Monday, October 05, 2020

The different types of polyphonic singing 1: parallel polyphony

Joseph Jordania has outlined nine different types of polyphonic singing in his book Choral singing in human evolution.

I will be working through each type trying to give a simplified explanation and some musical examples. This week it’s parallel polyphony.

I’ve already outlined what polyphonic singing is.

Not every culture has polyphonic or multi-part singing. When they do, it can take many different forms.

parallel polyphony

Parallel polyphony is when one or more voices follow the main melody in parallel. Roughly speaking, each part starts on a different note, and when the main melody goes up, they go up, and when it goes down, they go down.

There are two different kinds of parallel polyphony.

1. tonally unconnected parallelism

This is when one or more voices stay a fixed distance (or ‘interval’) away from the main melody. Basically, the harmony parts are singing exactly the same as the tune, the only difference being that they start on different notes.

For those of us in the West who are used to tonal music, this will sound very strange and be incredibly difficult to sing.

Try this for example. Get together with a friend and choose a simple song like Happy birthday. Decide on two different starting notes. Go off into a separate room and practice singing the tune from your starting note. Then come back together and sing at the same time. See what I mean?

In the West, this kind of polyphony was one of the first to develop. After the simple unison (or ‘homophonic’) singing of Gregorian chant, a second voice was added which sang in a strict parallel. This is known as strict or parallel organum.

Here is a short example: 


There is a tradition in Iceland called Tvísöngur (“two singing”) which dates back to medieval times, but which is still sung today.

Most of the time the parts move in parallel fifths in a strict tonally unconnected parallelism. But at certain moment the parts shift places (the top part goes lower and the low part goes higher than the top part).

Here are two examples:


There is an account of part-singing described as ‘popular’ in northern England in the latter years of the 12th century. It is by Giraldus Cambrensis (Gerald of Wales), who says that in the north of England they sang in two parts. It is probable that what Giraldus refers to is a form of tvísöngur.


2. tonally linked parallelism

This is the kind of harmony singing that we’re most familiar with in the West. It was very prevalent in popular songs of the 1950s and 1960s:

One or more voice parts follow the main melody (either above or below it), but the distance between these vocal parts and the melody (or ‘interval’) can vary.

Rather than singing in a strictly parallel fashion with a fixed interval, we adjust the interval from time to time to that the harmonies continue to sound ‘nice’ (or ‘tonal’) to our ears. It’s something we’re so familiar with that we probably don’t even notice it happening.

Other cultures which have a polyphonic singing tradition like this are many of the sub-Saharan African countries. Here’s an example from Zambia:

next week

In the next post in this series I’ll be looking at drone polyphony.


Get more posts like this delivered straight to your inbox!

Click to subscribe by email.


Chris Rowbury




Monthly Music Roundup: