Believing only what you can see
"So, you only believe in what you can see, hear, or touch?" It is a (somewhat derisive) phrase you might come across every now and again if you don't believe in the supernatural. It's inevitably followed up with something about believing in love, freedom, or subatomic particles. After all, the reasoning goes, if you don't believe in the supernatural because you can't see it, how can you believe in love, freedom, or subatomic particles?
There's a trick in that argument. While a naive skeptical position might be "If can't see, hear, taste, touch, or smell it, it doesn't exist," and you might find some people who use that language, very few people actually work that way, even those who say it.
I believe in, say, France. I've never been to France, so I've never seen France, I've never touched France, I've never heard France… but yet I believe in France. I believe in France because "France" is a theory that explains quite a bit (the history of western Europe, for example). Any alternative theory which proposes that France doesn't exist would have to explain everything too, plus explain why things look like they do without France existing.
It's possible all the history books, encyclopedias, airlines, and international travelers I've met are in on a conspiracy (or are merely wrong), but "France exists" seems like a well-established theory with plenty of explanatory power and lots of good supporting evidence.
A second example: We all know glass is a slow liquid, not a solid. We've never seen it flow, we've never felt it flow, but we all know it does. We know this because old window panes are wavy and thicker at the bottom because the glass flowed down due to gravity. It's a "supercooled liquid" and can flow very slowly.
But… that's almost certainly false. We all believe(d) it, but it's probably not true. Glass does not flow. Old window panes are wavy because ye-olde glass-making techniques produced wavy glass. The glass was often thicker on one end for the same reason. The people who put the glass in the frames then consciously put the thicker side on the bottom because that's more structurally sound. If glass really flowed, it would not just be thicker at the bottom, it should bulge, and even possibly flow out of the window frames.
Lots of people believe it, and they have lots of evidence (the old window panes that are wavy and thicker at the bottom), but it's not true. I've heard it repeated by engineers. I used to believe it, but now I don't.
What's the difference? Why do I still accept the "France exists" theory but I now reject the "glass flows" theory?
The "glass flows" theory attempted to answer the question "why are old windows ripply and thicker on the bottom?" I heard this as a child and accepted it because it answered that question. I really didn't think too critically about it. I didn't consider any alternative theories. I didn't consider that the theory didn't fit the data all that well (no bulging or flowing out of frames). I accepted it uncritically, partly because of the "hey, that's neat!" factor. Everyone does.
So if the evidence presented is actually evidence for something else (limitations of ye-olde glass-making technology), the only evidence left is "lots of people believe it". "Lots of people believe it" is really only evidence of one thing: that lots of people believe it.
Lots of people also believe "France exists", but that's not all the evidence there is. "France exists" gets support from many places: history, geography, archeology, travelers, linguistics, toast, bread, coffee pressing, and kissing. A tremendous number of things we understand about the history of Europe depend on France existing. The theory "France exists" answers a lot of questions which can't be answered so neatly otherwise.
Short of a giant conspiracy aimed at fooling me, specifically, I feel fully justified in believing France exists, even though I can't see it.
To me, the supernatural seems much more a "glass flows" belief than a "France exists" belief. A huge number of people believe in one supernatural thing or another (ghosts, gods, spirits, magic, etc), and it's used to explain a few things (strange feelings, strange sightings, strange coincidences). We hear about the predominant supernatural theory of our society when we are kids, accept it and say "that's neat!", and never really question the fundamental idea all that much again. Even though we can't see it, it answers a few questions, and that's good enough.
But the supernatural once explained a much larger chunk of our world. Pretty much everything was caused by something supernatural. Birth and death were miracles performed by spirits, demons, and angels. Life contained souls and chi. Crop success and failure were both acts of gods. Lightning, thunder, rain, hail, wind, tornadoes, storms, droughts, floods, ice, and hurricanes were all acts of spirits of some sort or another.
But proto-scientists began chipping away and finding better explanations for natural events. They tested things and explained whole swaths of the world once handled by "supernatural things we couldn't see".
Now, there are fewer and fewer true mysteries in need of a bottom-up explanation. Some of the existing explanations involve "things we can't see", but most of them are well supported. For example: subatomic particles. I "believe" subatomic particles exist: there's a chance the theory's wrong, and I've never directly seen them, but they are part of a theory which explains quite a lot of the world, and has been repeatedly validated.
We "believe in them" even though we can't see them directly because they answer many questions, and scientists can use the theory to predict how new materials will behave and engineers have used the theory to design circuits and microprocessors. This doesn't mean the theory is absolutely correct, of course, but it's the current "good enough" theory we can actually use.
But supernatural claims can't be validated (or at least haven't), and don't offer much explanatory power, and almost always introduce more questions than they "solve".
For example, if we believe that Pele the volcano god causes the mountain to erupt, that "solves" the mystery but introduces a much bigger mystery, and it gives us no predictive power. If there were other evidence that Pele existed, like a big voice coming from the mountain that said "I am Pele! Fear Me!", then we might use that theory. But if the reason the theory of Pele was proposed was to answer the question "Why is the mountain exploding", saying "Pele did it" doesn't explain anything, really. Even if I didn't know about plate tectonics or the Earth's internal heat, it'd be far better to say "I don't know why it's exploding" than to say "Pele did it!"
So skepticism is not limited to "believing what we can see, hear, or touch," but that doesn't rescue the supernatural.