Realisation of /l/

Does l-vocalisation have any effect on flapping for you? I was watching a TV show recently in which an AAVE speaker with l-vocalisation pronounced /Vlti/ as [Vo4i], and I thought it sounded strange. I suppose I have an underlying /l/ that still counts as a consonant in the application of the flapping rule. This underlying /l/ would also explain why I never have a linking /l/ after 'now' even though it is homophonous with 'Nile' for me as [naU] (though 'now' as often [na@] for me, while 'Nile' never is.)

I've been thinking about /ul/ and /Ul/ and I've realized why they haven't merged form me like other tense-lax pairs before historic /l/. It's because /ul/ has l-breaking, while /Ul/ doesn't. So 'pool' and 'pull' are clearly distinguished as [pu.Uo] and [pUo]. Before vowels, like 'the pool is' and 'pull it' they are [pu.Uo5Iz] and [pU5I?].

>>Does l-vocalisation have any effect on flapping for you? I was watching a TV show recently in which an AAVE speaker with l-vocalisation pronounced /Vlti/ as [Vo4i], and I thought it sounded strange. I suppose I have an underlying /l/ that still counts as a consonant in the application of the flapping rule. This underlying /l/ would also explain why I never have a linking /l/ after 'now' even though it is homophonous with 'Nile' for me as [naU] (though 'now' as often [na@] for me, while 'Nile' never is.)<<

l-vocalization seems to disable flapping for /t/ and /d/ immediately following it in my dialect, with /t/ following it normally being aspirated as well even if the following vowel is unstressed. Note that this is in contrast with /r/, after which /t/ and /d/ are flapped (or simply elided due to flap elision IMD) when followed by an unstressed vowel.

As for "now" and "Nile", the diphthong from /aU/ differs from that from both /al/ and (commonly) /aIl/, as the former [a_-U] involves a central starting point as has a rounded ending point while the latter [aM] involves a front starting point and an unrounded ending point. About /aI/, IMD it undergoes smoothing to [a] before high to mid back vowels as well as before /l/ and /w/, even across word boundaries, except in more careful speech; I would assume that something similar is the case with your dialect with your mention of "Nile" above.

>>I've been thinking about /ul/ and /Ul/ and I've realized why they haven't merged form me like other tense-lax pairs before historic /l/. It's because /ul/ has l-breaking, while /Ul/ doesn't. So 'pool' and 'pull' are clearly distinguished as [pu.Uo] and [pUo]. Before vowels, like 'the pool is' and 'pull it' they are [pu.Uo5Iz] and [pU5I?]<<

My dialect, on the other hand, consistently preserves all the tense-lax pairs even though, from what I have heard, it is quite common for English dialects with l-vocalization to merge such pairs. And that is even with my dialect lacking breaking except before /aI/, /aU/, and /OI/ (and even then such does not happen for my /aI/ if I smooth it to [a]). That makes me wonder why my dialect has not lost such pairs even for /ul/ versus /Ul/ (the pair which one would expect to be the least differentiated by far in my dialect).

Another thing I thought of is that pronouns seem to act like suffixes when it comes to realisations of /l/. For example, 'tell him' (often pronounced without /h/ in unstressed situations) is [tElIm] rather than [tEo5Im] as would be expected from my explanations above. This works with 'him' 'her' and 'it', and also applies to the particle 'a', as in 'fulla money'. So the sentence "Tell him about the bag full of money," would probably be [tElI~ m@bEU? D@ bE@g fUla mVni] in casual speech. Do you have your dark-l equivalent or light-l equivalent in these situations?

>>Another thing I thought of is that pronouns seem to act like suffixes when it comes to realisations of /l/. For example, 'tell him' (often pronounced without /h/ in unstressed situations) is [tElIm] rather than [tEo5Im] as would be expected from my explanations above. This works with 'him' 'her' and 'it', and also applies to the particle 'a', as in 'fulla money'. So the sentence "Tell him about the bag full of money," would probably be [tElI~ m@bEU? D@ bE@g fUla mVni] in casual speech. Do you have your dark-l equivalent or light-l equivalent in these situations?<<

I have a dark-l equivalent in these places, but it acts as if it were in an intervocalic position rather than a coda position here (as in it is commonly vocalized but vocalization is not obligatory outside of careful or emphatic speech unlike in coda positions).

I was reading a book about Canadian English today and it mentioned some accents in Newfoundland having [M\] for post-vocalic /l/. I was surprised, because most other Newfoundland accents have clear-l in all positions and I don't think I've ever heard this realisation here. Have you ever heard this Newfoundland [M\] and is it the same as yours?

>>Have you ever heard this Newfoundland [M\] and is it the same as yours?<<

I really would not know, never really having heard Newfoundland English to begin with.

>>I was reading a book about Canadian English today and it mentioned some accents in Newfoundland having [M\] for post-vocalic /l/. I was surprised, because most other Newfoundland accents have clear-l in all positions and I don't think I've ever heard this realisation here. Have you ever heard this Newfoundland [M\] and is it the same as yours?<<

One note, that I forgot to say earlier, is that I only have [M\] as an allophone of /l/ in "light" positions, and what I have in postvocalic "dark" positions and often in intervocalic "dark" positions is really [M_o], which is about the same height as my usual [U] but fully backed and unrounded, or its rounded counterpart after rounded vowels.