tractor wrote:Cainntear wrote:So are you suggesting that having a different phoneme map from a native speaker doesn't mean you haven't learned the language wrong...?.
If their pronunciation is correct, I see no reason to claim that they've learnt the language wrong. If their pronunciation is correct, how do you know that their phoneme map is different from that of a native speaker?
OK, fair enough. However, I think the gap between our positions is starting point vs end point.
Cainntear wrote:What is the right thing in the beginning?
The right thing is obviously the way native speakers speak. Beginner mistakes are not the right thing, whether it's incorrect pronunciation or incorrect grammar.
Which is correct, clearly, but how do we get there?
Errors are inevitable early on, as is patently obvious. But which errors can be corrected early, and which corrections are important -- i.e. which corrections are necessary to lead you down the path towards talking how native speakers talk?
This is why I favour phonology over "accent". You can make something that sounds superficially very similar to the target sound, but do it a fundamentally flawed way. If you take a superficial approach, that doesn't call for correction. But if you ask yourself why and think forward, correction becomes different.
For example, after finishing MT, I took formal classes in Spanish. I started at post-beginner (because I figured my weaknesses in phraseology would be addressed by knowing similar phrase in two other Romance languages that would be more than transferrable) and my /d/ sounds were probably corrected more than others' were. I didn't care because I knew mine were better.
The teacher would accept alveolar [d] and dental [ð] as correct, whereas I was making a point of using an apico-dental phoneme. (I just remembered apico-dental and googled it to make sure I was using it right. I'm going to use it repeatedly now to make sure I don't forget it again, so I'm sorry in advance if this is boring!!)
English /ð/ is a true dental (tip of the teeth) and English /d/ is alveolar (not touching the teeth, instead touching the flesh-covered bony ridge in the roof of the mouth. Spanish /d/ is apico-dental, meaning the tongue touches the gumline of the tooth (I hate the term apico-dental personally as to me the tip of the tooth is the peak, not the gumline, and my brain refused to stop seeing "apico" as anything other than "peak"). My classmates were using English /ð/ and English /d/, and I'm deliberately using the phoneme marks // instead of phone marks [] because I am convinced they were considering them different phonemes. Why was I convinced? Because when reading something they would specifically ask "is that
d or
ð?" they had a hard distinction and they were acting in a way that was consistent with believing that one letter could denote two distinct phonemes. The thing with allophones is that they're context dependent, and /d/ in an intervocalic (between vowels) position is realised as [ð] and in a non-intervocalic position its [d] and the native speaker doesn't choose between phonemes, but automatically selects the allophone based on context.
[Note: I deliberately made a distinction between "choosing" and "selecting" here as I feel that the word "choice" implies more agency.]Meanwhile, I had reasoned through it and decided that if I was to learn /d/ as a phoneme and [d] and [ð] as allophones then I had to start by taking an "architype" of the phoneme (which in the case of /d/ is [d]) and then get my brain and muscles to learn to be "lazy", because the intervocalic [ð] allophone is at its heart an attempt to pronounce [d] that doesn't quite make it and "rounds off the corners" -- the [ð] allophone has the tongue moving
towards the [d] allophone position, but it doesn't quite make it there.
I could only do this because I had been studying linguistics, and although it was done in a course on English language, the course was very cleverly designed to do multiple jobs, so English phonology was used as a demonstration case for linguistics that applied to multiple languages. As I said before, languages tend to have a tendency to either reduce vowels or to reduce consonants, and the whole thing about English schwa was deeply embedded in that context, so when I started on the post-beginner Spanish course about a year later, I was extremely conscious of that.
So if I said an intervocalic D in Spanish, I would consciously, deliberately overpronounce it, resulting in what the teacher detected as a [d] allophone which she would correct to [ð] allophone, but she never ever corrected people who were using the English /ð/ phoneme or English /d/ phoneme.
I have frequently been mistaken for a native speaker. I have no proof that the above is a necessary precondition to doing that, but I am convinced that it's a major part of what made me so good at Spanish.
Mind you, this is just the ordinary t. Not the English th, not the soft Russian t'. Just your ordinary t. Apparently, those are a little bit different in English and in Russian. And any teacher who spends time on this in an ordinary language course is wasting the students' time. 
(whose emoji I've just discovered includes the adjectival prefix "very").
