It’s not really a standard “d” there, but something called an alveolar flap. The phenomenon of turning Ts and Ds between a vowel into the same sound (for example, “metal” and “medal” are identical in my accent, unless I choose to stress the difference) is called intervocalic alveolar flapping. In my experience, it seems to be present in a majority of American accents. The reason I say this is because when somebody pronounces “little” with a clear “t” rather than an alveolar flap, it sticks out to me. This linguistics paper (PDF), says the phenomenon is a “characteristic of most varieties of North American English.” I would agree with that assessment.
This I don’t know for sure. My speech doesn’t exhibit any “t-glottilization” in the word “partner,” (although it will in words like “not” and “can’t”) but my “center” does come out as “cenner” in spontaneous speech, just like “twenty” becomes “twenny,” and I think that’s pretty common in American accents. I’m not sure if it’s present in most American accents, but at least very common.
Because in your dialect, those are not considered “correct” pronunciations. In North American English, those are not considered incorrect. Hell, most speakers probably can’t even tell or know that they’re using alveolar flaps, glottal stops, or eliding “nt” into “n,” just like many speakers here swear up and down that they don’t insert a consonant rhotic between vowels in words, when they clearly do.
Got a DVD a while back of the Beatles’ appearances on the Ed Sullivan show, and one of the things that really stood out to me was the way Sullivan emphasized the first syllable of “Beatles,” clearly pronouncing the ‘T’ and inserting a glottal stop between it and the ‘L.’ I’ve always pronounced both the singers and the insects “beadles”…
This. I think one of the biggest lessons I take from this thread is just how “sloppy” we all can be in our pronunciation and not realize we are doing it. Because I would swear I’m pronouncing a “t” in all of those situations, but actually listening to myself and trying to relax and be natural, I hear some of those features - “I’m not making a glottal stop, I’m saying a t” and then it comes out like a glottal stop. :mad:
But I keep reading that word as phenome and not knowing what that is.
But I don’t know what that means. And I try to avoid using words I don’t know.
Did you mean that in the first case, it is unpredictable? If so, then I think I see what you are saying. A “phoneme” is different when it relays different semantic context, but when it is just a variant sound for the same semantic context it is the same phomene. Correct?
Watch some of the later episodes of the 80’s Bill Cosby Show: I noticed the kids especially converting the trailing “d” in words into a hard glottal (throat) stop. “Dad” became “Da’”. I have not really heard that very much lately.
Your latter statement is perfectly correct. But if you think about it (it took me a while to get this, too!), this means that when you have two different sounds that do NOT have semantic value – that is, they represent the same phoneme – like, the “aspirated p” and “unaspirated p” in English – then you will come to understand that what I said is true:
It is perfectly PREDICTIBLE which sound English speakers will use, based on the SOUND CONTEXT. After or before certain sounds, we always use one of the sounds. After or before certain others, we always use the other sound. No semantic information is conveyed!
You can think of this in terms of basic information theory. If some variable always correlates with some other variable in the same exact way, then you don’t need both variables – you just need one of them to convey information. It’s when you introduce UNpredictibility that you start adding new information (or at least the possibility of it), as long as that unpredicibility isn’t random – and, these semantic-conveying phonemes are *not *random. We use one when we want to convey some meaning, and use the other when we want to convey some other meaning.
Perhaps a different example will help. For English speakers, an example of the reverse of the “aspirated/unaspirated p” thing is the “r - l thing”. The “R” and “L” sounds ARE different and distinct phonemes in English (never mind that the “r”, especially, can be pronounced different ways – the original point of this thread). But in Mandarin Chinese, they are different ways to pronounce the SAME phoneme (just like “aspirated p” and “unaspirated p” in English). Thus, native Chinese speakers have trouble remembering which of the two sounds to use in which situations when they are learning English. I’ve never seen a study on this, but I’ll bet they tend to use “r” in certain SOUND contexts, and “l” in certain other ones, forgetting that the two sounds DO convey semantic information in English.
True. I have never taken a class in linguistics, so correct me if I’m wrong, but sometimes allophones can vary “freely” – that is, more or less unpredictably (based on sound context) – while still failing to convey semantic information, because they are processed as essentially the same “sound” by the speaker and the listener.
I guess this would include things like the “t/d/alveolar flap/glottal stop” series we just discussed in words like “latter”, where the variation among allophones is less due to sound context, and more due to the REGISTER of speech (at least in some regional varieties of English): formal and/or slow vs. informal and/or fast.
I should amend what I said just a bit: if an individual uses one allophone when speaking in one register (e.g., formal), and another when speaking in a different register (e.g., informal), this change* can* convey something “semantic” in the broader sense – that is, something CONNOTATIONAL, rather than DENOTATIONAL. A famous example is that of President Obama (and Bush before him) “dropping -g’s” in words ending in “-ing”, when trying to convey “hey-I’m-just one-of-the-guys” when speaking for a particular audience.
Then predictable isn’t really a matter of relevance, as it is predictable in either case. In one case, it is predicatable by semantic intent, the other by sound context. I guess I was looking at it more from predictable = semantic content. You can pick which it needs to be by the meaning intended, then that is predictable use, whereas it varies depending on the letters around it isn’t predictable in that same context.
True. I think that linguists tend to use the word “predictible” to refer to the sound context, though, because it’s hard to enter the mind of any speaker, so the semantic information that is trying to be conveyed is in some sense less “known” than things like the sound context. It’s something you can guess at pretty well most of the time, but sometimes can only know for sure by asking the speaker, “what did you mean by that?”.
But based on sound context, one would think I should say “shtchreet”, because I say “tchree”. But I don’t. When I pair the “tr” with the leading “s”, the “ch” goes away.
Thinking about it, it seems when I say “str” I am pairing the “t” with the “s”, then flowing to the “r”, so the “t” is cleaner. But if I just say the “tr”, the “t” forms closer to the “ch” shape.
I note that “t” is formed with the tip of the tongue near the roof of the mouth, while “ch” is formed with the mid part of the tongue at the roof of the mouth. “R” is formed with the back edges of the tongue high and the tip low. So rolling “t” to “r” seems to form with the middle of the tongue rather than the tip high as it blends to the “r”.
Don’t let my username fool you. I’m American, born and raised, from American parents. I have a couple of Irish immigrant great grandparents, and that’s the closest link I know. I don’t speak Irish Gaelic.
I grew up in Arkansas, and have lived in Oklahoma and southeast Texas.
