THR-Flapping and Beyond

Stop! Say the word "through" out loud. What did it sound like? More specifically, what did the 'r' in "through" sound like? There's a possibility that you used what's called a 'flap' for that sound: the [ɾ] sound. This sound is very common in American English, but usually between vowels, as an alternate version (or allophone) of /t/ or /d/: "better," "ladder." It's not a common allophone for the /r/ sound in modern American or English English.

This pronunciation is nevertheless pretty common. However, we don't know how common it is. One of the first papers to deal with the subject specifically is by Joey Stanley: (thr)-Flapping in American English: Social factors and articulatory motivations. Stanley's paper focuses on Utah English, where he found it to be more common among Utahns than Washingtonians.

This feature, which he calls THR-flapping, can be found outside these states. I've collected a few examples of it, the earliest dating from 1911.

• "We'll go th[ɾ]ough the ceiling" - Come Josephine in my Flying Machine, Ada Jones (Lancashire, Philadelphia) 1911
• "Looking th[ɾ]ough your eyes" - I Write the Songs, Barry Manilow (New York City)
• "It's slipping th[ɾ]ough your fingers like sand" - Gold, Marina and the Diamonds (Wales)
• "But in your letter, you said we were th[ɾ]ough" - Lefty Frizzell (Texas, Arkansas)
• "Th[ɾ]ree" - every day struggle, Notorious BIG (Brooklyn, NYC)

You'll notice that most of these are in the word 'through,' with two examples of 'three'. This isn't unusual, as these two words were also overrepresented in Stanley's study:

"In this case, there was also a massive imbalance in what words were represented: 54% of the (thr) tokens in conversation were three, another 32% were forms of through, and the remaining 14% of the tokens came from 13 other lemmas."

THR-flapping can even be found in books, as a 1960s Norwegian textbook for English speakers I own explains that English speakers can make the 'r' sound by looking to the sound in 'through.'

Are there other consonants that let you flap the /r/ sound? I've uncovered only one other example of this, from a century ago, with /f/. Both /f/ and th are fricatives made in the front part of the mouth, but FR-flapping seems to be extremely rare besides this isolated instance. Note that Billy Murray uses an approximant /r/ in other situations.

• "When you're flush, your f[ɾ]iends are sunny" - A Good Old Dollar Bill, Billy Murray (1909)

It's not clear what motivates THR-flapping. Could it be retention of the Middle English realization of /r/ as [ɾ], preserved only in /thr/? Or is it just easier to flap the /r/ when you're already in a dental (or interdental) position than have to retract the tongue and form an approximant?

In any case, THR-flapping is a tremendously common part of English across dialects, and deserves a little more attention. I'd also be interested in seeing if there are more examples of FR-flapping, as in the Billy Murray example above. Is he a one-off or was FR-flapping more common in the past?

Random Updates

First off, an example of the COT-CAUGHT merger. "Spaw day":

Secondly, for those of you following the history of people saying things like 'happay' instead of 'happy' in pop music, I found an interesting early example. From Boney M, their 1976 cover of "Sunny."

"Sunnih, sunnay"

What does this mean for the theory? It's unclear - the lead singer, Liz Mitchell, was born in Jamaica and then moved to London at 11 years of age. Young enough for her to pick up influence from other Londoners, and that's where I posit the "-y" breaking originated. In any case, this is one of the clearest examples from the 70s I've heard, so enjoy, and have a sunny day.

Finally, a fun bit of etymology for you: 'cloud' and 'clod' share the same origin. I was alerted to this bit of trivia by the book "Clouds of Old Naples", and had it confirmed by Etymonline. What word did English speakers use for 'cloud' before comparing it to a rock? Welkin!

How Singing Resonance Affects Vowels

I've recently been delving deeper into the human voice and the actual anatomy behind it. Most of the audio I look at is sung speech, so I've been wanting to further categorize aspects of sung speech that may be relevant. Vocal science and voice descriptions are quite opaque, and descriptions of voices can vary wildly from one source to another. A big part of this is also because it is still not entirely clear how the vocal mechanism functions, and what anatomical configuration results in certain vocal colors.

Even with this lack of consensus, one can find useful ideas. One recent idea I've found that explains a fair amount is that there are two types of resonance that singers naturally find when singing. One is based around the 'oo' sound: image a loud 'woo-hoo' call. The other is based around the 'eh' sound: imagine calling out 'hey'.

The 'eh' type of resonance results in a voice that sounds brassy and even shrill. It is prominent in pop music, which I cover a lot. I suspect that the 'eh' type of resonance is also easily found with the 'oh' sound and the 'uh' sound: sounds that are mid-way between the top and the bottom of the mouth.

When I use 'eh' resonance, I can feel a vibration above the back of my tongue. Achieving this vibration is very easy if I use mid-height vowels like [e], [o], and [@]. The farther away you go from these vowels, the harder it is, but lower is even harder. [I] is harder than [e], but not by much, while [ɛ] requires some effort to get that brassy resonance. [a] is very hard to achieve, and requires a weird tongue posture where the front is low in the jaw while the back is raised.

I was never taught how to achieve this resonance; I simply noticed that when I sang on [e] and [o] vowels, my voice seemed to carry farther without having to use more air. From there, I tried modifying vowels to get them closer to mid-height where I could get away with it:

• [ɛ] got shifted up to [e]
• [ɪ] got lowered to [e] as well.
• [ʌ] could sound nigh-on Southern as [ɜ]
• [u] didn't decrease in height, but I did push it forward to [ʉ], which let me keep the back of the tongue high.

What is the impact of this? Well, it means that when analyzing a piece of sung text, we have to keep these vowel modifications in mind. Here's an example from a past Dialect Dissection: Britney Spears raising the vowel in BED:

"You used to say that I was sp[e]cial" - What You See Is What You Get

In the Britney Spears article, I listed this as a possible example of Southern influence. And it could be! But it could also be an example of a Pop Vowel Modification (time to name things :)). Considering that Britney is actually from the South and has a Southern accent in real life, I would give points to the "is using her Southern accent" theory. She's just also using features that work well within the aesthetic constraints of pop music.

This also plays a role in the popularization of ME-breaking and HAPPY-breaking (and to a lesser extent, HAPPY-laxing): we move from vowels that are harder to get 'eh' type resonance on ([i] and [ɪ]) to vowels that are easier to get resonance on ([i] moving to [ɪ], and [ɪ] moving to [ej]). This feature began as part of actual, real life accents: Southern American English (white and Black) and working class London English. It just so happens that they fit the Pop Vowel Modification.

I believe that music can be a great vehicle for understanding language and linguistic change. But in order to make use of it, we need to understand how sung speech differentiates from spoken speech, and how the pressures on sung speech are different from the pressures on spoken speech. To that end, I'll continue studying this area and writing about it.