Is Solipsism Simple?

Solipsism is the view that nothing exists but one's own stream of conscious experience. In The Problems of Philosophy (1912), Bertrand Russell says that although it's logically possible that solipsism is true, solipsism should be rejected as less simple than the hypothesis that an external world exists. But is realism about the external world really simpler than solipsism?

On the face of it, you might think solipsism is simpler. After all, it involves radically fewer entities. That's the great Ockhamesque beauty of it. Solipsism may be crazy, but at least it's simple!

Russell employs two arguments against the simplicity of solipsism. First:

If [a] cat exists whether I see it or not, we can understand from our own experience how it gets hungry between one meal and the next; but if it does not exist when I am not seeing it, it seems odd that appetite should grow during non-existence as fast as during existence (p. 23).

Second:

When human beings speak -- that is, when we hear certain noises which we associate with ideas, and simultaneously see certain motions of lips and expressions of face -- it is very difficult to suppose that what we hear is not the expression of thought, as we know it would be if we emitted the same sounds (p. 23-24).

Thus, he concludes, "every principle of simplicity urges us to adopt the natural view, that there really are objects other than ourselves" (p. 24).

Now, I'm inclined to think that Russell's argument here is very inadequate. But let me quickly say that I've found it surprisingly difficult to uncover what I'd consider to be better arguments against solipsism in the philosophical literature. Plus, Russell is famous! So let's take him seriously.

To see the core problem with Russell's argument, consider dreams. In dreams, cats can grow quite quickly hungry between appearances, despite their nonexistence in the interval. And the voices and faces seen in dreams reveal the real existence of no other independent mind. It seems no great violation of simplicity to suppose that, in a dream, the appearances of the cat and the voices and faces are concocted on the spot by me. No need to posit a giant, really existing universe, light-years upon light-years wide! And the solipsist, it seems, can just treat waking experiences the same way. Simple!

Russell is of course aware of dream skepticism, addressing it thus:

But dreams are more or less suggested by what we call waking life, and are capable of being more or less accounted for on scientific principles if we assume there really is a physical world (p. 24).

I think one might just as easily turn this argument on its head. If we assume solipsism rather than realism, we can invoke principles explaining why cats and people seem to behave as they do: They are imperfect projections of me upon my imagined world, based on what I know from introspection about myself. That theory is of course sketchy and incomplete, but so is the current scientific account of the content of dreams!

Now it might seem that postulating an external world behind appearances can at least explain correlations that must remain unexplained in solipsism. For example, if there is a real penny that I'm both looking at and manually rotating, the real existence of arm and coin explains why such-and-such changes in visual experience co-occur with such-and-such changes in tactile and proprioceptive experience.

I see two obvious replies for the solipsist:

First, why can't it simply be a law of my experience that such-and-such tactile and proprioceptive experiences will tend to co-occur with such-and-such visual experiences? Surely there's a theoretically discoverable structure to such co-occurrences -- a structure not so different, perhaps, and probably simpler, than that employed in the realist's account of tactile and visual perception and motor control and its relation to external objects. After all, realists' psychological theories, if they're really going to explain the relation among the experiences, require complicated overlapping and competing brain mechanisms for determining, among other things, visual shape and orientation from optical input.

And second, if simplicity really favors the theory with fewer unexplained coincidences, won't solipsism win hands down, even if it leaves a few things unexplained that the realist can explain? The small world of the solipsist will have vastly fewer such coincidences in total, and vastly fewer free parameters, than the enormously large, fine-textured, and richly populated world of the realist.