The rhythm of Happy Birthday, or The Star Spangled Banner

[lettersquash]

I just noticed something today. I was explaining what I meant by "sub-beats" within beats to a friend, and to that end discussed compound time (or meter) versus simple time/meter. I used various examples where you can count the main beats, but also shorter sub-beats within each beat. In simple time, note values are divided in half, repeatedly, and compound time allows for three sub-beats. I was delving into very long-term memory from a bit of formal music education about 50 years ago.

But in the meantime I've been a musician without studying, just working by ear, and I have always counted beats and "sub-beats" (I don't know if there's a correct name for them) to analyse pieces.

So, I described how I can count three beats to Happy Birthday to You, but I also hear three sub-beats going on, which align with the "happy" (two sub-beats on "ha-" and one on "ppy"). The way I hear this - or can think of it quite naturally - is with this pulse going on subliminally all the way through the piece. You can sing the song and emphasize it: Ha-a-py  bi-i-irth  da-ay-ee  to-oo-oo you-oo-oo -oo-oo-oo,... and the notes of "happy" fit perfectly.

I noticed that the same is true of the Star Spangled Banner. Each beat (it's again in 3/4) can be heard as a sort of unused triplet. Like Happy Birthday, it begins with an anacrusis: "O-oh / Say can you / see ...", and that "o-oh" sounds distinctly like two "sub-beats" to one to me.

When I googled about compound time, I realised I was basically right about that, but when I found scores of these two songs I saw they weren't notated as I heard them, and (from the definition of compound time) I realised that I would have to notate them as 9/8 to get the 2:1 ratios. Instead, they're notated as simple time, with a dotted eighth for "ha-" and a sixteenth for "ppy". This is something of a revelation to me.

In practice, they probably don't sound terribly different (I've listened to the scores in MuseScore, but as none of them have my imagined timing I can't compare in an unbiased machine output...until I write it that way...). On the other hand, I still feel that the triple sub-beats are kind of implied, the natural way to think of these pieces, and I can't help but think this notation is just an approximation because it's easier to write, and to read. If I deliberately count four sub-beats (which takes a bit of concentration and slowing it down), it feels wrong to divide those as 3:1.

So, my question is, do you know what I'm talking about - do you also hear those triples inside each beat? Do you agree it's a kind of approximation? (Incidentally, there are simpler versions of Happy Birthday that use an eighth note for each of those "ha" and "ppy" - good for wishing a robot happy birthday - humans like a bit more swing on the big day).

This raises another question for me. The other thing this suggests is triplets, so that the simple time can accommodate the three sub-beats where these apply, but the triplets I've seen involve every note being played, which wouldn't work here. So, do you ever come across a triplet split 2:1, with a dotted note and a disconnected beam?

[anacrusis]

So, do you ever come across a triplet split 2:1, with a dotted note and a disconnected beam?

If I understood you correctly, I would argue most pieces in, for example, 6/8 are like this. Check out the opening to Chopin's 2nd ballade, for example.

[lettersquash]

If I understood you correctly, I would argue most pieces in, for example, 6/8 are like this. Check out the opening to Chopin's 2nd ballade, for example.

No, that's my point, these aren't triplets, as far as I'm aware. The 6/8 time allows two pulses or main beats, each with three eighth notes. That is compound time, providing three equal note values, where simple time only allows division by two. So, in the Chopin, the 2:1 ratio of note values can be scored with a quarter note and an eighth note. However, in simple time, one can force a note value to be split in three by using a triplet - scored using the normal time values but not played as such. So a quarter note's time might be given a triplet of three eighth notes with a beam, and a little "3" above. We know not to play them as eighths despite the use of those icons (because there aren't any others in the system for thirds of other values).

These are, however, usually all played - each of the three notes, and I don't know if they're ever split into two, or just one of them played, etc., or how that might be notated. If that's not done, I guess a composer would have to switch to a compound time (6/8, 9/8, 12/8...) or use the approximation I talked about above, divide by 2, or 4, and then use the normal time values.

[brogers70]

These are, however, usually all played - each of the three notes, and I don't know if they're ever split into two, or just one of them played, etc., or how that might be notated. If that's not done, I guess a composer would have to switch to a compound time (6/8, 9/8, 12/8...) or use the approximation I talked about above, divide by 2, or 4, and then use the normal time values.

Sure, you can do that. So in 4/4 time you might have a quarter note followed by an eighth note, and there would be a bracket over them with a "3" above the bracket. That would indicate that you were to divide that beat into three thirds, with two of the thirds taken up by the quarter note and one of the thirds by the half note. Hope that's somewhat clear.

[anacrusis]

No, that's my point, these aren't triplets, as far as I'm aware. The 6/8 time allows two pulses or main beats, each with three eighth notes. That is compound time, providing three equal note values, where simple time only allows division by two. So, in the Chopin, the 2:1 ratio of note values can be scored with a quarter note and an eighth note. However, in simple time, one can force a note value to be split in three by using a triplet - scored using the normal time values but not played as such. So a quarter note's time might be given a triplet of three eighth notes with a beam, and a little "3" above. We know not to play them as eighths despite the use of those icons (because there aren't any others in the system for thirds of other values).

These are, however, usually all played - each of the three notes, and I don't know if they're ever split into two, or just one of them played, etc., or how that might be notated. If that's not done, I guess a composer would have to switch to a compound time (6/8, 9/8, 12/8...) or use the approximation I talked about above, divide by 2, or 4, and then use the normal time values.

Ok gotcha! As brogers said, you sometimes do see a quarter note followed by an eighth note (or the other way around), with a bracket and a 3 over them to indicate that they are played in the same time as a triplet.

It's not super common, simply because if a composer knows they're going to write a piece where they often need to divide the beat in three, they pick 3/8, 6/8, 9/8 or 12/8 and notate the piece that way. That's what those time signatures are for!

[lettersquash]

Sure, you can do that. So in 4/4 time you might have a quarter note followed by an eighth note, and there would be a bracket over them with a "3" above the bracket. That would indicate that you were to divide that beat into three thirds, with two of the thirds taken up by the quarter note and one of the thirds by the half note. Hope that's somewhat clear.

Oh, right - thanks (also anacrusis for your confirmation of that).

Any thoughts on my main point - assuming I've explained it well enough? Another way to think about it is, if you were going to play Happy Birthday to You with accompanying arpeggios in the left hand for some variety, making every beat a triplet (assuming it's in 3/4) would be perfectly natural (in fact, I saw at least one version where that's done), whereas a pattern of four notes would be awkward and sound machine-like. So, to my mind, those triplets are where the "Happy" should land, not a dotted eighth and a sixteenth. Maybe the whole question is pedantic, and the difference isn't actually that noticable, or it's understood that you play it rather more like a triplet. I'm just a bit taken aback that many pieces probably aren't notated how I think of them, because I've not really thought about how to notate them before. I've just counted these "sub-beats" as a matter of habit in all sorts of pieces. No doubt most of them are compound time - like an Irish jig will most likely be in 12/8 - and I happened to alight on these two examples that surprised me.

[brogers70]

Oh, right - thanks (also anacrusis for your confirmation of that).

Any thoughts on my main point - assuming I've explained it well enough? Another way to think about it is, if you were going to play Happy Birthday to You with accompanying arpeggios in the left hand for some variety, making every beat a triplet (assuming it's in 3/4) would be perfectly natural (in fact, I saw at least one version where that's done), whereas a pattern of four notes would be awkward and sound machine-like. So, to my mind, those triplets are where the "Happy" should land, not a dotted eighth and a sixteenth. Maybe the whole question is pedantic, and the difference isn't actually that noticable, or it's understood that you play it rather more like a triplet. I'm just a bit taken aback that many pieces probably aren't notated how I think of them, because I've not really thought about how to notate them before. I've just counted these "sub-beats" as a matter of habit in all sorts of pieces. No doubt most of them are compound time - like an Irish jig will most likely be in 12/8 - and I happened to alight on these two examples that surprised me.

I may not quite be understanding your question, but....If you have an accompaniment in triplets and you have dotted notes in the melody there are two ways to handle it. During the Baroque period (and according to some, even as late as Schubert) the common practice was to fit the dotted notes to the triplets, so that if you had a triplet of eighths in the accompaniment and a dotted eighth and sixteenth in the melody, you'd play the melody so that the sixteenth lined up with the third triplet.  That's because originally the dot just meant to lengthen the note a bit, not necessarily by exactly 50% of its value. So a Baroque dot could mean anything from a quarter and eighth under a "3" brace, to a standard dot, to a double dot, depending on the context and style of the piece. Nowadays, if you see such a thing, a triplet in the accompaniment against a dotted eighth and sixteenth in the melody, you are expected to play the 16th after the final triplet eighth, because we modern classical folks interpret a dot to mean a lengthening of the original note by exactly 50% of its value. But that was not always the case. So really you have to know how finnicky was the person who notated your version of Happy Birthday. ANd I guess it may be different in jazz.

[lettersquash]

I may not quite be understanding your question, but....If you have an accompaniment in triplets and you have dotted notes in the melody there are two ways to handle it. During the Baroque period (and according to some, even as late as Schubert) the common practice was to fit the dotted notes to the triplets, so that if you had a triplet of eighths in the accompaniment and a dotted eighth and sixteenth in the melody, you'd play the melody so that the sixteenth lined up with the third triplet.  That's because originally the dot just meant to lengthen the note a bit, not necessarily by exactly 50% of its value. So a Baroque dot could mean anything from a quarter and eighth under a "3" brace, to a standard dot, to a double dot, depending on the context and style of the piece. Nowadays, if you see such a thing, a triplet in the accompaniment against a dotted eighth and sixteenth in the melody, you are expected to play the 16th after the final triplet eighth, because we modern classical folks interpret a dot to mean a lengthening of the original note by exactly 50% of its value. But that was not always the case. So really you have to know how finnicky was the person who notated your version of Happy Birthday. ANd I guess it may be different in jazz.

That's interesting, and does suggest it might be the main reason. I'm reading the scores of these from the modern perspective, but they've probably got a longer history or at least (in the case of the Star Spangled Banner) draw on that older, looser tradition. Thanks.

[palmtree]

Nowadays, if you see such a thing, a triplet in the accompaniment against a dotted eighth and sixteenth in the melody, you are expected to play the 16th after the final triplet eighth, because we modern classical folks interpret a dot to mean a lengthening of the original note by exactly 50% of its value. But that was not always the case.

One interesting tidbit on this point is that different countries would have different customs notation. The most well known example of the treatment of dots being France, where (during the Baroque) dots were often treated as double dots, and no dots at all were often played as with single dots (just a fun music history sidenote).