Could [ l̰ ], [n͡l], [lⁿ] and [nˡ] be the same phoneme?

If you're new to these arts, this is the place to ask "stupid" questions and get directions!
Post Reply
Nloki
hieroglyphic
hieroglyphic
Posts: 29
Joined: 15 Dec 2018 16:01

Could [ l̰ ], [n͡l], [lⁿ] and [nˡ] be the same phoneme?

Post by Nloki » 23 Mar 2019 17:00

While trying to create a set of lateral consonants for my conlang, Nlokian, I noticed that the digraph "nl" in its name wasn't just a pair of separate phonemes. Try to pronunce [n] and [l] simultaneously. The problem is I found nothing on the Internet about nasalized lateral approximants so I quite don't know how it would be represented on IPA. So how do you think it should be written: [ l̰ ], [n͡l], [lⁿ] or [nˡ]?

User avatar
Creyeditor
mongolian
mongolian
Posts: 4590
Joined: 14 Aug 2012 18:32

Re: Could [ l̰ ], [n͡l], [lⁿ] and [nˡ] be the same phoneme?

Post by Creyeditor » 23 Mar 2019 18:50

It is definitely possible to nasalize laterals i.e. to have nasalization and a lateral articulation in the same segment. This would be written as an l with a tilde above [l̃].
[n͡l] would be an nasal-to-lateral affricate. I am not sure if this exists. This notation is sometimes also used to indicate prenasalization. In this case you would have a nasal closure preceding the lateral. This does not seem to be what you describe.
[lⁿ] is a nasally released lateral and [nˡ] is a laterally released nasal. Since usually only plosives and affricates have releases, this notation does not make much sense. In any case it would indicate temporal ordering, so not what you want.
[ l̰ ] is a creaky voiced lateral. Possible, but no direct relation to nasalization.

tl;dr I would suggest going with [l̃], judging from you description.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]

Nloki
hieroglyphic
hieroglyphic
Posts: 29
Joined: 15 Dec 2018 16:01

Re: Could [ l̰ ], [n͡l], [lⁿ] and [nˡ] be the same phoneme?

Post by Nloki » 22 Apr 2019 14:16

Thank you very much, Creyeditor. I'm sorry for not answering before, in fact I was quite embarrassed after realising that my question was worse than silly...
Anyway, I have come up with a new pair of doubts about phonology so far:
•Would it be sensible to use a set of obstruents where some of them are phonemically voiceless and others phonemically voiced even in the first stages of a conlang's evolution? For example, if I had a voiceless postalveolar non-sibilant fricative /ɹ̠̊˔/ but not its voiced phonemical equivalent, and viceversa, in case I would have just a voiced labiodental fricative? Or with stops: if I had voiceless bilabial, alveolar and velar plosives but a phonemic voiced palatal stop?
•Can a phoneme become an allophone depending on its setting within a syllable? For example: /l̃/ (aka: the-phoneme-we-were-just-discussing-about-in-the-first-posts) should only exist as word onset because placing it into the word itself would make it become an allophone of an onset /l/ placed after a coda /n/?
Thanks

User avatar
Creyeditor
mongolian
mongolian
Posts: 4590
Joined: 14 Aug 2012 18:32

Re: Could [ l̰ ], [n͡l], [lⁿ] and [nˡ] be the same phoneme?

Post by Creyeditor » 22 Apr 2019 21:00

I don't think the question was silly at all, and I see no need to be embarassed.

There is no general answer to your first question, but there are some tendencies: if you have only one series of stops, they tend to be voiceless. The same is often true for affricates. If you have only one series of fricatives they tend to be voiceless, too. The second tendency is weaker however, since sometimes the voiced fricatives pattern more like approximants. There are also some tendencies concerning place of articulation. Labial sounds are more likely to be voiced and non-labial sounds are more likely to be voiceless. Now coming to your more specific questions: a voiceless postalveolar non-sibilant fricative /ɹ̠̊˔/ without its voiced phonemical equivalent looks very naturalistic to me. The same is true for "just a voiced labiodental fricative" without a voiceless equivalent. Your suggestion for the stops looks much less realistic to me. I would expect the labial stop to be voiced instead.

IIUC, you have a contrast between /l̃/ and /l/ word initially. A cluster /nl/ cannot occur in this position. Word medially, a /nl/ cluster is realized as [ l̃]. To me, this looks like a neutralization. Word-medially, the contrast between /nl/ and / l̃/ is neutralized to [ l̃]. We cannot know anymore which of them it was underlyingly, unless we have independent evidence from other phonological processes. The situation looks natural to me.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]

Post Reply