Mind-brain identity theory

This is a somewhat ontological topic, so in one respect it shouldn't belong in my queries.
However, I have successfully snuck around being a hypocrit because I suggested that in order to be able to answer ontological questions, I could assume an empirical method. Otherwise I'm just stuck at Hume's roadblock.

The mind-brain identity theory suggests that minds and brains are NOT two different things; but rather, that all references to these are really to brains and brain states.

This theory stands in favor of a monistic materialism and in opposition to dualism. Can it be empirically argued for? Not really, since to prove this would be to disprove dualism, which can't be done since spiritual reality cannot be empirically observed.

I can use Occam's razor to get rid of the spirit world, though, but I haven't thought about Occam's razor in depth. So what else can I talk about here?

It is related to eliminative materialism, which predicts that at some point in the future, language will change to more accurately represent what is going on in the brain. Instead of saying "I am happy" you might say "I just had a bunch of endorphins released in sectors A4, B95..."
(This is kind of an extreme case, but the idea is as science moves forward, the gap between sensation and corresponding reality will become smaller). If the eliminative materialists are right, dualism would eventually become less popular of a perspective, less intuitive, and to de facto eliminated.

However, if I were to approach this subject from the same level of skepticism as I have approached, say, the existence of God, then I would be forced to conclude that as this is merely a theory, there will never come a point where we can for certain say one system is true over another.