February 28, 2018

The Logic of Mondegreens

Anyone who has ever listened to sung music has probably run into the situation where they either mishear the lyrics or don't understand them at all. Most people don't know that there is a technical name for misheard lyrics - they're called mondegreens, from a story of someone who misheard the lyric "and laid him on the green" as "and Lady Mondegreen."

Some researchers, like Steven Pinker (1994), have noted that many misheard lyrics tend to be pretty weird. One of my favorites is mishearing Kesha's "Cannibal" as "Cat nipple." Perhaps it is only the weirder mondegreens that we remember, since mundane mondegreens where one word is off aren't interesting. A recent example comes from "I Write Sins, Not Tragedies." "I chime in with a 'haven't you people ever heard of closing a goddamn door'" is the lyric in the lyrics booklet, but many people - even the singer, Brendan Urie - sing 'closing the goddamn door.' (Worth noting that many lyrics booklets have incorrect lyrics, but let's assume for the sake of argument this one is correct). Both are grammatical, both make sense, and the difference is literally a matter of changing an indefinite article for a definite article. This isn't going to get mythologized like "the cross I'd bear" turning into "the cross-eyed bear."

Humorous mondegreens are so popular that there was a brief trend of making misheard lyrics videos around 2007. The first one here is from Nightwish, a band where the singer speaks English as a second language and also uses operatic technique, which further distorts understanding. The second is from Evanescence, whose singer is a native English speaker who doesn't articulate clearly. Some of these mondegreens are clearly a result of someone trying to find a misheard lyric for the sake of filling in the video, but many of them are very plausible.

Mondegreens have many causes. Let's take a look at a couple of the causes.

Homophones

This is one of the most common causes of misheard lyrics - when two sound sequences simply sound the same. In this case, the problem is not that someone mishears the sounds themselves, but that they "decode" it incorrectly. One of the most famous examples of this is from a Jimi Hendrix song. "Excuse me while I kiss the sky" is misheard as "excuse me while I kiss this guy."

Let's look at the sequence that's giving us problem, the one that should be "the sky." We're going to show the sounds all together here, since there are no pauses when Jimi makes this sequence (and we don't use pauses between words in normal speech, either).

[ðəskaɪ]

Jimi intended us to hear word boundaries between ə and s:

/ðə skaɪ/ "The sky."

However, think about the phrase "this guy." First of all, it is more common to hear about someone kissing a person than an inanimate concept. When we hear and understand "kiss," our brains are primed to hear other words that we commonly hear with "kiss." We are more likely to hear "guy" than "sky."

"But isn't there a short 'i' in "this"? And there is a 'g' in guy, not a k!" Well, not precisely. Unstressed syllables in English have a tendency to move towards the schwa (the sound in about). In stressed position, "this" would indeed have a short i. In an unstressed position, it's not inconceivable that it would move towards a schwa, especially since 'i' is already a lax vowel.

As for the 'g' in 'guy'... what if I told you the 'g' in 'guy' and the 'k' in 'sky' are the same sound? Seriously! Try saying 'sguy.' Does it not sound just like 'sky'? In English, what differentiates the 'voiced' and 'unvoiced' members of a stop at the beginning of a syllable is not actually 'voicing' (there is some in the 'voiced' stops, but very little), but aspiration - the breath that comes after the stop. This is why Guy and Kai sound different. However, when you have a consonant cluster with an 's' in front of a 'voiceless aspirated' stop, there is no more aspiration. The 'k' in 'sky' and 'Kai' are not exactly the same sound. The hypothetical lyric "Excuse me while I kiss this Kai" is therefore actually less likely to be misheard as 'guy' because our brains hear the aspiration and excludes the sounds that it could be. The 'g' sound is never aspirated. It is therefore trivial for the brain to hear "ðəskaɪ" and interpret it as:

/ðəs kaɪ/ "This guy."

Jimi seemed to be aware of this homophony, as he played around with it in live performances, miming a kiss towards another band member.

Assimilation & Coalescence

This post was inspired by someone complaining about the Lana Del Rey song "Swan Song." The opening line goes "put your white tennis shoes on and follow me." They were complaining that she pronounced "tennis shoes" as "tenishoes" as opposed to keeping the 's' distinct. This was probably not helped by the fact that the stress fell on the second syllable of "tennis" in the song, while in spoken speech it should fall on the first syllable. This made it sound like "ten issues" as opposed to "tennis shoes." (Unfortunately I can no longer find this post, but this person is not the only one to have misheard this line.)

What this person was complaining about is assimilation. This is a process where one sound changes to sound like a similar sound near it. 's' and 'sh' are both coronal sibilant fricatives. The difference between the two sounds is one of a few millimeters - 's' is made on the alveolar ridge, and 'sh' is a post-alveolar, made slightly behind the alveolar ridge. The 's' changes to become another 'sh'. When you have two sounds that are so similar next to each other, it can actually become difficult to make the sound (articulation) and to understand the sound (audition). Assimilation is therefore a common phenomenon in languages. It is rarely an issue for native speakers who already know what the words sound like and whose brain compensates for assimilation. It is more likely to impede comprehension in people who are second-language speakers, since they are still learning to understand the ways that words change in context in the second language.

If you'd like to explain the phonetic and phonological processes that go into some particular lyric you misheard, send in your misheard lyrics! I can put them in some kind of recurring segment explaining misheard lyrics.

References

  • Steven Pinker (1994). The Language Instinct. New York: William Morrow. pp. 182–183

1 comment: