Taking All the Fun out of a Folk Song


Level of Authenticity



“Writing about music is like dancing about architecture.” (I can’t find a definitive attribution for this quote).


I got the folk song “Sovay” stuck in my head over the weekend and I decided that my mission was to investigate claims that it was in an unusual meter.  TL; DR: It’s not, but it does have some other interesting structural features.


I think I first heard this song sung by Cattea the Studious or Baron Halfdane some time in the late 1990s.  Later I acquired a version by The Pentangle (1968) which, try as I might, I could not love.  Finally some time after 2004 I acquired a version by Dave Swarbrick and Martin Carthy on an album helpfully and informatively titled “English Folk Anthology”.  I don’t know when this recording was made but it appears that Mr. Carthy has been singing this song since at least 1965.  (He ain’t proud….or tired….)

Because they are passed down orally, English folk songs are best thought of as belonging to “song families” which can be given date estimates, in contrast to pieces of literate/art music, which can often be said to exist in a particular, definitive form, to have been composed by a specific person, and in a specific time range.  According to the Ballad Index (accessed August 8, 2017), the Sovay/Female Highwayman song family dates to before 1845.


The Ballad Index indicates the Sovay/Female Highwayman family is found in the USA, Canada, England, Ireland, and Australia.


To see if I could!


I learned the song from listening to a recording, which is not exactly how the oral tradition used to work in the days before recordings, but is much closer than if I had learned it by reading it off a page.  Instead of just looking for a transcription, I decided to see if I could figure it out on my own.


The first thing I did was to write down the words to the first verse on a piece of staff paper.  I put the line breaks where the ends of the phrases in the verse were.


I feel like this is the most difficult part of music to convey in writing alone.  You can work out where the notes stand in relation to each other by doing a lot of math (eg, when you halve a string or air column, it vibrates at a frequency that we think of as being eight notes apart from one another).  Otherwise, you basically have to rely on your audience having heard a selection of music in your tradition of choice.  I did try here to provide songs I think will help you understand it, but if you are lost or want to know more let me know and I will answer your question in person or by recording and posting a video of me singing or something.

I started writing down the tones that went with each of the words in the verse.  How did I know what line to start on and which lines or spaces to put the tones on?  In modern music theory we think of there being seven degrees in a scale: do, re, mi, fa, sol, la, ti, (and back to do).  You should be able to locate these degrees in relation to each other if you sing “Do(e), A Deer” from The Sound of Music.  In medieval and renaissance music theory, they thought of there being six degrees: ut, re, mi, fa, sol, and la.  Contemps had their own versions of Doe A Deer to help them remember where the degrees were in relationship to each other, for example “Ut Queant Laxis” sung by Polyphonia in Montengarde.  In order to move on up or down the scale and reach notes beyond those six, you have to recalculate on the fly in your brain, because you can start again at ut once you reach either fa or sol.  Having learned modern music theory first, I find the on-the-go recalculation challenging to say the least.

If you think about the seven (or six) degrees, or about the white notes on a piano keyboard, the thing to note is that the spaces between the notes are not all equal.  On a piano keyboard, some white notes have a black note between them and others don’t.  The ones that do have a black note between are called “tones” and the ones that don’t have a black note in between are called “semitones”.  If you want to map these to your solfege, the semitones are between mi and fa, (and again between ti and do if you’re a modern).  If you use modern letter names for your piano keys, your semitones are between E and F, and again between B and C.

I sang the tune to myself until I figured out where I thought the semitones were.  My conclusion was that the semitone is between the notes that form the second to last and then last notes of the first phrase.  If you’re not intending for your piece of music to be played on a particular instrument at a particular pitch, but instead you just want to record where the notes of your piece are in relation to each other, it’s easiest to place your tune on the staff somewhere that doesn’t require any flats or sharps and, if you can manage it, somewhere that doesn’t require any ledger lines.  Medieval and renaissance folks got around the problem of ledger lines by being good at reading many different clefs (known as C clefs).  These notate where Middle C is on your staff and then you just switch them to a different line when your tune is threatening to go too far above or below the actual five lines of the staff.  I am only proficient at reading treble clef (which shows where G above Middle C is) and bass clef (which shows where F below Middle C is) so that constrained my options.  I decided to put my semitone between B and C (above Middle C) on the treble clef, which actually turned out so that I didn’t need any ledger lines or accidentals (sharps or flats).

Once I decided where to start writing my tune on the staff, I then went through, humming it as I went, and put in the pitch or pitches that I thought went with each word in the first verse.

Note Values

In the sixteenth century, people were talking about notes as being “lunga” (long) or “brevis” (short).  British English speaking musicians I *think* still use terms like “semibreve”, “minim”, “crotchet”, and “quaver” for note values.  Notice that the “half-long” note here is the longest normal note value used.  American and Canadian English speaking musicians use the much more straightforward “whole note”, “half note”, “quarter note”, “eighth note” etc., which is a lot more informative about what duration value the notes have compared to each other.  Most people think of one quarter note as equalling one beat.  It’s actually more complicated than that, and for my purposes here today it’s more important for you to think of one quarter note as having the same duration as two eighth notes.

My purpose in this project was to determine what kind of exotic meter the Sovay tune was in.  Most Western art music, even in the 20th century, is written to have two, three, or four beats in a bar.  Occasionally very adventurous composers (Holst “Mars”), jazz musicians (Dave Brubeck “Take Five”) and writers of dramatic film scores (Howard Shore “Isengard” leitmotifs) reach out of Western music’s comfort zone and write stuff that has five beats in a bar.  One of my favorite twentieth century composers, Bela Bartok, wrote quite a lot of piano music with unusual rhythms (which he called “Bulgarian Rhythm” even though there are several variants) or with frequent switches in meter.  Bartok almost always chooses to write Bulgarian rhythms so that one eighth note equals one beat, and that was the model I followed here.

So I sang along with the tune again, pointing to each pitch I’d already notated and figuring out whether it took one or two beats.  It turned out there were no notes that took more than two beats.  Then I put little ticks or lines above each note that I thought had an emphasized beat.  If you sing a song, you’ll notice that you don’t put the same amount of emphasis on each beat, rather there will be one or sometimes two beats in the pattern that are more emphasized than others, and it’s usually very regular throughout the whole song.  One of the things that’s unusual about Sovay is that it seems not to have an entirely regular pattern.  I then counted how many beats there appeared to be in between all the ticks, and as you can see at the top half of my page I got a bunch of numbers that are not all the same and don’t appear to repeat in a regular pattern: four, ONE two, ONE two, ONE two, ONE two three, ONE rest three, ONE two, ONE two, ONE two, ONE two three, ONE two three….

I was then ready to write out a new copy of the tune with bar lines and with meter changes notated everywhere they seemed to be called for.  As I went along, much to my disappointment, I discovered it was actually possible to fit the tune in quite easily to a rhythm that has two beats in a bar, every single bar.  Sure, there were a couple of places where the emphasis didn’t seem to land on the ONE of the bar, but this is a legit compositional technique called syncopation, and there weren’t any places where I had either an extra beat or a missing beat that I needed to deal with somehow.

Then I did something else for a while that evening, but I was still disappointed that the Sovay tune I knew hadn’t turned out to be something interesting, so I decided to notate it again using Bartok’s rapid meter-switching notations anyways.  We will truly know if I succeeded only if we can find someone who reads music but has never heard the Sovay tune to sing it back to me based on what I wrote to determine if it sounds the way I think it’s supposed to sound.


But Is That Really It?!

I was really convinced that despite the meter disappointment there was something different about the tune.  Well, there is.  Most music from the mid seventeenth century on is very structurally predictable.  Every phrase is made up of four bars, and most tunes are made up of some number of phrases that is a multiple of two, four, or eight.  Think of the Mozart theme and variations on “Ah! Vous dirais-je, Maman” (more likely known to you as “Twinkle, Twinkle, Little Star” or the Alphabet Song):

Twinkle, twinkle, little star,
1 2 3 4
How I wonder what you are.
1 2 3 4
Up a- -bove the world so high,
1 2 3 4
Like a diamond in the sky,
1 2 3 4
Twinkle, twinkle, little star,
1 2 3 4
How I wonder what you are!

Now let’s look at the 2/8 version of Sovay I notated:

So – -vay, So- -vay, all on a day – – She
0.5 1 2 3 4 5 6
dressed her- -self in men’s ar-ray- – With a
1 2 3 4 5 6
sword and a pis- -tol all by her side To
1 2 3 4 5 6
meet her true love, to meet her true love, a- -way did ride
1 2 3 4 5 6 7 8

So, there is something a little unusual about the Sovay tune, but it’s not the meter, which can be notated as a perfectly reasonable two-in-a-bar.  It’s that the phrases within each verse are not all the same length as each other, and most of them are also not the same length as is customary.

Stuff I Didn’t Talk About

Oh my god, SO much stuff.

  1.  Whether A&S and Bardic are/deserve to be two separate things that are judged on two separate standards (all of the YES)
  2. Whether demonstrably post-period material nonetheless belongs at an SCA bardic circle (I change my opinion on this as often as I change my underwear)
  3. Comparison of oral/folk music tradition with literate/art music tradition (SO interesting) and whether one is somehow “better” than the other (hahaha NO)
  4. More about medieval music notation and how their solfege system works (hey maybe if I work on explaining it to others I will finally figure it out myself)
  5. Seriously, I could talk your ear off about music, but I want you to remain friends with me.

Works Referenced


The Ballad Index

Master Thorvald Grimsson

Thomas Morley “A Plaine and Easie Introduction to Practicall Musicke” (1597)

Bela Bartok “For Children” and “Mikrokosmos”

Richard Taruskin “Oxford History of Western Music”

Arlo Guthrie “Alice’s Restaurant”

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s