“The” Self

Wednesday December 29, 2021 — Taipei, Taiwan

I was talking to someone recently about the concept of “self” — we disagreed on whether it made sense to consider people as having a singular self (I don’t, they did), and perhaps on what it means to be a “self” at all. These are some notes and thoughts from that conversation.

First off, we should try to sketch out a definition of what “self” is. This definition, like most definitions, will likely fall apart under scrutiny, but it’s good to start somewhere, and we can always adjust as we go along. The person I was talking to started off by saying that the self was something contained within the brain — that things in the brain could affect the self, but things outside the brain could not, and that since every person has a single brain, they can only have a single self.

That’s fair enough, but I see a couple problems with it. The first is that it’s not clear to me that the brain is actually a single entity — split-brain experiments make a reasonable case that could be more accurate to think of a brain as multiple independent subsystems communicating and collaborating. Why do we choose to say that a brain is the unit of “self,” when it seems that there could be a smaller unit that exhibits self-like properties (having memory, making decisions, etc)?

One might say that the reason we choose a single brain as the unit of self is that it controls a single body, but that brings us to a second problem: there are many organs in the body that secrete hormones that have a significant effect on the dynamics of cognition. If you accept this, then you need to expand the self to include not just the brain, but other parts of the body as well. When I mentioned this, my conversational partner expanded their definition of what contained the self to include organs that secrete hormones that affect the brain. This brings up new problems, though: I have a insulin pump, which excretes a hormone that has a indirect but significant effect on the brain.Insulin, which affects blood sugar, which in turn affects brain function. If you accept that organs which regulate brain function are part of the self, then it doesn’t make much sense to say that an insulin pump, which is essentially a synthetic version of one of those organs, cannot be part of the self.

But if you accept that an insulin pump can be part of the self, what else might be? Music, for instance, probably has a more significant impact on cognition than insulin does,Assuming one has fairly typical insulin levels. so should we consider a music player that someone always carries with them as a part of their “self”? A computer, similarly, can have a huge impact on how one thinks — could that be a part of the “self”?

It’s because of these difficulties that I don’t think that thinking of self as a singular entity makes sense. Instead, a more useful model to me is that “self” is a lens that allows us to split the world into agents, and those agents could exist at levels smaller than the brain (different halves of the brain, or perhaps even smaller functional subdivisions) or larger (groups of people working together can have a sense of selfhood collectively). Many smaller selves cooperating can make up a larger self.

Where we choose to draw these lines should depend on what sort of analysis we’re doing in the moment, but restricting ourselves to the idea that a single physical system can only have a single self is limiting and harmful towards nuanced analysis.