Strangely enough, I'm reading Schoenberg's Theory of Harmony right now. I recommend the opening chapter to everyone interested in this topic, because it's one of the most well-written rants in all of music theory.
What Schoenberg opposed was the idea, which he claimed to be prevalent among music theorists in the late 19th and early 20th century, that we could discover "laws of beauty" which could be applied to make beautiful art.
Well, yes and no. Schoenberg certainly admits that certain intervals are more pleasing than others, and that perception was based on how closely they conform to the harmonic series. (Which, to stay on-topic, happens to be exactly what the researchers in this study contend). Schoenberg's argument was that "consonant" and "dissonant" tones are not opposites, as the words imply, but differ only by a matter of degree--how far out the series you go.
We can't imagine what was in the minds of the people who rioted at the premiere of The Rite of Spring.
Most reports, other than Stravinsky's self-aggrandizing story, point to the choreography, not the music, being the target of derision. For most of the performance, the audience couldn't even hear the music!
...we now understand that music theory, and the "rules" therein, are descriptive, not prescriptive. They are a language for understanding and talking about music in the tradition of the European common practice era.
I think no one understood that better (at the time) than Schoenberg, who wrote the Theory of Harmony from his own observation and not as a compilation of rules that he had been taught.
In that sense, it's like category theory in mathematics or design patterns in software engineering. they're not recipes on how to write programs or do maths, they are a vocabulary for understanding, reasoning about and talking about programs or mathematical structure.
And that is why we still read Theory of Harmony, today. It is important to note that Schoenberg did not turn against modal tonality until later.