Unknown sound

[Kezdő]

Whilst reading through my print version of the LCK, I noticed that The Zomp made reference to one amateur conlang with a sound described as "midway between [n] and [m]." You'd think they meant [ɱ] at first, but I got to thinking at first that it meant an entirely different sound altogether. I've searched the IPA up and down and inside out, and as far as I know this sound simply does not occur naturally. However, I've been quite able to pronounce it.

So think of a nasal consonant. While you hold your lips as if pronouncing [m], you also touch your tongue to the dental area like [n]. It sounds a bit like "mn". (But don't be mistaken - it's a single consonant, not simply a doubly articulated thing.) Now, if the IPA has no characters for it then what do I use to represent it? Also, what's the technical term for this sound?

[Nortaneous]

That's a doubly-articulated consonant, a labial-alveolar, written n͡m.

And yes, it occurs naturally.

[Xephyr]

And yes, it is "simply a doubly articulated thing". "Holding your lips as if pronouncing [m]" while also "touching your tongue to the dental area like [n]" is double-articulation.

[Herr Dunkel]

Tho ɱ is a nice sound too, and doesn't require an entire series of coarticulations that are otherwise pesky and retarded

[Radius Solis]

Perhaps by "doubly articulated", you meant articulated in sequence? I.e. that what you describe is not articulated in sequence, but rather together? This is not the usual use of that term, but it's the only meaning I can think of that would make your sentence correct. And yes, a consonant can have two articulations and still be a single consonant; this is transcribed with a tie-bar.

[Kezdő]

Perhaps by "doubly articulated", you meant articulated in sequence? I.e. that what you describe is not articulated in sequence, but rather together? This is not the usual use of that term, but it's the only meaning I can think of that would make your sentence correct. And yes, a consonant can have two articulations and still be a single consonant; this is transcribed with a tie-bar.

Uh, yeah, I think. So doubly articulated means simultaneous? And it does occur naturally? Well, I'll be darned! Of course, the chances of me knowing about Yélî Dnye were pretty low.

Alright, thanks so much for the answers, guys! Behold, I am enlightened now!

[Pole, the]

So doubly articulated means simultaneous?

Yep. The same concerns e.g. [kʷ] or [pʲ].
Exceptions are sounds like [t͜s] or [ʔʰ], where the first component is a stop and the second one is a fricative, they obviously can't be pronounced simultaneously.
The fricative element then occurs directly after the stop element (when normally the plosion would happen).

[Nortaneous]

Are secondary articulations considered a subset of double articulations? I was under the impression that they aren't, but what do I know.

Also, does [ʔʰ] actually occur in any natlangs, or is it just a Dumb Conlang Thing like ɧ and using thorn everywhere?

[Pole, the]

Sorry, they both are types of co-articulation.

[Drydic]

Are secondary articulations considered a subset of double articulations? I was under the impression that they aren't, but what do I know.

Also, does [ʔʰ] actually occur in any natlangs, or is it just a Dumb Conlang Thing like ɧ and using thorn everywhere?

I've seen it somewhere, though for the life of me I can't remember exactly where. Probably North American though.

[Miekko]

Are secondary articulations considered a subset of double articulations? I was under the impression that they aren't, but what do I know.

Also, does [ʔʰ] actually occur in any natlangs, or is it just a Dumb Conlang Thing like ɧ and using thorn everywhere?

There's nothing stupid about [ʔʰ] - it's just as likely as with any other stop that voicing onset time would be non-zero for the next sound.

[zompist]

Note that the point of the comment in the LCK is that my friend didn't know anything about phonetics, so there is no answer to the question of what he meant. He was just combining letters as naive conlangers do.

(At least I think so. He may be in Chicago this year so I may see if he can produce his sound. If he really had a specific sound in mind, my guess is [ɱ] since I don't think an English monolingual would readily come up with co-articulation.)

[Aurora Rossa]

He could also have been referring to the linguolabial point of articulation, where you put the tip of your tongue on the upper lip. One could describe that as having qualities between /m/ and /n/ although your mileage may vary there.

[Kezdő]

Note that the point of the comment in the LCK is that my friend didn't know anything about phonetics, so there is no answer to the question of what he meant. He was just combining letters as naive conlangers do.

(At least I think so. He may be in Chicago this year so I may see if he can produce his sound. If he really had a specific sound in mind, my guess is [ɱ] since I don't think an English monolingual would readily come up with co-articulation.)

I'm an English monolingual and co-articulation was the first thought I had. Then again, I'm also a conlanger.

By the way, how's progress on the LCK2? Swimmingly, I suppose?

[zompist]

Pretty good, thanks. I feel it needs about 30 more pages though.

[Nortaneous]

There's nothing stupid about [ʔʰ] - it's just as likely as with any other stop that voicing onset time would be non-zero for the next sound.

The only Google results for it that aren't random data or conlangs are this, which also claims the existence of a nasalized glottal stop, and this, which claims some dialect of Chinese has it corresponding to a velar stop in others.

[Miekko]

There's nothing stupid about [ʔʰ] - it's just as likely as with any other stop that voicing onset time would be non-zero for the next sound.

The only Google results for it that aren't random data or conlangs are this, which also claims the existence of a nasalized glottal stop, and this, which claims some dialect of Chinese has it corresponding to a velar stop in others.

There's probably not that much of a point in pointing out that the glottal stop is aspirated if, say, all voiceless stops are aspirated. Pure phonetics isn't that interesting really, I guess?

[ná'oolkiłí]

No, the reason [ʔʰ] probably doesn't really exist is because the glottal stop operates differently from other stops. During the production of [t], for instance, the vocal folds are completely open, even during the period the tongue creates closure at the teeth/alveolar ridge. The VFs remain open for a period after this closure is released until they are brought together to allow them to vibrate (to produce a vowel, say). The manipulation of the length of this period of voicelessness after the burst (ie, VOT) is the difference between [t] and [tʰ]. To produce [ʔ], though, the VFs themselves create the closure (in this way, then, [ʔ] can be seen as a type of phonation). The VFs vibrate when they are close but not closed, so to begin a vowel after a glottal stop, the VFs need simply to abduct slightly. For [ʔʰ], they would go from completely closed to completely open to close and vibrating—not an inconceivable articulatory gesture, but one that is quite different from what goes on for the other stops. Also worth noting is what [ʔ] vs [t] look like on a spectrogram. A [t], no matter speech rate, is almost always a period of silence—a blank column surrounded by sounds with nontrivial intensity. [ʔ] can appear like this, but often, especially as speech rate increases, it is realized instead as a period of irregular and widely spaced glottal pulses (vibrations escaping the VFs) not unlike what creaky voice looks like.

TLDR: [ʔʰ] doesn't really exist because [ʔ] is different from other stops.

[Qwynegold]

Yikes! I have /?_j/ and /?_w/ in a conlang project. Are these possible?

[Miekko]

Yikes! I have /?_j/ and /?_w/ in a conlang project. Are these possible?

To me, it seems rather likely that the labial or palatal components would be perceived as the primary thing for those, and the ?-part would be some kind of secondary thing.

[Solarius]

Yikes! I have /?_j/ and /?_w/ in a conlang project. Are these possible?

I think Hausa has the former.

[Aurora Rossa]

To me, it seems rather likely that the labial or palatal components would be perceived as the primary thing for those, and the ?-part would be some kind of secondary thing.

Yeah, although it depends on the linguistic context as well. If the language features glottalized resonants otherwise, it would make sense to analyze them as glottalized /j/ and /w/. If the language features extensive palatal and labial coarticulation instead, one could reasonably analyze them as palatalized and labialized glottal stops. I recall reading that both interpretations have appeared, with one favored for some language families and the other favored for other ones.

[ná'oolkiłí]

Yeah, in languages that have things like that there's apparently disagreement about the best way to analyze them: I've seen all of [ʔʷ w' wˀ w̰]. Phonologically I'd label them what makes most sense... do they pattern with other labialized/palatized stops or ejectives or glottalized resonants? Just be aware their phonetic implementation might not exactly line up with your label (but that's true of any phoneme).

[Miekko]

Yeah, in languages that have things like that there's apparently disagreement about the best way to analyze them: I've seen all of [ʔʷ w' wˀ w̰]. Phonologically I'd label them what makes most sense... do they pattern with other labialized/palatized stops or ejectives or glottalized resonants? Just be aware their phonetic implementation might not exactly line up with your label (but that's true of any phoneme).

Could you tell us which languages those are supposed to exist in?

[Nortaneous]

Yikes! I have /?_j/ and /?_w/ in a conlang project. Are these possible?

Yes.

There's probably not going to be a contrast in any one language between, say, /ʔʷ/ and /w̰/ (note that w' and wˀ are just different ways of writing w̰, depending on how you're analyzing them), but that's mostly just because the inventory would have to be huge to support the series necessary to keep those stable.

/ʔʷ/ is attested in 'Auhelawa, Inor, Adyghe, Lao, etc.
/ʔʲ/ seems to be mostly an areal feature of languages spoken in/around Nigeria, such as Hausa, Fula, and Cipu, which has a very strange set of glottals. Part of that is probably transcription, though; it seems to be common in East Asia, although there it's written /ʔj/, even if it's analyzed as one consonant.
I'm too lazy to go through the rest of them but anything with <ˀ> probably 1. is North American and 2. patterns with ejectives.