More on Gravity

Scott Aaron wrote to me to comment on my entry about gravity. I thought it was interesting enough to quote in (mostly) its entirety, with his permission:

Just read your blog posting on this subject. Don’t confuse the theory with the phenomenon. Gravity is not a fact. That a pencil will fall when picked up and let go is a fact. Gravity is the theory that purports to explain why it falls and the mechanics of the fall. The facts are the observables, i.e. the phenomenon. The theory is what explains the phenomenon.

There are many theories of gravity. There’s Newton’s theory, Einstein’s General Relativity, quantum gravity. When you say gravity is a fact, to which theory or version of gravity are you referring? When you say gravity has been tested, it is the theories that have been tested.

Passing experimental tests does not make a theory “fact”. Newton’s theory of gravity has been extensively tested for centuries. But it’s wrong (isn’t relativistic). General Relativity has been tested, but is wrong (isn’t quantized).

* * *
Andrew here.

I have one problem with this, and there are a few things in particular I like.

My problem is with this: “That a pencil will fall when picked up and let go is a fact. Gravity is the theory that purports to explain why it falls and the mechanics of the fall.”

This strikes me as semantics. We — English speakers — have chosen to use the word “gravity” as the name for the phenomenon of objects being attracted to each other due to their mass. So gravity is not a theory; we live it every day.

Scott wrote, “The facts are the observables, i.e. the phenomenon. The theory is what explains the phenomenon.” And to me, pencils (and people, and meteors, etc.) falling is the observable, and therefore it’s the fact. We have chosen to call that “gravity.”

But why does it work? Was Newton right? Einstein? That’s getting into gravitational theory.

What I like about Scott’s note is the discussion of testing. That’s what science does — comes up with a theory to explain something (gravity), then tests those theories. If they fail the test (or can’t be tested), it must be something else.