Not in science, they don't. We're trying to establish the reality of consciousness and its replicability. We NEED to define those words.
Yes, we do. We have a big fat dictionary, and about half the words represent things that we can describe with physics, and half of them represent words that we
can't describe with physics. Some of those represent words that we don't think exist, or which may not exist. Many, many others represent things that we
do thing exist. And we have no solid definition for them.
There are several approaches to being unable to define "happiness", or "pain". We can use the lack of a definition as a reason to assume that they don't represent anything real, in the same sense as "pig" or "electron". This has the disadvantage that most of us do regard them as real things, and act accordingly. We can classify them in terms of the behaviour they tend to produce - but while this is a helpful way to identify the feelings, it's certainly not a definition. We can make claims about how they are produced, but if not backed by a physical theory, such claims need more justification.
My preference is to accept that we are dealing with real quantities, of some unknown nature, and to try to investigate them.
We already know. But we need to see past our illusions and delusions to get to reality. Otherwise we'd still think the sun spins around the Earth.
I'm for good and against evil myself. Hooray for knowledge! Down with ignorance!
How does one experience something "indirectly" ?
Everything in the world is experienced indirectly. We have no perfect perception of anything, except the very fact of sensation itself.
You still haven't answered the question. What do you mean by "spontaneously" ?
If you mean "without prior programming" then you are wrong. Everything you do is either deterministic or random (IF, and only IF, quantum fluctuations can actually affect us at that level), so there is no room for free will in the old sense of the word. Everything we do is pre-programmed. A computer just needs to be able to rewrite its programming using new experiences in order to simulate this, and some programs already do that.
If a program is incapable of claiming consciousness spontaneously, then it's incapable. I suspect that it may be. However, if a program
isn't directly programmed to produce consciousness, but in some manner asserts it, then that would certainly be evidence - possibly the first evidence - that computational consciousness is possible.
However, to directly write a program to assert consciousness is trivial and meaningless.
So the solipsists claim. But they're wrong.
It's nothing to do with solipsism. It's obvious that we experience the external world imperfectly. That is not the same thing as claiming that there is no external world, or that everything is generated internally. I can't see any link whatsoever with an assertion that consciousness represents a problem in physics that has not been solved. The dualist accusation, while not justified, at least has something to do with the argument. Solipsism seems to be just thrown in as one of those "bad philosophical ideas" that can be used to label an opponent. What next - idealism I suppose. That usually comes up.