Deck the halls / with boughs of holly / fa la la la la / la la la la
Na na na na / na na na na / hey hey hey / goodbye
Wop bada lumop / a wam bam boom
Hickory dickory dock
High ho, high ho
What do all of these songs have in common?
They’re all examples of musical nonsense syllables, the technical term for which is non-lexical vocables. Non-lexical, because they don’t have any sort of meaning that you could find in a dictionary or lexicon, and vocables, because they’re things you vocalize. (Lexical vocables, on the other hand, are more commonly known as “words,” although sometimes people call the non-lexical type just “vocables.”)
It’s tempting to ask where non-lexical vocables come from, but unfortunately that’s probably about the same as asking where language or music comes from. In fact, like children who babbles “mamamamama” as they’re learning to speak, it’s quite likely that we made meaningless, rhythmic sounds before we even had specific meanings associated with them, although we can’t know for sure. And if you’ve ever struggled to write a rhyme or to remember the lyrics to a song, you can understand the appeal in singing using low-pressure nonsense syllables, which probably explains why they stuck around even when we started associating words with meanings.
Many languages have nonsense syllables
Non-lexical vocables are found in music from a wide variety of cultures and time periods, including Scottish mouth music, Saami joik, Appalachian eefing, Blackfoot chants and other Native American music, Inuit throat-singing, Jewish nigun, Mbenga-Mbuti music, Maldivian music, and others. In English, we have non-lexical vocables from Shakespeare, such as the song “Hey Nonny Nonny” in Much Ado About Nothing, all the way to scat, do-wop, and improvisational jazz singers such as Ella Fitzgerald and Bobby McFerrin, and even to modern songs like “What Does the Fox Say?”
Nonsense syllables do show patterns
Despite the fact that non-lexical vocables don’t have a particular meaning in themselves, there are a few general patterns that we can notice about them.
The first pattern is that despite their lack of meaning, the choice of sounds to combine into nonsense syllables is somewhat predictable. One of them is repetition: we often get strings of the same one or two syllables, such as fa la la la la or na na na na hey hey, rather than a series of different ones such as fa lo nee boo hey, which is an example I just made up. Even the more complicated strings of syllables tend to stick to a small subset of sounds: notice the repetition of “b” and “op” and “w” in bop bop shoo wada wop or the rhyming in hickory dickory dock.
The second pattern that we can notice is that nonsense syllables tend to be uncomplicated, most commonly a consonant plus a vowel as in fa, la, na, ho, hey, high and so on. Even when they’re a bit more intricate, such as shoo-wop shoo-wadada-wop or boom chicka wow wow, each individual syllable doesn’t normally get more complicated than consonant+vowel+consonant or CVC. (Of course, this is assuming that you look at the sounds, not the letters: “oo” and “sh” may be spelled with two letters, but they each make a single sound.)
You wouldn’t necessarily expect that nonsense syllables would be so simple: after all, English is fine with consonant clusters as in tree or hands or even the longest monosyllabic English word strengths. But we sing fa la la la la instead of fra spla spla spla spla. Why might this be?
Nonsense syllables and vowels are easier to sing
Well, we can think about what the purpose of non-lexical vocables is: to give us something easy to sing. Which one is easier to hold a note on, la or strengths? Try it. It’s pretty clearly la, and there’s a good reason for this. Vowels as a category are more sonorous than consonants: even if you’re tone-deaf, you can hold some sort of note on aaaaaaa and eeeeeee. Some consonants are fairly sonorous too: you can probably also manage to hold a note on mmmmm and llllllll. However, many consonants aren’t sonorous at all: you can’t hold a note on ppppppp or sssshhhhh without adding a vowel in there somehow.
So the more sonorous something is, the easier it is to sing, which means that vowels are great, but consonants should be used only in strict moderation. If you’re writing a song and you actually want it to mean something, you might be stuck using some words with less-sonorous consonant clusters because you like their meanings. But if you’re just going to sing meaningless sounds, why not pick ones that are really easy to sing?
In fact, another great reason to sing meaningless, easily-articulated sounds is if you’re just learning how sing in the first place, so you don’t have the difficulty of remembering lyrics, enunciating difficult consonant clusters, and carrying a tune all at the same time. So nonsense syllables like do re mi* are used for teaching singing, in a system called solfège. And even though the song from The Sound of Music might suggest that they stand for something, a sequence like mi re do is definitely nonsense: it’s not really trying to assert anything about female deer or drops of golden sun.
Here’s a bonus question: what would happen if you go to the other extreme and use nonsense sounds composed primarily of consonants? Actually, this is possible too, but it sounds more like beatboxing than fa la la la la. Don’t believe me? If you put the right string of consonants into Google Translate and then click on the little speaker icon, you can hear it beatbox to you. Here’s a demo, and here’s a guide to composing your own rhythms.
* Musicians have pointed out that do, re, and mi aren't nonsense because they are the names of notes. We see their point—do, re and mi are in the dictionary, so they are lexical vocables. The important point for this article is that do, re, and mi fit the pattern of not having consonant clusters.