Template talk:ang-IPA

Uppercase

@Benwing2, Mahagaja Could the module convert its input to lowercase itself before doing any further processing? It's what I did with Module:sl-IPA and it seems to work just fine. —Rua (mew) 13:52, 13 December 2019 (UTC)

@Rua Yup, I have a change to commit that does exactly that. Benwing2 (talk) 17:17, 13 December 2019 (UTC)

underiendlic

It's generating the wrong stress here, the second syllable is stressed, not the third. —Rua (mew) 17:38, 22 March 2020 (UTC)

That's because the module is misinterpreting "under" as a verbal prefix. Since the actual prefix here is the adjectival negator un-, which I understand to usually take primary stress in Old English (as in modern German, e.g. ˈunschädlich), shouldn't the actual pronunciation be /ˈunˌder.jend.liːt͡ʃ/, with the first syllable stressed?--Urszag (talk) 00:47, 24 March 2020 (UTC)

Distinction between /j/ and /dʒ/

On menġan the template generates /j/ as the phoneme, but this is incorrect. /j/ and /dʒ/ were distinguished after nasals; menġu (“multitude”) is [ˈmenju] and not [ˈmendʒu]. —Rua (mew) 09:06, 20 April 2020 (UTC)

It would be pretty strange to analyze OE as having a phonemic distinction that exists only after /n/, so if /j/ and /dʒ/ are treated as distinct phonemes in that context, it would be best to transcribe them as distinct phonemes in other contexts as well. However, the allophonic analysis is still possible if we stipulate that there was some kind of difference in the underlying phonological structure or prosody of menġu and menġan; e.g., if menġu is treated as having synchronic and not only diachronic elision of a vowel between the n and ġ. If allophony of [j] and [dʒ] is denied because of pairs like [ˈmenju] and [ˈmend͡ʒɑn], then by analogy allophony of [ɣ] and [g] would also have to be denied because of pairs like [ˈsynɣiɑn] and [ˈbringɑn] (Ringe and Taylor 2014 page 4). In general, there are a number of different approaches to the phonemic analysis of Old English dorsal consonants (for example, some scholars call ġ and ċ allophones of g and c—even though they are not straightforwardly in complementary distribution—the nature of "geminate" ġġ and ċċ is disputed, the timing of their development to affricates is unclear, and the timing of the sc > [ʃ] change and its phonologization is disputed). In many areas, I don't think we will be able to identify a single analysis that is unanimously regarded as correct.--Urszag (talk) 00:43, 21 April 2020 (UTC)

@Urszag: Maybe not, but in practical terms, if someone were to create menġu and put this template on it, it will give the wrong result. syngian already has the wrong result. So clearly some distinction has to be made. —Rua (mew) 09:04, 30 April 2020 (UTC)

The current template does already support transcribing [nj] and [nɣ] by including square brackets around the phone. I just haven't edited the entry for syngian because I hadn't seen a source aside from Ringe and Taylor that mentioned the existence of a distinction between [ng] and [nɣ], so I was still a little unsure about how certain or unanimous the reconstruction with [nɣ] is. I guess I'll add it along with a citation.--Urszag (talk) 09:23, 30 April 2020 (UTC)

making parameter 1 optional

Can the template and/or module be rewritten so that |1= can be omitted if it's identical to the page name (i.e. no diacritics necessary)? That's the way most autogenerated pronunciation templates for other languages work. —Mahāgaja · talk 12:34, 31 March 2021 (UTC)

initial f=v

How does one force an initial f as a /v/ ? As in fersian = "versian" ? Leasnam (talk) 18:22, 16 September 2023 (UTC)

I don't think that's possible currently. It seems pretty questionable to use /v/: given our existing transcriptions, that would imply a phonemic contrast with /f/, which I believe is contra the consensus. Lass 1993 argues that West Saxon, Kentish, and West Mercian likely followed the rule that "fricatives are (generally) voiced except in obstruent clusters and finally", "categorically" for /θ/ and "probably the majority" for /f, s/ (which would mean our current phonetic transcriptions are misleading/incomplete). As an alternative interpretation, Lass cites Campbell 1959 as saying that Fergilius and fers show substitution of [f] for Latin [v]. Neither of them seems to assume a contrast in word-initial position between such loans and native vocabulary.--Urszag (talk) 00:16, 17 September 2023 (UTC)