Opening the book…
A statement counts as scientific only if it forbids something — if there is a conceivable observation that would prove it wrong. Falsifiability is what separates physics from assertion: a theory earns credibility by surviving genuine attempts to break it, not by explaining everything after the fact. A claim compatible with every possible result tells you nothing.
For any theory or claim, ask what specific observation would refute it, then design the measurement that could deliver that result. Prefer hypotheses that make sharp, risky predictions over vague ones that fit anything. Track where a theory is most likely to fail and probe there; a decisive negative is worth more than another easy confirmation.
In practice no single result kills a theory outright — auxiliary assumptions can be blamed, so refutation is a judgment across many experiments. Some foundational principles are hard to test directly and are held provisionally for their fruitfulness.