Possible sound changes for [ɲ]

[Das Baron]

[J] in X-sampa, in case the IPA character in the title doesn't show up. What are some possible sound changes for it? It's a sound I don't like and rarely include in my languages, but a recent conlang developed it as a result of a change I had to include for other reasons and now I'm not sure how to get rid of it naturally. It seems too easy to have it just change back to [n].

[äreo]

IIRC, French (in at least some varieties) has /J/ > /n/ before /i/ and > /jn/ finally.

[Wattmann]

/J/ > /n`/ / __{a, o, u}
/J/ > /n/ / __V
/J/ > /j/

That would give you three different phonemes from one /J/. I'd expect /J/ > (/n`_j/ >) /n`/ as the full for the first.
Though, this might not be terribly realistic, I find it natural for my mouth.

[Zhen Lin]

The 日 initial in Chinese is thought to have been some form of palatal nasal (or even palatal nasal affricate...!). The reflexes in modern languages are quite varied:

Mandarin: voiced retroflex fricative (pinyin <r>), or null
Cantonese: palatal approximant, or null
Min Nan: voiced alvelo-palatal fricative, or lateral approximant
Japanese: palatalised alveolar nasal, or voiced alveolo-palatal affricate
Korean: palatal approximant, or null
Vietnamese: palatal nasal

[Rory]

Change it to /nj/, as happened in Latin American Spanish.

EDIT: Although it didn't happen where I thought it did (New Mexico), it still seems to me to be a plausible enough change, especially in a situation where there is a large population of non-native speakers whose L1 does not have palatal nasals.

[Ser]

Change it to /nj/, as happened in Latin American Spanish.

Source? And in what dialects of Latin America? I've never heard of this before, though my dialect does have one word where this happened: doña, reduced to ña and pronounced [nja]. Otherwise it's a plain [J].

[----]

How about ɲ > ĩ / j̃ like Portuguese does sometimes? That could make for some interesting diachronics.

[Rory]

Change it to /nj/, as happened in Latin American Spanish.

Source? And in what dialects of Latin America? I've never heard of this before, though my dialect does have one word where this happened: doña, reduced to ña and pronounced [nja]. Otherwise it's a plain [J].

Wow. I had a quick look for sources, but I must have dreamed it up. I thought "oh, wait, it's just New Mexican Spanish", but apparently I dreamed that too. (New Mexican Spanish does have interesting verbal analogical changes though, but that's totally different.) Oops, I messed up. I'll edit my original post to reflect that.

[Aurora Rossa]

I would have it change to /n/ and have it affect the surrounding vowels in some way. Maybe following back vowels become fronted so /a o u/ become /æ ø y/ or something.

[Jetboy]

I believe Ancient Greek, and maybe also French, changed /nʲ/ to /jn/, which formed diphthongs with proceeding vowels: /banjō/ > /bainō/, /ktenjō/ > /kteinjō/, etc.. The neat thing here is that since the /j/ was only present in the present stem, the other stems retained the original vowel, provding some nice alterations.

<banjō>. Hah.

[Šm Mepuyoš ab Duhen]

Kajkavian dialect of Croatian changes /ɲ/ to /jn/.
I have idea of changing /ɲ/ to /ni/.

[Curlyjimsam]

Sound changes I tend to apply to this sound include [ɲ] > [j] and [ɲ] > [ŋ]. The latter hasn't been mentioned in this thread yet but seems quite plausible to me; if anything, I'd predict it to be more likely than [ɲ] > [n] as there is a greater similarity in the articulation.

[Bedelato]

I would suggest one of either [n], [ŋ], [nj], [ŋj], or [j].

[linguoboy]

Sound changes I tend to apply to this sound include [ɲ] > [j] and [ɲ] > [ŋ]. The latter hasn't been mentioned in this thread yet but seems quite plausible to me; if anything, I'd predict it to be more likely than [ɲ] > [n] as there is a greater similarity in the articulation.

Vietnamese (Hanoi dialect) has /ɲ/ > [ŋ], e.g. lính ['lïŋ˧ˀ˥] "soldier". This causes fronting of back vowels, e.g. banh ['ɓæ̈ŋ̟˧ˀ˥].

[Wattmann]

Sound changes I tend to apply to this sound include [ɲ] > [j] and [ɲ] > [ŋ]. The latter hasn't been mentioned in this thread yet but seems quite plausible to me; if anything, I'd predict it to be more likely than [ɲ] > [n] as there is a greater similarity in the articulation.

But isn't [J] closer to [n`] than [N]?

[----]

Sound changes I tend to apply to this sound include [ɲ] > [j] and [ɲ] > [ŋ]. The latter hasn't been mentioned in this thread yet but seems quite plausible to me; if anything, I'd predict it to be more likely than [ɲ] > [n] as there is a greater similarity in the articulation.

Vietnamese (Hanoi dialect) has /ɲ/ > [ŋ], e.g. lính ['lïŋ˧ˀ˥] "soldier". This causes fronting of back vowels, e.g. banh ['ɓæ̈ŋ̟˧ˀ˥].

It's not so much fronting of back vowels, really. According to the grammar I've been using, orthographic a before nh/ch finals is better analyzed as the phoneme represented by <ă> in other positions, which for many speakers, the primary realization is in fact /æ/. This makes sense, because other than that vowel, only /i/ and /e/ can appear before the palatal finals (i.e. No back vowels can precede it). This leads many phoneticians to analyze /-ɲ/ as /ŋ/ affected by a front vowel, which might have been where our conflicting interpretations arose from.
Nevertheless, other Vietnamese dialects do similar things with it, particularly the Saigon/HCMC dialect. Using the same example, 'lính', we get this:
/lɪɲ/ > [lɨn] (tone not marked because of irrelevance)
That /ɨ/ can even appear as [ɯ] for a significant number of speakers. So what we have is backing/centralization of vowels, and the palatal nasal becomes alveolar.

[AnTeallach]

Sound changes I tend to apply to this sound include [ɲ] > [j] and [ɲ] > [ŋ]. The latter hasn't been mentioned in this thread yet but seems quite plausible to me; if anything, I'd predict it to be more likely than [ɲ] > [n] as there is a greater similarity in the articulation.

But isn't [J] closer to [n`] than [N]?

Well, the default meaning of [ɲ] is palatal, and palatal articulations get classified under the "dorsal" (body of the tongue, like velars and uvulars) category, rather than "coronal" (blade and tip of the tongue, like alveolars). So in that sense [ɲ] has more in common with [ŋ] than [n]. But it perhaps should be noted that the [ɲ] symbol is apparently also used for palatalised postalveolars (e.g. in Polish according to Wikipedia), which sound similar, and those are classified as coronal.

Medieval Scots apparently had a palatal-like nasal in Gaelic loanwords, written nȝ as in MacKenȝie, and this sometimes became [ŋ] as in Menȝies (now written with a z of course). So, if you're prepared to invoke a bizarre orthographic confusion followed by a spelling pronunciation, you could even change [ɲ] into [nz]...

[Nortaneous]

Well, the default meaning of [ɲ] is palatal, and palatal articulations get classified under the "dorsal" (body of the tongue, like velars and uvulars) category, rather than "coronal" (blade and tip of the tongue, like alveolars). So in that sense [ɲ] has more in common with [ŋ] than [n]. But it perhaps should be noted that the [ɲ] symbol is apparently also used for palatalised postalveolars (e.g. in Polish according to Wikipedia), which sound similar, and those are classified as coronal.

There's a separate series of symbols for coronal palatals, as outlined in that article (ȶ ȡ ɕ ʑ ȵ ȴ), but it seems like nobody uses any of them, besides the ones for sibilants.

[Ser]

There's a separate series of symbols for coronal palatals, as outlined in that article (ȶ ȡ ɕ ʑ ȵ ȴ), but it seems like nobody uses any of them, besides the ones for sibilants.

Nobody but Sinologists, at least.

[Zhen Lin]

Medieval Scots apparently had a palatal-like nasal in Gaelic loanwords, written nȝ as in MacKenȝie, and this sometimes became [ŋ] as in Menȝies (now written with a z of course). So, if you're prepared to invoke a bizarre orthographic confusion followed by a spelling pronunciation, you could even change [ɲ] into [nz]...

Well, given that [ɲ] and [ȵʑ] are two of the reconstructed Early Middle Chinese values for the 日 initial, I don't think we really need to invoke spelling pronunciation...

[Skomakar'n]

Sound changes I tend to apply to this sound include [ɲ] > [j] and [ɲ] > [ŋ]. The latter hasn't been mentioned in this thread yet but seems quite plausible to me; if anything, I'd predict it to be more likely than [ɲ] > [n] as there is a greater similarity in the articulation.

I did [ɲ] > [ŋ] in my conlang, at least.

[Hakaku]

[nj], [ɲ] > [ŋ] occurred in Canadian French. [nj], [ɲ] > [n] occurred in Ryukyuan.

[Tropylium]

For a Uralic survey: *ń (which may have been [ɲ] or [nʲ]) is fairly stable — a few languages change it to /n/ or /j/ (Mari does both: /n/ initially, /j/ medially), and a change to *jn has been posited (but not very commobly accepted, AIUI) in Finnic, but there's nothing odder than that.

Denasalization to [ɟ] or [dʲ] (or something these can further decay to) should also work, tho I'd expect to see denasalization of other nasals too in at least some positions in conjunction.