There are a lot of problems with substance dualism, but for me one of the most compelling is also one that has received less attention than it deserves. It is an objection that confronted the poster-boy for substance dualism, the French philosopher Rene Descartes. It can be summed up in a question: If dualism is true, then how is unconsciousness possible?
...there is still something odd about about the soul carrying out unconscious processes, especially where these processes are presumably at least largely independent of what is going on in the brain. And it is still the supposed independence of the soul from the brain that creates problems for the dualist, since it looks very much like the properties that the soul is being invoked to explain are determined by the state of the brain.
This strikes me as a pretty unintelligent critique of dualism:
i) Although the notion of unconscious thought might seem to be self-contradictory, we know for a fact that reasoning can operate at a subliminal level. Consider scientists and mathematicians who struggle to solve a problem, fail, sleep on the problem, and awaken with a solution.
Likewise, consider the creative process. When a novelist writes a story, where are these ideas coming from? Although writing is a conscious activity, the ideas well up from the subconscious.
ii) To my knowledge, we generally forget what we dream about unless we wake up during or right after the dream. There may be a lot of mental activity going on when we sleep, but we just don't remember it.
iii) In substance dualism, the soul is independent of the brain in the sense that it can survive apart from the brain. The soul is ontologically independent of the brain. But when a soul is coupled with a brain, mental awareness may be dependent on the brain.
iv) To take another comparison: words and sentences are concrete. They can be visible or audible. Yet meaning is abstract. Meaning is not reducible to sound waves or the chemical composition of paper and ink. That symbolizes meaning. That encodes meaning in physical carriers. If you destroy the carrier, there's loss of information. The listener or reader can no longer access the information. But it's the conduit that's destroyed–not what it conveys.
v) Take the position of Rupert Sheldrake, which he develops from Henri Bergson:
“Where?” is the wrong question. Memory is a relationship in time, not in space. The idea that a memory has to be somewhere when it’s not being remembered is a theoretical inference, not an observation of reality. When I met you this morning, I recognized you from yesterday. There’s no photographic representation of you in my brain. I just recognize you. What I suggest is that memory depends on a direct relationship across time between past experiences and present ones. The brain is more like a television receiver. The television doesn’t store all the images and programs you watch on it; it tunes in to them invisibly.
I don't necessarily agree with that. But it represents a paradigm-shift, and it's worth exploring from that perspective.