Daniel Dennett, as I mentioned, has criticised the Cartesian idea of the self extensively. He has argued that assuming there is a “central point” where everything comes together ignores how much work is already been done within peripheral regions of the brain. He also argues (and this is what I want to concentrate on here) that the “mind” or “self” is best understood as a construct which we achieve with the aid of language. To show how this better helps us understand “who we are”, he draws an extended analogy with the idea of a centre of gravity (Dennett 1992).
The centre of gravity of e.g. a chair is not one more physical thing in the world in addition to all the atoms that make up the chair. It is a kind of theoretical fiction, but one with a well-defined role in physics, and one which describes certain real properties of objects. Although it is nothing which we can perceive directly, it can still play a role in certain kinds of causal explanation.
For instance, if someone says “why doesn’t that chair fall over? ” we can say “because it is stable thanks to the position of its centre of gravity”. That this is a causal explanation is evident if we contrast it with an alternative, like “because it’s fixed to the floor”. The concept of a centre of gravity is therefore of something which “does work” in the physical world; having a centre of gravity makes a difference to how things behave.
Just as the physicist uses a mixture of observation and mathematical deduction to establish the position of an object’s centre of gravity, Dennett argues, we all typically perform an act of interpretation on the actions of other human beings in order to posit a central point in relation to which the story of those actions can be made sense of. We could, in principle, make sense of John’s actions by telling a story about physical causal mechanisms, all the way down to the level of interactions between the basic constituents of John’s body, but it is much more useful to us to utilise our capacities for telling stories to weave a narrative around a central character called John, a self. This allows us to make judgements about character, form predictions about what John will do next, and so on. It also allows us to identify someone across time, in relation to the position of their body.
But this does not indicate that the self is identical with the body, or indeed with any part of the body. Just as the centre of gravity of the chair is not (and indeed, cannot be) identical with any part of the chair, the self is not identical with the body, or indeed any part of it. The self differs from a centre of gravity insofar as it has a spatio-temporal position that is only “grossly defined”:
Roughly speaking, in the normal case if there are three human beings sitting on a park bench, there are three selves there, all in a row and roughly equidistant from the fountain they face. Or we might use a rather antique turn of phrase and talk about how many souls are located in the park. (“All twenty souls in the starboard lifeboat were saved, but those that remained on deck perished.”Dennett 1992
This means, therefore, that the self cannot be accurately localised anywhere within a room, within a human body or even within a brain. It is, for Dennett, an effect of interpretation that serves certain evolutionarily useful purposes (increasing our chances of predicting what others are going to do).
The other corollary of this view is that our own selves are exactly as real as the selves of others: we are the centres of stories we tell ourselves over time, actors and patients within dramas which we weave in concert with others. We have no direct, transparent access to something like our “true self”: in fact, our own selves are also the products of interpretations we carry out upon our own thoughts, desires, beliefs and actions.
Dennett, D. (1992). “The Self as a Center of Narrative Gravity” in F. Kessel, P. Cole and D. Johnson, eds, Self and Consciousness: Multiple Perspectives, Hillsdale, NJ: Erlbaum, 1992.