Scientific method
The core of the scientific method is that different observers of reality agree that evidence is in accord (to some degree) with a new hypothesized concept or model of "how things work." True or false can be established, but errors are so easily made (for many reasons) that peer review is essential.
21st century science has many issues because physical experimentation is so energy-intensive (think CERN in physics) or so time-consuming, that physical verification is troublesome to say the least.
When projections suggest disaster at some future point, as with oceanic degradation, waiting for disaster is an unsatisfactory option even though classical science posits that to be fully confirmed, reality has to accord with the projection. The best that can be done is model validation: evidence thus far confirms it.
As a consequence, a great deal of modern science consists of "thought experiments," working through the logic of a hypothesis, or even a conjecture, to see if it is consistent with prior knowledge, or a plausible modification or extension of it. This allows for more contention between different ideas, of course.
Scientific use of key terms use differs from colloquial use:
- Theory: A concept or model supported by some observed evidence, but not strongly enough to rule out other explanations. The term is often used by scoffers to denigrate what they believe is a harebrained concept -- or one inimical to their interests.
- Hypothesis: A concept or model structure in testable form, so evidence can prove it "true or false," or given the nature of experimental and observational error, it is highly probably or highly improbable.
- Law: A theory supported by all known evidence and refuted by none.
A "black swan" observation is one in which a single bit of evidence contradicts what has previously been regarded as a law: all previously seen swans are white.