Pronunciation-voiced th-assimilation

[euncu]

"As with many English consonants, a process of assimilation can result in the substitution of other speech sounds in certain phonetic environments. Most surprising to native speakers, who do this subconsciously, is the use of [n] and [l] as realisations of /ð/ in the following phrases:[1]
join the army: /ˈdʒɔɪn ðiː ˈɑːmi/ → [ˈdʒɔɪn niː ˈɑːmi]fail the test: /feɪl ðə ˈtɛst/ → [feɪl lə ˈtɛst] "

I quoted the above text from ;Pronunciation of English th - Wikipedia, the free encyclopedia




My question is that is this a general rule, and is it also correct for the words starting with the voiced th other than the definate article?


I'm asking this because I've long been trying to use this sound correctly when I'm speaking but it slows my speech down for the sake of doing it right and even then I still don't know whether it is OK. So if there is a general principle confirming the text quoted above it will at least be a small relieve that there are times I get to evade pronunciating this sounds.


Thanks for your answers in advance.

[BobK]

Note the warning at the beginning of the article: it needs attention; in particular it needs useful references. The first sentence has a footnote that refers to 'Collins and Mees', but this is not listed in the bibliography. I presume it refers to Practical Phonetics and Phonology: A Resource Book for Students Routledge English Language Introductions: Amazon.co.uk: Beverley S. Collins, Inger M. Mees: Books.

I can almost hear the author's Linguistics 101 lecture saying 'Get out your Collins and Mees'. And the note in Wikipedia only says that its examples (not its statements) are from that book. Generally I think the way the article expresses itself is questionable. Assimilation of the kind described does sometimes occur. But if you go out of your way to say things like /lʊk ɪn nə bɒks/ you will only raise eyebrows.

b

[birdeen's call]

I think [,ɪnə'bɑːks] is a possible realization of "in the box" in America. I think I've heard it and I would guess th-alveolarization could be the culprit. After changing [ð] to [d], it seem quite natural to elide it, as in [ˈwɪnər] for "winter".