Linguistic Miscellany Thread

[Salmoneus]

Which is interesting because etymologically they're the same word even in English. Spelling pronunciation can be subtle. I have them the same, but i think if someone said the word in isolation i would be able to guess which one they meant from intonation. As is the case with wood/would.

I just looked that up and it's true. It seems the monosyllabic pronunciation is the older one, and the disyllabic one is the innovation. But yes, spelling pronunciation has definitely reinforced the two separate pronunciations.

Coincidentally, however, I just happened to be reading some poems by Greville, and in one (XCIV) he specifically writes flow'rs. Greville makes a point of using apostrophes when his metre demands it - so for instance the -ed ending is usually <ed>, and presumably /@d/, except where the metre requires <'d>, presumably /d/. This seems strongly to suggest to me that, for Greville, the disyllabic pronunciation of "flower" is at least known, and probably the standard, otherwise he wouldn't have bothered specifically noting the absence of the second syllable in this poem - he's not a writer who just throws in apostrophes wherever a vowel has dropped out regularly.

Greville* is apparently believed to have "started" his cycle in the 1580s, and must have finished it before he died in 1628, suggesting that the disyllabic pronunciation must have developed rather quickly in Early Modern English (Greville's generally a conservative writer in his diction, I think). So it may be an 'innovation', but it's not a recent one!


*Fulke Greville, later 1st Baron Brooke; 1554-1628. Chancellor of the Exchequer and poet.

[Pabappa]

Looking for a word for sound symbolism that isn't onomatopoeia...like how "moon" has /u/ because the moon is round....I saw it just recently but cant remember....it's not "iconic" or "phonesthetic" either. Something that's used as a noun... "moon is a _____" .
thanks

[Ryusenshi]

An interesting note from a translation of Hesiod's Theogony (quoted from memory).

Whoever pours it in libation and swears a false oath falls in a coma* for a completed year.
* Translator's note: the word "coma" seems weirdly anachronistic... even though the original text uses the exact same word.

This is a common problem when translating from ancient texts: some words (here, κῶμα) have taken a more specialized meaning since then. In this case, do you use the same word anyway (here, "coma")? Or do you say something more generic (most translations say "lies breathless")?

Bible translations face a similar dilemma. How do you translate βάπτισμα? The easiest option is to use the direct cognate "baptism". But, in English, this word means a specific ceremony; while the Greek word originally meant "immersion". Similarly, the English word resurrection specifically means "raising from the dead", but in Latin the word simply meant "getting up".

[Ryusenshi]

What exactly is the difference between clausal subordination and coördination? What makes conjunctions like "yet" and "for" coördinating and those like "because" and "although" subordinating?

I wondered about that back in high school, but didn't dare ask the teachers.

Indeed, at first glance, they seem similar:
I arrived late because I overslept. / I arrived late for I overslept. (the second one seems a bit "literary" but is otherwise correct)
I arrived in time although I overslept. I overslept yet I arrived in time.

One of the reasons, I think, is that clauses with "because" can be moved around and used as complements in another clause:
Because I overslept, I arrived late.
The reason I arrived late is because I overslept.
You can't say:
*For I overslept, I arrived late.
*The reason I arrived late is for I overslept.

Similarly:
Although I overslept, I arrived in time.
*Yet I arrived in time, I overslept.
I'm not sure a clause with "although" can be used as a verb object. Maybe it can work in some sentences, and I'm just not finding them.

[zompist]

I think you're on the right track, Ryusenshi.

There's a sort of gradation from conjunctions to prepositions, and all these words are in between somewhere. I don't have time to work it all out right now, however!

[malloc]

Well, I have been grappling with the problem of subordinate clauses in my main conlang. The language is thoroughly head-marking and polysynthetic with pronominal suffixes on heads agreeing with their dependents. This system faces extraordinary complications with multiple clauses in one sentence and especially subordination. If subordinating conjunctions function like other heads in the language, they ought to agree with their dependents, meaning the verb or clause they are subordinating. But verbs have none of the features, like animacy and obviation, that normally determine agreement in this language. The language would need something like verbal "gender" to distinguish multiple verbs for agreement purposes, which seems both unprecedented in linguistics and nightmarish in complexity.

[Zju]

Looking for a word for sound symbolism that isn't onomatopoeia...like how "moon" has /u/ because the moon is round....I saw it just recently but cant remember....it's not "iconic" or "phonesthetic" either. Something that's used as a noun... "moon is a _____" .
thanks

Visual onomatopoeia maybe? IIRC I've heard the takete/bouba effect being referrerd to in this way.

[Pabappa]

Thanks for your quick reply, ... id forgot id asked ..... i now remember the word I actually wanted was ideophone. Onomatopoeia is just a subset of this.

[Nerulent]

Well, I have been grappling with the problem of subordinate clauses in my main conlang. The language is thoroughly head-marking and polysynthetic with pronominal suffixes on heads agreeing with their dependents. This system faces extraordinary complications with multiple clauses in one sentence and especially subordination. If subordinating conjunctions function like other heads in the language, they ought to agree with their dependents, meaning the verb or clause they are subordinating. But verbs have none of the features, like animacy and obviation, that normally determine agreement in this language. The language would need something like verbal "gender" to distinguish multiple verbs for agreement purposes, which seems both unprecedented in linguistics and nightmarish in complexity.

Is there a reason that conjunctions can't just be exceptional in this regard? Especially if there just isn't a verbal category for them to agree with.

Agreement with a subordinate verb phrase would be pretty cool though; any verbal category such as voice, mood, tense, plurality, activeness (stative verbs or experiencer verbs or even verbs of motion often behave slightly differently in some languages) would be up for agreement. It doesn't necessarily have to agree with the verb over, say, the subject either.

[Space60]

"Isn't it?" vs. "Is it not?" Why is the order different for the contracted and noncontracted versions?

It's cold, isn't it?

It is cold, is it not?

*It is cold, is not it?

*It's cold, is itn't?

[Travis B.]

"Isn't it?" vs. "Is it not?" Why is the order different for the contracted and noncontracted versions?

It's cold, isn't it?

It is cold, is it not?

*It is cold, is not it?

*It's cold, is itn't?

It is because -n't is an inflection and not a clitic, so it moves with the word it is attached to.

[Travis B.]

I have noticed that in a number of German names around here, including the Milwaukee street name Teutonia, <eu> is pronounced /aɪ/ (or properly [əe], since none of these cases that I am aware of precede lenis consonants or vowels). Is there a reason for this, as if the pronunciation was taken directly from Standard German one would expect /ɔɪ/, and if the pronunciation was a simple English spelling pronunciation one would expect /ju/? It almost reminds me of the pronunciation of Dutch <ui> in English as /aɪ/.

[Nortaneous]

Palatinate German has eu > ei. (Pennsylvania German preserves this.)

[Pabappa]

It is because -n't is an inflection and not a clitic, so it moves with the word it is attached to.

reminds me a bit of how English 'd behaves like a PoS all it's own. E.g. "I really'd rather not"... it seems to go after the most convenient vowel final word.

[missals]

Does anyone happen to know of any languages that permit only one particular consonant in the coda, other than /n/ or an archiphonemic nasal? (like in Japanese?)

E.g. a language that only allows /s/ in the coda. I do know there's Iau with its word-final [p̚] < /f/, but that's very unusual if I'm not mistaken. Is there anything in the middle range of commonality?

[Estav]

Does anyone happen to know of any languages that permit only one particular consonant in the coda, other than /n/ or an archiphonemic nasal? (like in Japanese?)

E.g. a language that only allows /s/ in the coda. I do know there's Iau with its word-final [p̚] < /f/, but that's very unusual if I'm not mistaken. Is there anything in the middle range of commonality?

Japanese also allows "Q" as a coda (an obstruent that is identical to the onset of the following syllable); that seems plausible as another possible candidate for the only coda consonant. Apparently, in word-internal position coda consonants are only permitted in Trukese as the first part of geminates, although Trukese does have word-final consonants (which apparently act as "extraprosodic" for the purpose of minimum word-length requirements, and are analyzed as resulting from a process of word-final vowel deletion) ("On the Representation of Initial Geminates", Stuart Davis, Phonology (1999)). Trukese also allows word-initial geminates, so I'm not sure how we can tell that word-medial geminates are divided between syllables.


I know that there are languages that have geminates but not nasal-obstruent clusters (e.g. Inuit; although that isn't an answer to the original question).

Only having /h/ as a coda consonant also seems somewhat plausible to me.

[Vijay]

I think Chatino (or at least the San Juan Quiahije variety) only allows glottal stops as syllabic codas.

[Frislander]

I can't recall such languages off the top of my head, but I have some near examples.

Rotinese reportedly only permits /k/ and /s/, both of which apparently by a process of suffixation, and there might be other ones in the Austronesian family. Also Ancient Greek had /s n r/ only in the coda, so I can imagine a variety which assimilates /n r/ into geminates would lead to /s/ being the only surface coda consonant.

[Linguoboy]

I have noticed that in a number of German names around here, including the Milwaukee street name Teutonia, <eu> is pronounced /aɪ/ (or properly [əe], since none of these cases that I am aware of precede lenis consonants or vowels). Is there a reason for this, as if the pronunciation was taken directly from Standard German one would expect /ɔɪ/, and if the pronunciation was a simple English spelling pronunciation one would expect /ju/? It almost reminds me of the pronunciation of Dutch <ui> in English as /aɪ/.

Im Deutschen gilt Entrundung der mittelhochdeutschen Umlaute „ö“, „ü“ und „eu/äu“ zu „e“, „i“ und „ai“ in den meisten Dialekten des Oberdeutschen und des Mitteldeutschen.

[Nortaneous]

Some reconstructions of Tangut have -w as the only coda consonant.