Hacker News Reader

Comment by PaulDavisThe1st | Hacker News Reader

The t-glottalization mentioned in the article is now picking up speed in the USA too, most notably (I think) among female speakers. I'm still shocked (as someone who grew up with a cockney-leaning accent) to hear presenters on NPR do t-glottalization, and on the occasions when I'm around millenials, especially millenial women, it's an every-fifth-sentence phenomenon. Given its role as a class status/marker in the UK during my youth, it feels very surprising, but also a great living reminder of how language changes in real time.

View on HN

Glottal stops for t's are not new. A lot of t's in American English are glottalized. For example

1. button

2. gotten

3. mountain

4. eaten

5. wait

In all of these examples the t is dropped entirely in all but the most formal speech. Unless you mean something different this is not a new phenomenon and is actually a characteristic marker of American English. In fact everyone I know, even people 30-40 years older than me, drop their t's everywhere. It actually feels weird to me to pronounce them in the above example and it sounds like I am scolding someone. I don't really know the linguistic history of it and my family has only been here a few generations, but everyone back to my grandparents spoke this way. Perhaps it wasn't as common before the 30's though.

As for news and what-not I'm so used to it I didn't even notice any change in inflection. NPR tends to be very hip these days (unfortunately) and so I am not surprised they are speaking more casually than a normal newcaster.

Glottalized is not the same sound as dropped or flapped. The American English dropped t's in these examples are either completely dropped, or pronounced as a (weak) flap. If not completely dropped, the t in "button" would usually be a flap, almost the same sound as the r in Spanish "cara". The final t in "wait" would most likely be an unreleased stop, if pronounced at all. Both regular and flapped t are articulated towards the front of the mouth, whereas glottalization happens the very back.

See https://en.wikipedia.org/wiki/Flapping and https://en.wikipedia.org/wiki/T-glottalization

While the parent is incorrect about wait, he/she is correct about T-glottallization in the word button. The glottalization of the T in button is even listed in the wikipedia article you linked.

Yes, but even though T-glottalization in button is happening in the US now, that was not true historically. It is changing from budd'n to 'buh-un'

It seems I messed up the definition of glottalized but unless I am deliberate I also barely pronounce the t in wait.

in typical north american english, in #1 #2 and #4 the t's are pronounced /d/ and the final syllable shortened (budd'n, godd'n, ead'n) and in #3 they are omitted but not glottalized (moun'n). in #5, the final t is simply not aspirated, but rarely glottalized.

historically, linguists have noted that language changes tend to arise among young women. They are more flexible due to age and more socialized than their male peers.

The studies show that women tend to copy high status features ("prestige"), and men tend to copy low status features ("covert prestige"). That is to say, women want to sound like they're class above where they really are (richer, fancier, etc), while men want to sound like they're a class below (rougher, manlier, etc). In fact, upper middle class women have in certain studies been measured to more consistently use prestige features than actual upper class people.

This sounds like a very categorical claim. It seems hard to measure, quantize and produce rigorous and replicable results.

What seems so hard? It sounds pretty trivial to me -- all you're doing is recording samples of people's speech and objectively identifying pronunciation patterns. And it's pretty easy to do before/after with the same people at differing ages (e.g. 15 vs 20) or life stages (e.g. pre-university, university, post-university) to watch change over time.

Time-consuming for sure. But not even remotely difficult in terms of rigor or replicability. And if you're familiar with the International Phonetic Alphabet, you'll see that pronunciation is something that can be analyzed extremely objectively.

>This sounds like a very categorical claim.

The claim is explicitly non-categorical, as the usage of the word "tend" shows.

>It seems hard to measure, quantize and produce rigorous and replicable results.

Not really.

First you split your informers into groups, based on sex, social class and/or education level. Then you pick a linguistic variation that you know to be stigmatised, but still present in that group regardless. Then you analyse the frequency that that feature appears, in comparison with another variation for the same feature.

If I recall correctly one of the first to notice those gender discrepancies in speech was Labov, so you might have some luck if you websearch "gender paradox Labov".

I took a linguistics class years ago that talked about this!

The hypothesis was that upper class women pronounced the "r" as in "Fourth", but lower class women "rolled the r" so "Fourth" became "fowth".

The researchers entered department stores with high and low end customers and asked for a product that was "on the Fourth floor", then they counted the number of times the "r" was pronounced or "rolled"

I know which study you're talking about. I don't recall if Labov addressed gender differences in that specific study but, if he did: in that situation, lower class men would be even less likely to use the rhotic (with /ɹ/) pronunciation than lower class women, and prefer the non-rhotic (without /ɹ/) one instead.

Although I am British, I moved away to another country when young, and got some of my English lessons privately, plus my family is quite posh, or at least, likes to talk that way. I moved back to the UK and on joining primary school there, my received accent was the first thing to go, much to my parents dismay, as it invites teasing and bullying.

I can still talk in a reasonably posh way, but almost never use it, and adopt all sorts of mannerisms from my travels, my accent is, at a guess, a little bit estuary english, with other things thrown in.

I've never really felt the need to up my accent, it doesn't make you the most popular person professionally to act better than everyone else, especially if you don't have vastly superior skills or superior standing to back it up, in fact it annoys me if someone tries it on me. I wouldn't say I act lower class than my standing, rather, I tend to fit in with my peers.

I’ve noticed this shift too in my decade+ in the US. It’s actually really nice to learn someone else noticed it too, because no one I’ve mentioned it to has known what the hell I'm talking about.

I think I only heard in it in some African American English accents when I first came here, but now it’s common among younger Americans more generally. I’ve also heard it applied to other consonants - you can replace quite a few of them with a glottal stop and still be understood :D

Is this why i find that british accents have worsened in past decade? I can watch a speech from Thatcher pretty well, she speaks quite clearly . But in recent movies and TV series i almost always turn subtitles on

I think they have always been that hard to understand, it's just that TV series started adapting the more vernacular registers. Also, you will probably still be able to understand a speech from Liz Truss.

Ugh that's incredibly annoying if true in the US. It's almost definitely people trying to sound "fancier"/higher class, as the British people/accents are stereotyped as such here.

Haven't noticed it yet (New Yorker), but will surely keep my distance from/cringe at such silly people.

I think it was appropriated, if you’ll pardon the pun, from African American accents, but I’m only an amateur linguist noticing things and making connections, so in the cold light of science, my “think” is no better than your “almost definitely” :-)

Also an amateur linguist, and you could be right, the pattern is similar to AAVE. Somewhat related, I was recently listening to a ton of YouTube videos that go through the differences in pronunciation of German spoken by those in Bavaria and those elsewhere, and the dropping of certain consonant sounds in “Bavarian German” seems familiar.

Note: not fluent yet, please if you speak German correct me if my depiction is not accurate. “Bavarian German” does seem to drop a lot of the condiment sounds, and also seems to drift closer to Latin in some ways, but I haven’t had time to actually look into that.

Reply on news.ycombinator.com