Thursday, 14 December 2017

Lies, Fictions, Yarns And More

See Lies, Fictions And Yarns.

Sean pointed out another kind of untrue statement, an error. Thus:

Untrue Statements
"yarns" (?)

But there is also a kind of statement that is intermediate between true and untrue. A scientific theory:

is the best current approximation to the truth;
explains some phenomena;
but is provisional - always subject to revision.

Some sf shows the scientific process:

in James Blish's They Shall Have Stars/Year 2018!, new discoveries are made despite personal conflicts between some scientists;

in Blish's The Quincunx Of Time, messages from the future assume different scientific paradigms;

in Poul Anderson's Starfarers, interstellar explorers encounter quantum intelligences that receive tachyonic messages from their future.

If a novel about a long interstellar expedition is based entirely on current knowledge and theories about remote stellar regions, then it is bound to be inaccurate because such an expedition would gain new knowledge and revise current theories.


Paul Shackley said...

Kaor, Paul!

I would consider a scientific theory to be a "reasoned speculation."

And I like the last paragraph of this blog piece. My hope is that actual expeditions will behave exactly as you described!

Merry Christmas! Sean

S.M. Stirling said...

In science, a hypothesis is speculation.

A theory is the currently accepted explanation which has been repeatedly tested and has not (yet) been falsified. (Science doesn't deal in truth, of course.)

The linguistic problem here is that in common usage, "theory" means what "hypothesis" does in scientific discourse. That's unfortunate.

Paul Shackley said...

Richard Dawkins calls evolution a "fact," I think merely to counteract the argument that, if it is only a "theory," then it has not been proved yet.

S.M. Stirling said...

Well, yeah, but he's a polemicist speaking demotic. And not a very scrupulous one, at that.

Paul Shackley said...

Sheila once used the word "theory," meaning theory as opposed to practice, whereas the guy she was talking to took her to mean theory as opposed to fact. Big difference.

Paul Shackley said...

Kaor, Paul!

I thought theories are attempts to explain or understand things. And which can be tested.


S.M. Stirling said...

Specifically, a scientific theory must be falsifiable -- that is, there must be a way to show, if it's false, that it is false.

General relativity is an exemplar of this. There are multiple ways it could be falsified if it was incorrect; predictions of the effect of light on gravity, or of high speeds on time dilation.

Quantum mechanics is full of them too - entanglement and action at a distance, for instance.

In both cases, the theories have never -been- falsified; they passed all the tests, and increasingly exhaustive and precise ones as our experimental techniques became better.

The amusing thing is that they flatly contradict each other, from which one can deduce that both are incomplete descriptions.

A classic example of a non-falsifiable hypothesis (and hence not a scientific theory at all) is Freudian psychoanalysis.

There is literally no way to falsify it. It "explains" everything, and therefore nothing.

A non-falsifiable statement is worse than wrong: it's actually meaningless, a semantic null set.