Secondary Dominants – part 1


This is my personal favorite harmonic device. It’s something even budding composers are familiar with but it seems to me that it is tragically underused in recent years.

This even though it is a staple of classic popular and traditional music, not to mention showtunes. For those not too familiar with this term, it is any dominant used in a song that is not the primary – or diatonic – dominant (so G7 in C major is primary whereas E7 would be secondary). It is, along with using chords from the parallel major or minor scale, one of the only theoretically acceptable uses of nondiatonic chords.

I don’t know where I would be without this tool for spicing up the harmony in some of the most refreshing and unpredictable ways. The spice results from the secondary dominant suggesting a different tonality, or key. It greatly enriches the harmony and keeps the listener interested. Here are some of the contexts secondary dominants appear in in practice (that I can think of):

  • in anticipation of their tonic (eg. E7 preceding Am)
  • as a successive enhancement of the corresponding minor (eg. A7 following Am)
  • as a substitution for their tonic or chord of resolution (eg. E7 instead of Am)
  • as a substitution for the corresponding minor chord (eg. A7 in for Am)
  • as an unresolving (or nonfunctional) dominant (eg. E7 not resolving to Am)
  • as a strong means of departure
  • as part of a succession of secondary dominants (dominants of dominants)

The first and second applications seem to be the most common and often appear together. Oftentimes you are introduced to a new chord or reintroduced to a familiar one via a secondary dominant. This momentarily reframes the melody (which remains the same) by treating its original accompanying chord as a tonic chord now, adding piquancy and emphasis as a result. It is often employed as a means of harmonic progression to signal a change in the direction of a song

A very strong effect is having a secondary dominant follow its corresponding minor, unexpectedly tugging at – or bending – the harmony and propelling the melody into the upcoming transitory tonality in a greater swell of emotion. Also, cascading dominants resolving to dominants (usually no more than 3-4) often result in very satisfying cadences but can also provide the framework for melodies proper.

And secondary dominants usually appear in a combination of the practical applications discussed. This is obviously decided by the demands of the melody or the unmitigated direction of the song. The best way to illustrate their use will be in looking at a specific piece of music, which I will do in part 2.


When does a song need a bridge and when is it better off without one? A musical question best solved musically – by ear – of course. Still, what is it then about a particular song that makes the songwriter’s ear decide to go with or without one?

The Oscar-winning songwriter Joel Hirschhorn advised to always write a bridge anyway and decide later… Great advice and something that I usually subscribe to. Usually, because some songs just have such a direct and visceral message or melody or such a delicate ambience that inserting a bridge would ruin the effect. And everyone – songwriters and laymen alike – has heard the jarring (even cringing)  result of a bridge shoehorned into a song. Less is more should always be the overarching rule, trumping all other considerations.

As for songs where a bridge fits beautifully into its surroundings, one might wonder how the song would have fared without its pleasantly breaking up the scenery… Where it is seamless, the listener can’t even imagine it ever being absent. That’s the delight of a great bridge – of a refreshing departure from the tonality, rhythm, lyrical theme or style, and sometimes even tempo and meter of a song. It’s like drinking just the right wine with a great meal.

Another interesting thing to consider is whether bridges could be musically interchangeable between songs – especially in the repertoire of the author. This is because, if there are songwriters out there like me, their bridges would have been considered (and almost approved) for use in other works. The world of music composition is a whirlwind of accidents and inspiration, involving endless bouncing and shuffling around of ideas new and old, all while staring into the void… Anything goes really – especially if someone is pointing a gun at the composer.

Since the very raison d’être of a bridge is to depart from the general proceedings, a chance inclusion in a song should be even less surprising. Still, it perhaps subverts the idea of songs being musically integral and singular. Anyway, a subject worthy of many more posts…