xkcd #2942: Fluid Speech
Alt text:
Thank you to linguist Gretchen McCulloch for teaching me about phonetic assimilation, and for teaching me that if you stand around in public reading texts from a linguist and murmuring example phrases to yourself, people will eventually ask if you’re okay.
If you’re a native-level speaker, no you don’t. You think you do. Assimilation is a real thing and is a huge part of all native language. NOBODY pronounces the way they think (and often loudly claim) that they do.
Just like the people who claim they don’t have a “j” sound in “could you”.
How do you know that no-one enunicates the t sound? I just asked my partner to say hot potato and she definitely does.
You’re skewing the results by a bad test.
Don’t “ask your partner” to say a particular word or phrase. The very act of asking that will have changed results. (This is experiment design 101 stuff here!) Ask her to read a lot of stuff that has “hot potato” in it in various places. (We tend to use sandhi in flowing streams of speech, not isolated clips.) Or, ideally, engage her in conversation and get her to say “hot potato” naturally as an organic outgrowth of the conversation.
But … make sure you record what she says. Your own brain, as a listener, fills in stuff that’s not there while removing stuff that is. You have to play it back, concentrating on only the sounds, not the words, and do it repeatedly, ideally isolating this one phoneme at a time.
Really, sandhi is a thing, and it’s a thing that literally every native speaker of every language in the world uses. There is variance by dialect, naturally (entire phones vanish or come out of nothing from dialect to dialect), but some elements of sandhi (like consonant assimilation) happen no matter what your dialect unless you’re specifically concentrating on having it not happen.
I don’t need to, i know she that correctly. There are definitely words we pronounce incorrectly but nit that one. You and the OP are conflating your local experience for a global one. I don’t live in the US, we enunciate differenrly
Well, the only way to check beyond me muttering at myself would be to have a recording of me talking casually about hot potatoes :D
And yeah, I definitely pronounce “could you” as “couja” when relaxed. Hanging out with people from different countries makes you pretty conscious about your accent some times. Mostly when half the voice chat can’t understand what you just said and the other half can’t understand why they’re having an issue.
Sandhi is a real thing. (Source: I had to study this shit to teach pronunciation classes.)
It took me WEEKS to recognize that what I thought I was saying and what noises I was actually making are completely and utterly different. There’s often no relationship (like “coodja” for “could you” or “chrain” for “train”) between the intended sound and the actual sound … but since everybody does it you don’t notice until its forced into your face. The only time you make distinct sounds as per the “official” description (and even then not as often as you think: I submit “train” once again as evidence) is when you’re deliberately speaking slowly and distinctly. Which is almost never (and comes across as condescending in actual interaction).
Weeks, I say again. WEEKS. And this was under constant training that included the playback of what we’d actually said showing us what we were doing. The denial is embedded deeply in our psyche.
I’d say part of this is the intended / official descriptipn isn’t actually that. The spoken word existed first, then someone tried to capture that spoken word using a finite list of characters and character combinations that map back to phenomes. The written word isn’t phonetically accurate to the letters it is composed of, and the written word is just close approximation of the spoken word itself.
My six year old daughter is getting the hanging of the spelling and whatnot, but earlier on in her Kindergarten year, words like “driver,” to her, started with a J. I had never thought about it, but it absolutely (at least in our NJ dialect) has a J sound, because, as you say, we all talk fucked up (paraphrasing).
Sandhi is an amazing area of study. It’s doubly fun in tonal languages: all the confusion of atonal languages with an added layer of shitfuckery.