Zompist Bboard Again

Akangka wrote: ↑Sat Jun 15, 2019 7:06 am I wonder why does my previous mandarin speaker says "b dibaca p dan p dibaca b" (b is pronounced as p and p is pronounced as b). Even though they're actually /p/ and /pʰ/, respectively. Is it possible that /pʰ/ is registered as lenis one instead of /p/. Or is it just a byproduct that javanese pronounce /b/ as slack voiced and /p/ as stiff voiced?

According to Wikipedia:

The Javanese "voiced" phonemes are not in fact voiced but voiceless, with breathy voice on the following vowel.[7] The relevant distinction in phonation of the plosives is described as stiff voice versus slack voice.[8]

Or do you have a sample video with the two consonants?

Space60 wrote: ↑Fri Jun 14, 2019 2:06 pm

Vijay wrote: ↑Fri Jun 14, 2019 2:00 pm What about a bathroom? Or toilet or water closet or something?

For me, the "toilet" is the device inside the room.

It is for me, too! But it's not for a lot of non-Americans, at least.

Qwynegold wrote: ↑Sat Jun 15, 2019 9:55 am*English is complicated, because even though I can come up with words that contain some of these sequences (like hamster or angry), I don't know how common it is for speakers to insert an epenthetic consonant in the middle, like [hæmpstɚ], [æŋɡrɪ].

I don't know of such a resource, sorry...but I will say that at least many English speakers (myself included) probably insert an epenthetic /p/ in words like "hamster," and...as far as I'm aware, /ŋɡr/ is the only acceptable cluster in "angry." I've certainly never heard anyone say it with [ŋɹ], and it is also historically derived from "anger" plus a suffix, so I would imagine it has always had /ŋɡ/.

Qwynegold wrote: ↑Sat Jun 15, 2019 9:55 am I'm looking for info on phonotactics of various languages, but can't find anything detailed. Specifically I want to find out if the following consonant sequences are allowed (medially): mn, mt, mk, mtʃ*, ms, mh, mr, nh, ŋm, ŋn, ŋp, ŋt, ŋtʃ*, ŋf, ŋs, ŋr.

I think these are all OK in English:

amnesia, empty, Aramco, Kamchatka, homesick, ham-handed, home-wrecker

inhale

ringmaster, hangnail, ping-pong, ringtone, song thrush, songfest, seamstress, ringwraith

Some of these may be phonetically assimilated-- e.g. "empty" is going to have a labiodental m.

Many of these could probably be improved; these were the first ones I found for each combo...

Anyone know of a language or languages with prefixed past tense(s) but suffixed future(s)?

(A mood marker that can be used with a future sense would also work. I'm not really bothered if the markers are particles or clitics rather than affixes. I'd be especially interested in any such language that also has an overt present tense of some sort.)

zompist wrote: ↑Sat Jun 15, 2019 4:37 pm Some of these may be phonetically assimilated-- e.g. "empty" is going to have a labiodental m.

"Empty" for me is ['ɛmti~'ɛmpti]; the /m/ definitely isn't a labiodental. But there's also hometown.

dhok wrote: ↑Sun Jun 16, 2019 1:50 pm

zompist wrote: ↑Sat Jun 15, 2019 4:37 pm Some of these may be phonetically assimilated-- e.g. "empty" is going to have a labiodental m.

"Empty" for me is ['ɛmti~'ɛmpti]; the /m/ definitely isn't a labiodental. But there's also hometown.

Well, today I agree with you. When I was compiling the list it seemed like some of the /m/'s might not be [m]'s. A labiodental in 'empty' doesn't feel wrong or strange to me, but right now the labial seems to predominate.

Empty is also [ˈɜ̃m(p)ti(ː)] for me, with no labiodental.

akam chinjir wrote: ↑Sun Jun 16, 2019 3:16 am Anyone know of a language or languages with prefixed past tense(s) but suffixed future(s)?

(A mood marker that can be used with a future sense would also work. I'm not really bothered if the markers are particles or clitics rather than affixes. I'd be especially interested in any such language that also has an overt present tense of some sort.)

Early Indo-European languages with augments worked rather like that. The augment was an (optionally) prefixed indicator of past tense, while the future was marked by a suffix. The pattern was spoilt a bit by the sigmatic aorist. The -i of the singular of the primary tenses is generally interpreted as an overt present tense marker.

Thanks Richard W! Been a long time since I've gotten lost in Sanskrit verbs...

Qwynegold wrote: ↑Sat Jun 15, 2019 9:55 am I'm looking for info on phonotactics of various languages, but can't find anything detailed. Specifically I want to find out if the following consonant sequences are allowed (medially): mn, mt, mk, mtʃ*, ms, mh, mr, nh, ŋm, ŋn, ŋp, ŋt, ŋtʃ*, ŋf, ŋs, ŋr.
*Or some other coronal affricate.

Spanish: mn in indemnizar, nx in enjambre. All the others are impossible within a word due to nasal assimilation for the point-of-articulation of the next consonant (mn and nm (as in inmaduro) are the main exceptions).

Spanish has some words ending in -m, which makes all the missing mC clusters possible across words: mt in el réquiem terminó, mk in el réquiem cambió, mθ in con el réquiem celebraron, mtʃ in el réquiem chino, ms in el réquiem solemne, mx in el réquiem gitano, mr in el réquiem real.

There's no /ŋ/ so the ŋC clusters do not apply.

Answering the question for Swahili is pretty easy, all because of the m- prefix.

mnazi '(coconut) palm' (cf. nazi 'coconut'), mtaalamu 'expert', mkuki 'spear', mchai 'tea plant' (cf. chai 'tea'), mshumaa 'candle', msoma 'reader', mhenga 'ancestor', mraba 'square'.

I couldn't find any common word with the cluster nh, but there are some place names in Tanzania with it (Nhundulu, Nhobora, Nhinhi), which I think are relevant.

Swahili has a distinct /ŋ/, but it only appears before a vowel or before /k/, not seemingly occurring in word-final position.

German allows all of these across morpheme boundaries and/or in recent borrowings. /mn mt ms/ are also found in synchronically monomorphemic words (here marked with °), but most of the best-known examples are actually older loans that have been nativized. /mt ms ŋt ŋs/ are special in that they frequently occur in native words for morphological reasons, including word-finally, and sometimes even in monomorphemic contexts. /mr ŋr/ can occur in derivations as a result of medial syncope. Also note that [tʃ] is not clearly a full unitary phoneme in German, at least not in onset position, but /ts/ is (usually written <z>).

demnach, Geheimnis, Gymnastik°, Hymne°
gesamt(e), bestimmt(e), Ämter, Amt°, kommt
Gemeinsamkeit, umkehren, Dummkopf
demzufolge, Umzug
demselben*, umsonst*, Umsatz*, Amsel*°, (des) Raums
namhaft, umher, Baumhöhle
umringen, Baumreihe, dämmrig (cf. Dämmerung)
Vergangenheit, Wohnhaus, Anhöhe, anhand
engmaschig, Ringmauer
Stellungnahme, Gefängnis, Fangnetz
Engpass, Recyclingpapier
gelangt(e), drängt(e), bringt, singt
Langzeitgedächtnis, Sprungziel
geringfügig, Klangfolge, dingfest
langsam*, Mitbringsel*, allerdings, (des) Frühlings
umfangreich, hungrig (cf. Hunger)

*) <ms ngs> are pronounced [mz ŋz] in these words, but they doesn't contrast with *[ms ŋs].

zompist wrote: ↑Sat Jun 15, 2019 4:37 pmI think these are all OK in English:

amnesia, empty, Aramco, Kamchatka, homesick, ham-handed, home-wrecker

inhale

ringmaster, hangnail, ping-pong, ringtone, song thrush, songfest, seamstress, ringwraith

Most of these examples are at morpheme boundaries though. Would people do differently if the sequences were word-internal? I tried looking up some more words I thought of, but for example magnet and signal are pronounced with [gn] according to Wiktionary, and single and Anglo-Saxon with [ŋgl] (I'm assuming /ŋr/ would be pronounced similarly to /ŋl/).

Ser wrote: ↑Mon Jun 17, 2019 12:23 am Spanish: mn in indemnizar, nx in enjambre. All the others are impossible within a word due to nasal assimilation for the point-of-articulation of the next consonant (mn and nm (as in inmaduro) are the main exceptions).

Spanish has some words ending in -m, which makes all the missing mC clusters possible across words: mt in el réquiem terminó, mk in el réquiem cambió, mθ in con el réquiem celebraron, mtʃ in el réquiem chino, ms in el réquiem solemne, mx in el réquiem gitano, mr in el réquiem real.

There's no /ŋ/ so the ŋC clusters do not apply.

Oh, so the nasal in enjambre does not become [ŋ]? That's interesting. What does Spanish do with /ŋ/ in loan words?

cedh wrote: ↑Mon Jun 17, 2019 5:16 am dingfest

lol

Okay, thank you, and everyone else who's contributed. At least I have a little more data now.

Qwynegold wrote: ↑Mon Jun 17, 2019 12:13 pm

Ser wrote: ↑Mon Jun 17, 2019 12:23 amSpanish: mn in indemnizar, nx in enjambre. All the others are impossible within a word due to nasal assimilation for the point-of-articulation of the next consonant (mn and nm (as in inmaduro) are the main exceptions).

Spanish has some words ending in -m, which makes all the missing mC clusters possible across words: mt in el réquiem terminó, mk in el réquiem cambió, mθ in con el réquiem celebraron, mtʃ in el réquiem chino, ms in el réquiem solemne, mx in el réquiem gitano, mr in el réquiem real.

There's no /ŋ/ so the ŋC clusters do not apply.

Oh, so the nasal in enjambre does not become [ŋ]? That's interesting. What does Spanish do with /ŋ/ in loan words?

Notice I used slashes ("//"), not square brackets ("[]"), which means I was talking about the phoneme inventory, not phonetic realizations. I personally pronounce /nx/ variously as [nh], [ŋh] or [Ṽh] ([ẽˈhambɾe]).

Spanish speakers typically adapt foreign /ŋ/ [ŋ] as /n/. The Kong of Donkey Kong is typically pronounced /kon/ (although, to confirm this, I listened to some YouTube videos discussing Donkey Kong, and I heard one guy consistently pronounce it /ˈdonkinkon/ lol). People in Spain have borrowed the suffix -ing forming new words such as puenting /ˈpwentin/ 'bungee jumping'.

Allright, thanks!

abortion and orbation are anagrams and both relate to losing a child.

I think that irregularities in English nominal number can be classified accordingly:
- Irregular plurals of normal nouns (goose - geese, wife - wives, child - children).
- Identical singular and plurals (one sheep - three sheep).
- Strangely formed nouns with a modifier placed after (mothers-in-law, courts martial).
- Learnèd irregular plurals (crisis - crises, cherub - cherubim/cherubs, one ninja - three ninja/ninjas). For a lot of these there's a struggle between the learnèd plural and the regular plural, with different levels of acceptance for either (matrix - matrices and millennium - millennia are very accepted, campus - campi and stadium - stadia mostly occur as rare jokes).
- Academic abstract plurals (sociologies, wisdoms, stupidities).
- Singulars that look like plurals (darts the game, mathematics, the United States).
- British collective nouns (the team say that, the government say that).
- All other miscellaneous cases, sometimes involving separate plurals according to meaning (person/people/persons/peoples, hovercraft - hovercraft/hovercrafts, fish - fish/fishes, mouse - mice/mouses, datum/data as a count noun vs. data the mass noun).