Under the bridge - Intro chords

The intro's in the key of "D" and the song changes to the key of "E" when the verse kicks in.

Music theory is based on the analysis of "common practice" music - classical music from roughly 1600 through 1900. Popular music is often based on music outside that tradition (blues, etc), and since the 1960s a lot of popular music has been written by people who aren't aware of the 'rules' of theory. They write what sounds good to their ears, and don't worry about how it conforms.

In fact, this is how music theory evolves - theorists develop explanations for why things work when they don't fit the current explanations.

In terms of popular music, a common exception to the 'rules' is found in parallel motion - you can take whatever the current chord is, and move it. As long as the chord roots are found in the same scale, you still get a sense of key. An extreme case of this is Otis Redding's "Dock of the Bay"; that makes use of eight different major chords - the harmonized scale is G mixolydian, with Bb major used as a passing chord between B and A.

However, the intro is based around two arpeggiated chords: D major and F# major, which is strange because not only do those chords not occur naturally in the key of E major, but they don't occur in ANY single key. Even if we look at the chords as secondary dominants, F# would be the secondary dominant of B (the dominant chord of E major, so that makes sense), but D major is the secondary dominant of G major, which doesn't occur in the key of E major.

I'd also suggest that it's better to analyse chords based on what they're doing rather than what they're called. F# could be a secondary dominant , but only if it acts like one, i.e., especially as a 7th chord, resolving to its parent tonic chord. (F#7-B/Bm). In the intro you described, it's not doing anything at all to promote the key of B major or minor, so it's not a secondary dominant.
Another quick point, secondary dominants don't have to point to a key that's related to the original key. You said D major could be the secondary dominant of G major except that (the chord) G major doesn't appear in the key of E. That's fine. It doesn't have to. The chord just has to resolve (preferably as a 7th) to its own tonic a 4th higher and promote that chord (G/Gm) as a new tonic of the song (either briefly or permanently). In fact, it's the commonest way of modulating to remote keys.

In this case though, as with F#, it's not resolving anywhere, so it's also not a secondary dominant.

I'd also suggest that it's better to analyse chords based on what they're doing rather than what they're called. F# could be a secondary dominant , but only if it acts like one, i.e., especially as a 7th chord, resolving to its parent tonic chord. (F#7-B/Bm).

Oops, that's true, my bad.

Musenfreund, if the intro is in the key of D, that explains the D major chord but not the F# major chord. Unless, of course, this is what Noteboat was talking about with parallel motion; does this also apply to keeping the tonic chord and changing the chord type? (for example, playing a C major chord for one bar, then playing a C minor chord for one bar) because this would also explain the ending of the song, which pretty much revolves around these chords: A Am G F

Hmm, might be. I wasn't thinking so much of the F# when I answered as I was about the D. It's sort of like a parallel major chord maybe????

Steve-o, yes - you need to keep something constant... if you don't, the results sound disconnected. So you can

- keep the key constant (as in E to F#m)

- keep the chord type constant (as in E to F#), or

- keep the root constant (as in E to Em)

Cool, that makes alot of sense actually because I've seen that major to minor change (with the same root note) in alot of songs before. Thanks!