For Reasons™[1] I have a terribly hard time pronouncing [k], [p], and [t]. Also, some specific examples of allophony happen very naturally for me (naturally as opposed to deliberately, not natural as opposed to artificial), and I wonder if I ought to try to aim to avoid that. So I have a series of questions about the acceptability/desireability to you of specific pronounciations. Not as universal judgments of all nasin(!), merely as guidance for me to develop mine.
vowel length variation (for example, long vowels on the first vowel of a multi-syllable word as part of marking stress, although I'm not sure I want to commit to always doing that)
null onset sometimes but not necessarily pronounced as glottal stop (I think for me it would happen: after coda n, after a pause, or in distinct speech)
/k p t/ consistently [kʰ pʰ tʰ] word-initially and [g~k~kʰ b~p~pʰ ɾ~d~t~tʰ] otherwise (this is my default)
/k p t/ consistently [kʰ pʰ tʰ] word-initially and [g~k b~p ɾ~d~t] otherwise
/k p t/ [g~k b~p d~t] (at the time of writing, as in, before getting feedback, this is the option I am leaning towards; I can barely distinguish those pairs)
/k p t/ consistently [g b d] (with some care I think I can get here)
/o/ as [ɒ̝] before coda /n/ (it's a very similar vowel to Received Pronunciation and General American "oral" and "forum"; very very open I know, but genuinely a thing for me; I do aim to avoid this, but I would like to get some feedback on how hard I ought to try)
coda /n/ being pronounced [m] before /p/
Finally, a fun fact about my struggles with toki pona phonology: "ji" is phonotactically banned, but I actually have a similar problem but in reverse: for me, [j] generally disappears after [i], so words like "ijo", "sijelo", and "kijetesantakalu" require special attention.
Also, in making this post I realized that I have been mixing up [ e~e̝ ] and now I've learned to elicit 5 different front unrounded vowels from myself (not 100% sure where they fall, but I think [i~e̝~e~ɛ̞~a] (maaaybe ), which I am quite happy with. I can also distinguish a sixth one in between [e~ɛ̞], but can't produce anything in the [ e̞~ɛ ] range on demand (yet?).
Native Danish speaker, a language with 0 (or, in my otherwise pretty General American accent, "sero") phonemic voicedness distinctions, using aspiration instead ↩︎
if you pu, lengthening the vowel/syllable is described in the book as a way to mark syllable stress
I do glottal stops between words. Can't really stop that. In fact, when I was learning, I was specifically told: Why would you want to try to get rid of that?
voiced and unvoiced plosives should be fine. Switching between those could be a tiny little strange and some listeners might have to adjust, but yea, it's commonly accepted that "dogi bona" is a viable way to pronounce "toki pona"
so many speakers (and I think that includes me) do aspirated plosives, so that should also not be an issue?
coda /n/ being pronounced [m] before /p/
"assimilation" of n-coda is pretty standard, "anpa" -> /ampa/, "ma Sonko" -> /ma Soŋko/ etc.
for me, [j] generally disappears after [i], so words like "ijo", "sijelo", and "kijetesantakalu" require special attention
I don't see an issue with j "dissapearing" (but maybe someone else does?). I don't know if there'd be a meaningful difference, to my ears, between "ijo" and "io", "sijelo" and "sielo", "kijetesantakalu" and "kietesantakalu". Unless this is part of your difficulty, that having 2 vowels next to each other can be an issue?
Shock! Horror! Ike! I have discovered that I actually tend to do this before coda /n/ way more generally! Not entirely sure what my brain's rule is, it happens with "lon" but not "pona" (EDIT: I figured out the rule, it's before any coda /n/ which is of course why it doesn't happen in "pona" because that doesn't have coda /n/). As such I am now slightly more concerned with feedback on this, although I am still inclined to make the effort to avoid it.
My main concern with the plosives is consistency, mostly because I, in the spirit of minimalism, want to keep my phonology simple, without importing too many rules from other languages I speak about when to voice/unvoice/aspirate.
Not a problem at all, I just thought it would cause ambiguity because I forgot that the phonotactics forbid it
as a native Mandarin speaker, i have a similar but more extreme problem of my zeroth nasin aspirating every plosive.
i have been deliberately deaspirating them after lipamanka saying something like "there's so much aspiration i can't focus" after hearing a recording of my pronounciation (that was a long time ago so my memory is definitely inaccurate)
At least that is consistent, though! For me, I didn't consider the aspiration itself a problem, only the fact that I find it very hard to control it, and so would end up with an idiophonology that one would need to describe in terms of rules for allophony of /p t k/ in addition to the the allophony that toki pona is already broadly considered to have ([m] in nanpa, for instance), which does not align with my personal toki pona goals. If I could make myself aspirate always I might have gone for that, but I don't trust myself to consistently aspirate /t/ between vowels in spontaneous speech.
That is really cool :D I love all the things you can do with vowels other than just pick a spot in vowel space: nasalization, creaky voice my beloved exists on the spectrum from full voicelessness to complete glottal closure (with creaky voice opposite to breathy voice), and there's even harsh/ventricular vs yawny/hollow/faucalized on the spectrum of "expanding or contracting the larynx". Also tone! And syllabic consonants! :D
Huh... I do that too and never questioned why, but that's probably the reason. And I'm not even fluent in Mandarin! But it's the only other language I've studied with only nasal codas, so I guess despite the complete lack of nasalization in my native language (German) this has somehow affected my Toki Pona accent. Amazing.
vowel length variation and a glottal stop null onset both seem completely normal to me!
aspiration on /k p t/ feels a little odd to me, but i think i would understand it; i think /k p t/ [g~k b~p d~t] is a great option, although [g~k~kʰ b~p~pʰ d~t~tʰ] would also be fine. thinking more about it, i think aspiration in toki pona gives me english speaker vibes but not much else
the vowel in "oral" and "forum", at least in my (californian english) speech, is definitely fine for toki pona; it doesn't sound off at all
assimilation of coda /n/ is very, very common among toki pona speakers; i bet you'd have a harder time finding someone who doesn't do it
i think for ijo, sijelo, and kijetesantakalu you don't necessarily need the /j/, they sound fine without it—i think my /j/ also near-disappears after /i/, and no one's ever commented on it. /j/ is basically a syllabic /i/ so i think it's hard not to elide it when they're placed right next to each other
overall i think your pronunciation is completely fine! aspiration is the only thing that feels just a bit off but it's still completely understandable
Thanks for your feedback :D (and also thank you to @retsamys, I forgot to say that)
In the end, in the interest of consistency (or rather of not making my phonology more complicated than what is already common), I decided to try to go for [b d g], but still with coda n assimilation. I think I'll also aim for short vowels but without being too concerned if some long ones slip in there (if I notice too many tho, I'll try to instead use them word-initially consistently).
As for the vowel, I have decided that the combined existence of the father-bother/PALM-LOT merger, where lager (with what wikipedia calls the PALM vowel) and logger (with LOT) become homophones, the CLOTH-THOUGHT merger, and the cot-caught/LOT-THOUGHT merger, which together result in some North American speakers having the same vowel for all 4 for PALM-LOT-CLOTH-THOUGHT, for example having baht and bought be homophones(!), means that I consider [ɒ̝] too easily confused with toki pona /a/, and so I I'll try to avoid it. Like, "father" is given in pu as an example word for /a/, but there are apparently English speakers who have the same vowel for "father" and "rock" (I wonder if that means they use the same vowel for both syllables of "oral"?).
