Sounds like someone is having fun with semantics at this thread's expense.
Quite the contrary.
From the distortion spectra, research has shown that certain distortion profiles will produce certain "sound"- compare the distortion spectra of many tube amps vs ss a/b vs class d and you find many describe the tube sound as "warm", "rich", "lush", etc. Class d as "neutral", "clear", etc. Certain harmonic profiles lead to similar perceptions, thus, from the measured performance, we can indeed make certain predictions on how an amp will "sound". Will everyone "like" that sound? No. This isn't semantics, it's psychoacoustics, an admittedly soft science.