others.
So stories and narrative competencies have an important role in social cognition and intersubjectivity. How about in the continuation of selfhood?
I think it definitely has an important function in social cognition. As far as concerns unified and continuing selfhood: we are animals, and our continuing identity as an animal is safeguarded by the biological mechanisms that keep us intact and let us navigate the world. So, I guess the question is, is there some level of personhood that is somehow more than animal, but has no social function? I suspect not, actually. I think that our continuing nature, in so far as it makes us a kind of distinctive thinking, planning entity, is bound up with our social existence. So I think narratives are essential to our nature, but that is because our nature is essentially social.
What do you think happens to the self if our narrative capacities are for some reason lost or diminished, like in some cognitive impairments?
It is a fascinating question, but really an empirical one. I think it depends a lot on the details of the case – amnesia, dementia etc. There will presumably be different consequences relating to different impairments. Take K. C., a canadian patient who became unable to form new episodic memories after a motorbike crash. He seems to have been unable to imagine himself in time – he couldn’t plan for the future or remember important auto-biographical information, but he could learn and retain factual information. So, it seems that K. C. could function as an agent from moment to moment, but not as an agent extended in time, and his abilities as a social animal were greatly impaired.
How do you see the difference between “self-representations” and “self-narratives”?
Good question. You might think of a self-representation as something that is updated quite quickly, like a snapshot of what you are like right now, and it could include a sense of where you are physically, what’s around you in space, whether your feet hurt, what you might do next, and so. Whereas a narrative would involve editing and constructing a representation of the most important stuff, stretched out over time, based on what you retain from successive representations. So you can imagine self-representations feeding to the narratives.
There might, though, be a kind of master representation, which is a bunch of important information about yourself that is not tied directly to the narrative – a sort of summary of who you are, rather than a historical account of who you are. That might constrain and inform particular representations at a time. The brain seems to alternate between enabling and suppressing the ‘default system’ or ‘default network’, which is a connected set of brain circuits that seem to do a lot of day-dreaming and self-related autobiographical thinking. It is possible that we have a guiding self-representation that is in charge of letting information into the master narrative, though it may be that the guiding systems can be hijacked by the default network and overwhelmed by it under some circumstances.
Reference & Bibliography
