There is a good deal of nonsense circulating about historical temperaments, and the saddest thing is that some of it comes from actual musicologists. Kyle Gann even claims that equal temperament could not be tuned exactly until 1917; if that were really true, I wonder why Werckmeister advocated it for performance as early as 1697. I cannot imagine that he was advocating something that could not be done. Frescobaldi and Doni were already arguing about it in 1640, pushing the date even earlier; and by 1753 C. Ph. E. Bach even inserted a description of how to get it. (It is slightly confused, but the general idea, that the fifths must all be tempered and every key must have the same intervals as any others, is there.) And in fact, one of the other key figures who most forcefully argued in favour of it was Rameau, who rejected the very idea that just intonation was natural. From the time of J. S. Bach on, the voices against equal temperament are a dwindling, reactionary minority; a vocal minority, to be sure, but still a negligible one.
A lot of this may come from the fact that we have been so bogged down by teaching advanced theories of analysis that many people forget the basics. Schenker, for instance, omits them in his account of his theories, presumably because he thought they were obvious and well-known to everyone. They certainly were well-known to everyone in his time, but they were certainly not obvious; now that they're not well-known to everyone, most people have forgotten them, and a lot of analyses betray the fact that the writer is trying to run before he or she can walk. One of the funniest and saddest results is that very few people actually seem to know what a modulation is in Classical tonality (hint: it's not just a chord attacked by its dominant).
In the interest of space, I am not going to go into very much detail about how tonality developed after the 18th century; suffice it to say that the increasingly pervasive chromaticism makes just intonation as a theory even more laughable and useless for Western classical music in the 19th and 20th centuries. So I will cover only the most complex case, that of 18th-century triadic tonality, where the tonic triad is the master of all the chords and everything else is construed as consonant or dissonant in terms of it, even if it does not seem dissonant at first glance and is still a perfect triad.
This is, of course, a simplification, but the fuller explanation will have to take up a post by itself.
(To be continued...)
I have recently resumed writing up the first part of this exposé; in the meantime, here is a sketch of a table of contents.
Part I: Harmony
[The overtone series and the major and minor triads]
[Circle of fifths]
[Dominant vs subdominant]
[More exotic harmonies]
[The minor mode and how it allies with chromaticism]
[The relationship of equal temperament with just intonation and why the former is a viable theoretical basis for late 18th-century music and the latter isn't]
[Hierarchy; when are we in a key and when are we just on a key?]
[Affective key characteristics; absolute and relative relationships]
Note that I write "Part I"; this is because tonality in the Classical sense is not merely a means of harmonic organisation, and I plan to later add parts on other musical elements, such as rhythm, melody, texture, and how they work together to create form. The recognition of these as having an equal role to harmony, and the possibility of rhythmic and textural consonance and dissonance, is necessary for properly understanding 20th-century music (not all of it atonal; Debussy likewise demands being understood in those terms). Of course, in the late 18th century, these concerns are usually subsidiary to harmonic dissonance, and only become the foreground at extraordinary moments - which is in a sense the inverse of Berg's technique of introducing tonal quotations at extraordinary moments.