Linguistic Miscellany Thread

[M Mira]

As a native Mandarin speaker, I see it as a rather straightforward derivation from the original sense "family". It went from "a household practicing a certain trade" to "a person practicing a certain trade". Depending on the nature of the trade, some retain the former meaning more, while some have drifted all the way to the latter.

Farming is strongly tied to the household, so "farm"+"house" is still primarily "farming family":
https://en.wiktionary.org/wiki/農家

Shops were, until this decade, strongly tied to a physical building, but not really to families, so it can mean both shopkeepers or businesses.
https://en.wiktionary.org/wiki/商家

Boats aren't really tied to either a family or a house, so it refers to the person.
https://en.wiktionary.org/wiki/船家

Painting is a craft linked to only personal skills, so it refers to the person.
https://en.wiktionary.org/wiki/畫家

Additionally, there's some sort of "anonymity" with #1~3: you can say "X gave me a warm reception", "X (collective) are unhappy about the gas price hike" for all four, but "He is a X" only works for #4.

[Moose-tache]

So to reiterate, if you have a compound word and one element ceases to exist independently, you technically have a bound morpheme. But you can still retain all the trappings of a "real" word, such as using kanji instead of kana, an entry in the dictionary, etc. In reality there are going to be plenty of morphemes on a spectrum from "still a transparent compound" to "grammatical or derivational affix." For example, in English:

blackberry - still a stand alone word with the same meaning. Interestingly snozberry might fit here, since anyone familiar with snozberries could probably parse a sentence that contained the phrase "a snoz."

cranberry - pretty clearly derived from crane, which is still a stand alone word, but here its meaning is not obvious and its pronunciation is conditioned by environment. The "memberberries" from Southpark might fit here as well.

wereberry - (a human-berry hybrid subject to lunar transformations) historically "were" would have been a stand alone word, but not only has it disappeared in that sense, its second life as a compound element is only tangentially related to its original meaning.

re-berry - (the act of replacing berries on a bush that have been lost) in this case the productive prefix re is analogized from Latin, where it derived from a noun res centuries ago. The original noun is not present in English in anything approaching a predictable form or meaning. The tendency to use a hyphen suggests that English speakers understand the productive nature of the prefix, not present in compounds like "mulberry."

As it happens, 2 through 4 are bound morphemes. We could probably draw a line anywhere along the list and give that group a fancy name like "de-nominal elemental fragments" or something, but if you want to use existing terminology like bound morpheme, you should use it the way it is used in linguistics, which will not separate 2 through 4 into separate categories based on etymology or other concerns.

[Raphael]

Completely unrelated: Are there direct or indirect loanwords from Latin in Modern Greek, and if so, how many?

[Pabappa]

Completely unrelated: Are there direct or indirect loanwords from Latin in Modern Greek, and if so, how many?

i think so, yes. Greek tries its hardest to come up with new words out of native stock ... panepistimio etc ,,,, but has some loans from Latin that came in at a time when they werent as proud. spíti "house", saíta "arrow" etc. those are just the ones that are for common objects. there are plenty of other words that are culturally specific to Rome, I imagine.

[Travis B.]

Rhomaioi

[dɮ the phoneme]

With Japanese it's even trickier, since the same kanji is used for what (to a linguist) are completely unrelated words. I mean, do you really think Japanese consider the -ka morpheme to be the "same morpheme" as ie or uchi 'house', though it means something completely different, just because they're both written with 家? I mean, maybe they do, but maybe they just shrug and treat them as a graphic pun, like the fact that X means both "multiply" and "Christ".

Anyway, the issue is a reminder that in most Western languages, 'word' is an everyday concept, but this doesn't correspond to an everyday concept in Chinese. (I don't know how it goes in Japanese.) Neither 字 zi (as in kanji) nor 词 ci quite corresponds to 'word' or to any linguistic concept.

In Japanese, at least in my experience, 'word' defintely is an everyday concept, probably since Japanese morphology makes (most) word boundaries pretty obvious. And certainly, I've never met a speaker who considers such disparate uses of the same Kanji as -ka and ie or uchi to be "the same word." It would seem to me that /ka/ <家> is unambiguously a bound morpheme.

[Raphael]

Completely unrelated: Are there direct or indirect loanwords from Latin in Modern Greek, and if so, how many?

i think so, yes. Greek tries its hardest to come up with new words out of native stock ... panepistimio etc ,,,, but has some loans from Latin that came in at a time when they werent as proud. spíti "house", saíta "arrow" etc. those are just the ones that are for common objects. there are plenty of other words that are culturally specific to Rome, I imagine.

Ah, thank you.

[Vijay]

I'm not sure whether this is the first time this has happened, but I just realized this morning that this Nepali movie song is in the format of a ghazal. The hamd (i.e. rhyme in the first line and every even-numbered line after that) is -e timilai (-ए तिमीलाई):
https://www.youtube.com/watch?v=fMfkAGZ8wbE

नबिर्से तिमीलाई नपाए तिमीलाई
बिना अर्थ दिलमा सजाए तिमीलाई
Nabirse timilai napāe timilai
Binā artha dilmā sajāe timilai

त्यो यात्रा सुनौलो अनि साथ तिम्रो
सम्झेर भेट्न बोलाए तिमीलाई
Tyo yātrā sunaulo ani sātha timro
Samjhera bheṭna bolāe timilai

आयौ समिपई जब तिमी निदरीमा
यो हलचल यो धड्कन सुनाए तिमीलाई
Āyau samipai jaba timi nidarimā
Yo halchal yo dhaḍkan sunāe timilai

रमाए एकैछिन पछि आसुँ झर्यो
अनि आसुँ पुछ्दै पठाए तिमीलाई
Ramāe ekaichhin pachhi ā̃su jharyo
Ani ā̃su puchhdai paṭhāe timilai

भयो आश रित्तो र यो मन निराश भो
अनि बेहोसीमै कराए तिमीलाई
Bhayo ās ritto ra yo man nirās bho
Ani behosimai karāe timilai

अनायासै मेरो दुख्यो जीन्दगीमा
परेली भिजाई बगाए तिमीलाई
Anāyāsai mero dukhyo jindagimā
Pareli bhijāi bagāe timilai

[Qwynegold]

Just a little anecdote about something I experienced recently. I was watching TV and the host said that we are going to meet two people, one of whom was called Hugo. A little later two people were presented, but none was called Hugo. I didn't think much about it at first, but then I realised: I must've misheard. One of these people was called Rudolf. So I made an analysis of how this happened. The TV host spoke Scanian dialect, so she must've pronounced the name something like [ˈʁʉːdɔlf]. So I misheard [ʁ] as [h]; I've heard that some Brazilian Portuguese dialect has [h] for their /r/. And I misheard [d] as [g]; they're both voiced plosives, only the POA is different. And apparently I completely missed out on the last consonants. So this way we get [ˈʁʉːdɔlf] > [ˈhʉːgɔ].

[Salmoneus]

I'm not sure whether this is the first time this has happened, but I just realized this morning that this Nepali movie song is in the format of a ghazal.

Song lyrics have a strong tendency toward formalist poetry, presumably because the structures of music mirror and are complemented by the structures of poetic form. Indeed, it may well be that poetic forms are primarily a result of musical structures, rather than vice versa. Certainly, many specific forms are derived from songs (prominent examples include the mediaeval French virelai and the more recent English ballad).

[Vijay]

Yes, of course; in fact, there are South Asian movies where all of the songs are simply renditions of poems, like Vanaprastham in Malayalam where they're all excerpts from Kathakali performances. Some of these movies are about a poet and use only that poet's works as movie songs, such as Mirza Ghalib and the Kashmiri movie Habba Khatoon (1978) about the poet of the same name.

But while ghazals in particular are attested at least from Turkey to Nepal, I don't seem to have found any yet in South Asian languages other than Hindi/Urdu, Tamil(!), Punjabi, Gujarati, Nepali, Kashmiri, Sindhi, Pashto, and Persian (if that counts). There are a lot of songs that are called "ghazals" in Malayalam and even Malay, for example, but they don't seem to have the same format, and I suspect the motivation in these cases for calling such songs "ghazals" is based on other factors. (In the case of Malayalam, ghazals mostly if not all seem to be associated with the Muslim community in northern Kerala, and they also don't seem to be part of the Muslim folk tradition in that area. In the case of Malay, ghazals seem to refer to strongly Indian-influenced music).

[Travis B.]

From listening to people on the radio in things such as non-local commercials and syndicated programs, I get the impression that dialects other than that here are less different from that here than the transcriptions that people give of how they speak would indicate. I certainly hear final devoicing (as I have noted before), l-vocalization, affrication of /t/ before /w ʊ uː ər/, raising of /ɑː/ to [ʌ] before /r/ followed by fortis obstruents, and so on. This almost makes me wonder whether people are not giving accurate transcriptions of their own speech, or whether, conversely, it just turns out that a lot of people on radio are from the Inland North.

[linguistcat]

I have already looked and haven't found anything, but I don't want to assume it's impossible for this to exist.

So are there any languages in which long prenasalized stops exist, even arguably? If not, are there languages with long voiced stops? If not again, what would be the closest I could reasonably get to either of those?

[akam chinjir]

I have already looked and haven't found anything, but I don't want to assume it's impossible for this to exist.

So are there any languages in which long prenasalized stops exist, even arguably? If not, are there languages with long voiced stops? If not again, what would be the closest I could reasonably get to either of those?

The article on "Partially Nasal Segments" in The Blackwell Companion to Phonology (by Anastasia K. Riehl and Abigail C. Cohn) has a brief sectionon exactly this. I'll summarise.

They've argued that prenasalised stops are phonologically distinct from clusters; e.g., ⁿd is distinct from nd. They then mention some languages in which these appear to contrast, focusing on Sinhala, where you appear to get contrasts such as laⁿda "thicket" vs landa "blind." Now, Sinhala lets consonants (including voiced plosives) geminate, and it turns out that ⁿd patterns with singleton consonants and the ostensible nd patterns with geminates. For example, apparently pluralisation can involve geminating a consonant, so balala cat becomes balallu cats---and polə^ŋga viper becomes poleŋgu vipers. Which is to say that, phonologically speaking, ŋg looks like it's really ^ŋgː.

They generalise by saying that as far as I know, prenasalised stops appear to contrast with clusters only in languages with consonant length---mentioning as other examples Fula and Selayarese.

So it looks like there's a good case to be made for geminate prenasalised stops in at least some languages.

(And there are certainly languages with geminate voiced plosives.)

[linguistcat]

Thank you. That's perfect, and I'm really happy to hear about this. I didn't want to try and puzzle out what else to do with those.

[Linguoboy]

Both the word nawr and its palindrome rŵan mean "now" in modern Welsh.

Etymologically, both words contain awr "hour". Nawr is an aphetic spoken variant of yn awr and rŵan derives from yr awr hon "this hour".

[mae]

-

[malloc]

This probably sounds dumb, but perhaps because the heart is considered the source of emotion and emotions are mercurial, prone to change at a moment's notice. That would be my guess anyway.

[Zaarin]

This probably sounds dumb, but perhaps because the heart is considered the source of emotion and emotions are mercurial, prone to change at a moment's notice. That would be my guess anyway.

The heart is not universally regarded as the seat of emotion. I have no idea about Arabic, but in its cousin Biblical Hebrew the kidneys were the seat of emotion.

[Linguoboy]

This probably sounds dumb, but perhaps because the heart is considered the source of emotion and emotions are mercurial, prone to change at a moment's notice. That would be my guess anyway.

The heart is not universally regarded as the seat of emotion. I have no idea about Arabic, but in its cousin Biblical Hebrew the kidneys were the seat of emotion.

IIRC, in South and East Asia, it’s usually the liver.