Zompist Bboard Again

bradrn wrote: ↑Mon Aug 30, 2021 4:02 am I found this construction interesting:

zompist wrote: ↑Mon Aug 30, 2021 3:59 am He's an RL friend.

Personally, I would say ‘a RL friend’, with the other article. Which option does everyone else prefer?

I prefer to rephrase such constructions so I don't have to decide

I think I would use "an IRL friend"

He was an SS officer.
* He was a SS officer.

Works fine for me.

It probably depends on whether you're pronouncing the letter names, or the underlying words.

I would write "He's a RL friend" (well, I'd actually write "He's a friend IRL") but I would write "He was an SS officer", because in the former case I would expand "RL" to real-life when actually speaking, whereas I would not expand "SS" while speaking.

Rounin Ryuuji wrote: ↑Mon Aug 30, 2021 8:43 am It probably depends on whether you're pronouncing the letter names, or the underlying words.

Exactly, and I was thinking of RL as an abbreviation, not a way of avoiding typing.

The complication here is that none of us are talking, so whether to use "a" or "an" is kind of artificial. Normally we use the form corresponding to how we'd say it if we were speaking, but there's no underlying phonetic facts about words that stay in our heads.

Which makes me wonder... you emoticon users, would you write "an ", "a ", etc.?

To build on that, I think, in this case, the article would condition the reading — a RL /riːl laif/ friend, an RL /ɑːɾ ɛl/ friend.

Travis B. wrote: ↑Wed Jul 14, 2021 8:51 pm Does anyone else here pronounce initial /kw ɡw/ in English as [kʷʰ ɡʷ~kʷ], i.e. not as two distinct segments but as a single labialized consonant?

Travis B. wrote: ↑Wed Jul 14, 2021 10:31 pm

Rounin Ryuuji wrote: ↑Wed Jul 14, 2021 9:58 pm I'm pretty sure my /kw/ is phonetically [kʷʰ] — it's articulated very quickly, like two sounds superposed on top of each-other, or with only a very brief offglide — but as for /gw/, I think it's, weirdly enough, [gw]; I feel like I take noticeably longer to articulate the second one (I also partially-devoice initial voiceless consonants, but not fully, as some speakers do, so it isn't really truly [kw]), and it feels like the [w] is more fully realised than in /kw/.

What I notice is that I round my lips before I even pronounce the stop, indicating that the stop itself is labialized; there is no transition from an unlabialized stop to a labialized glide for me in either case.

The phenomenon is often referred to as coarticulation, and it is normal, although the degree of coarticulation may vary depending on speech rate, among other things. Coarticulation basically refers to the overlap during the transition between different articulatory gestures. Remember that our speech organs are more or less moving the entire time that we speak, and readjusting the articulators take some time. It is also not necessarily the case that the different articulators move at the exact same time. I think you can probably move you tongue quicker than you can readjust your lips.

I don't think there's a cross-linguistically valid phonetic difference between [kw] and [kʷ], whether a language has one or the other is more of a phonological question. The [k] of [kw] will typically have anticipatory rounding due to the fact that readjusting the lips takes some time, and [kʷ] will typically have a [w]-like off-glide (maybe depending on the following vowel) due to the fact that readjusting the tongue and lips takes some time. This is not to say that a language couldn't distinguish /kʷ/ and /kw/ through some means, for example by timing or type of lip rounding (although I don't know of any such examples).

See also Hardcastle, William J. and Hewlett Nigel (1999) Coarticulation: Theory, Data and Techniques.

This type of phonetic coarticulation should be distinguished from phonemically coarticulated consonants.

Ephraim wrote: ↑Sat Sep 04, 2021 4:55 pm I don't think there's a cross-linguistically valid phonetic difference between [kw] and [kʷ], whether a language has one or the other is more of a phonological question. The [k] of [kw] will typically have anticipatory rounding due to the fact that readjusting the lips takes some time, and [kʷ] will typically have a [w]-like off-glide (maybe depending on the following vowel) due to the fact that readjusting the tongue and lips takes some time. This is not to say that a language couldn't distinguish /kʷ/ and /kw/ through some means, for example by timing or type of lip rounding (although I don't know of any such examples).

The obvious example is PIE *h₁ek̂wos 'horse' where 'k̂w' is not kʷ.

I have a suspicion that in my idiolect, court and quart have different initials, which is definitely not /kw/.

Weren't the "dorsal" series of Proto-Indo-European probably something like *[c ɟ ɟʰ]? I would expect */k̂w~cw/ to be phonoetically something like *[cɥ~c^ɥ] (I've also encountered the theory that */k̂ k/ were *[k q], but I think *[c k] a bit more likely, given that one yields affricated reflexes in some varieties), rather than having */k̂w/ a sequence differentiated only from */k̂^w/ by the degree of coarticulation.

Rounin Ryuuji wrote: ↑Sat Sep 04, 2021 6:17 pm Weren't the "dorsal" series of Proto-Indo-European probably something like *[c ɟ ɟʰ]? I would expect */k̂w~cw/ to be phonoetically something like *[cɥ~c^ɥ] (I've also encountered the theory that */k̂ k/ were *[k q], but I think *[c k] a bit more likely, given that one yields affricated reflexes in some varieties), rather than having */k̂w/ a sequence differentiated only from */k̂^w/ by the degree of coarticulation.

The problem with that is that */k̂/ was more common than */k/, if I recall correctly, which is consistent with if they were *[k q], whereas typologically having *[c] be more common than *[k] does not seem right to me.

But would a chain shift of *[k q] > *[c k] be terribly plausible? Certainly a shift of */c/ > */k/, or */q/ > */k/ on its own would be, but I find the chain shift that would otherwise be required for */k̂/ to various fricatives and affricates to occur... strange? I've never seen such a change attested, but perhaps it is in some language with which I'm unfamiliar.

Rounin Ryuuji wrote: ↑Sat Sep 04, 2021 8:02 pm But would a chain shift of *[k q] > *[c k] be terribly plausible? Certainly a shift of */c/ > */k/, or */q/ > */k/ on its own would be, but I find the chain shift that would otherwise be required for */k̂/ to various fricatives and affricates to occur... strange? I've never seen such a change attested, but perhaps it is in some language with which I'm unfamiliar.

Turkish, also iirc some of Athabaskan?

There seem to be a few PIE roots with initial *kw or *gw; the only one to survive the sceptics at Wiktionary is *kwep with notions of 'steam, odour, boil'.

There's some discussion of unconditioned palatalisation in a previous incarnation at origin of Arabic /ɟ/ (plus centum/satem musings).

To me at least, unconditional palatalization, which is known to happen as in the Arabic example, is much more plausible than unconditional depalatalization to velars, which is required for centum languages if one assumes that */k̂/ was [c]. Palatalization in general is far, far more common than backing of palatals to velars.

These are all fair and actually quite convincing points; I suppose my range of knowledge of sound changes simply wasn't extensive enough.

zompist wrote: ↑Wed Sep 01, 2021 4:49 pmWhich makes me wonder... you emoticon users, would you write "an ", "a ", etc.?

I only ever see (and use) emoji as "illustrations", e.g. "I know you like the mouse living downstairs ". They don't replace words.

Historical question now. I saw this in the Wikipedia article on "Syllable":

There are many arguments for a hierarchical relationship, rather than a linear one, between the syllable constituents. One hierarchical model groups the syllable nucleus and coda into an intermediate level, the rime. The hierarchical model accounts for the role that the nucleus+coda constituent plays in verse (i.e., rhyming words such as cat and bat are formed by matching both the nucleus and coda, or the entire rime), and for the distinction between heavy and light syllables, which plays a role in phonological processes such as, for example, sound change in Old English scipu and wordu.^{[10][further explanation needed]}

What happened to scipu and wordu that exemplifies a distinction between heavy and light syllables? Is the example perhaps misguided?

Kuchigakatai wrote: ↑Wed Nov 10, 2021 4:09 pm What happened scipu and wordu that exemplifies a distinction between heavy and light syllables? Is the example perhaps misguided?

The deletion of word-final u in Old English was weight sensitive. U was retained in scipu, where it was preceded by a single light syllable, but deleted in wordu > word, where it was preceded by a heavy syllable. Deletion also applied in stems consisting of two light syllables (which are not that common iirc).

Estav wrote: ↑Thu Nov 11, 2021 11:01 amThe deletion of word-final u in Old English was weight sensitive. U was retained in scipu, where it was preceded by a single light syllable, but deleted in wordu > word, where it was preceded by a heavy syllable. Deletion also applied in stems consisting of two light syllables (which are not that common iirc).

Oh I see. So Old English at some point had "scipu" but "word". Is "wordu" attested at all, or is it simply reconstructed?