(L&N) Q&A Thread - Quick questions go here

A forum for discussing linguistics or just languages in general.
User avatar
Sequor
sinic
sinic
Posts: 327
Joined: 30 Jun 2012 06:13

Re: (L&N) Q&A Thread - Quick questions go here

Post by Sequor »

I think languages with alienable vs. inalienable possession tend to just have it as a rather arbitrary property of categories of words, not so much subject to variation depending on the real-world situation. For example, Italian marks all family relationships (and only family relationships) as inalienable*, regardless of how close or far two relatives might be.

In Mandarin, inalienable possession is marked by being able to commonly (but optionally!) drop the particle 的 "de" between a pronoun and a noun. And yes, 家 jiā "house" is one of the nouns that can be used this way, as in 我家 wǒ jiā 'my house, my home'.


* The difference is found in whether the article is used with the possessive adjective (alienable) or not (inalienable). Contrast la tua casa 'your house' (alienable) vs. tua matre 'your mother' (inalienable).
hīc sunt linguificēs. hēr bēoþ tungemakeras.
Salmoneus
MVP
MVP
Posts: 2320
Joined: 19 Sep 2011 19:37

Re: (L&N) Q&A Thread - Quick questions go here

Post by Salmoneus »

My understanding is that - at least as regards Austronesian and I think it's broadly true elsewhere - inalienability is closely tied to mereology: inalienable possessions are generally real (body parts) or metaphorical (emotions, actions, social relationships) parts of the possessor. Sometimes, random non-part items can be inalienable if they're intensely associated with the individual - eg limes (carried on the person at all times for use in betel-chewing), or penis gourds.

On the one hand, houses seem like they might fit into the 'part of person' category, because they're certainly socially important, particularly if the 'house' is identified with the clan (eg with longhouses). On the other hand, I wonder if the reason they aren't (sfaia) inalienable in austronesia (where many cultures are very strong lifelong attachments to the clan house) is that the relationship goes the wrong way: the house is bigger than the person (both physically and metaphorically). Maybe it would be more likely to have "inhabitant" as an inalienable possession of a house, rather than vice versa? [though, of course, this may run into restrictions on who can own an inalienable]
Sequor wrote: 26 Mar 2021 15:33 I think languages with alienable vs. inalienable possession tend to just have it as a rather arbitrary property of categories of words, not so much subject to variation depending on the real-world situation. For example, Italian marks all family relationships (and only family relationships) as inalienable*, regardless of how close or far two relatives might be.
It's not usually dependent on closeness, no (not on a case by case basis - some languages do mark distant relationships as alienable but close ones as inalienable, on a lexical basis).

However, it often can vary depending on the real world situation! Austronesian possession is often sensitive to function (hence the possessive classifiers - 'my coconut for eating', 'my coconut for drinking', 'my coconut for use as an ornament', etc), and this can extent to demoting inalienables to alienables if the relationship is not of the prototypical type.

So, a language might distinguish "my kidney" (that I use to process fluids - direct) from "my kidney" (in a bag, that I am selling to an organ trader for cash". If a language DID have 'my house (that I live in)' as inalienable, it might nonetheless have an alternative alienable construction for "my house (that I rent out)".


EDIT: for example, Loniu apparently distinguishes inalienable "the female (of a species)" and alienable "the woman (belonging to me)".
* The difference is found in whether the article is used with the possessive adjective (alienable) or not (inalienable). Contrast la tua casa 'your house' (alienable) vs. tua matre 'your mother' (inalienable).
Oh, cool!
User avatar
Omzinesý
runic
runic
Posts: 3055
Joined: 27 Aug 2010 08:17
Location: nowhere [naʊhɪɚ]

Re: (L&N) Q&A Thread - Quick questions go here

Post by Omzinesý »

What is the history of Hungarian /ɟ/ <gy>?

I know egy 'one' is related to Finnish yksi. With this one word, I infer it derives from PU *kt, but does anybody have more information?
My meta-thread: viewtopic.php?f=6&t=5760
User avatar
All4Ɇn
mayan
mayan
Posts: 1652
Joined: 01 Mar 2014 07:19

Re: (L&N) Q&A Thread - Quick questions go here

Post by All4Ɇn »

Omzinesý wrote: 02 Apr 2021 15:02 What is the history of Hungarian /ɟ/ <gy>?

I know egy 'one' is related to Finnish yksi. With this one word, I infer it derives from PU *kt, but does anybody have more information?
I know that it also can come from proto-Uralic *j as in the case of gyökér from proto-Uralic *jᴕ̈kkɜ-rɜ as well as /sj/ as seen in the subjunctive stem for some verbs like eszik where it becomes egy-
User avatar
Xonen
moderator
moderator
Posts: 1095
Joined: 16 May 2010 00:25

Re: (L&N) Q&A Thread - Quick questions go here

Post by Xonen »

Omzinesý wrote: 02 Apr 2021 15:02 What is the history of Hungarian /ɟ/ <gy>?

I know egy 'one' is related to Finnish yksi. With this one word, I infer it derives from PU *kt, but does anybody have more information?
According to LSS Ánte (forthcoming), it's a medial reflex of PU *ď (whatever sound that actually was).

That's clearly not the only source, though; as you note, it appears to correspond to PU *kt in at least one word EDIT: apparently, egy does not in fact correspond to PU *ükti (or at least there's serious doubt about it), so scratch that. Still, there are words like húgy < *kuńśi and agy < *?ajŋi, so clearly at least some clusters do result in /ɟ/.
User avatar
Omzinesý
runic
runic
Posts: 3055
Joined: 27 Aug 2010 08:17
Location: nowhere [naʊhɪɚ]

Re: (L&N) Q&A Thread - Quick questions go here

Post by Omzinesý »

Xonen wrote: 03 Apr 2021 16:33
Omzinesý wrote: 02 Apr 2021 15:02 What is the history of Hungarian /ɟ/ <gy>?

I know egy 'one' is related to Finnish yksi. With this one word, I infer it derives from PU *kt, but does anybody have more information?
According to LSS Ánte (forthcoming), it's a medial reflex of PU *ď (whatever sound that actually was).

That's clearly not the only source, though; as you note, it appears to correspond to PU *kt in at least one word EDIT: apparently, egy does not in fact correspond to PU *ükti (or at least there's serious doubt about it), so scratch that. Still, there are words like húgy < *kuńśi and agy < *?ajŋi, so clearly at least some clusters do result in /ɟ/.
Thanks, both of you.

I could imagine that, because all the other Hungarian voiced stops derive from nasal + stop, /ɟ/ could also derived from a palatal nasal + stop-ish cluster like kuńśi.
Still waiting for Ánte's etymological dictionary :)
My meta-thread: viewtopic.php?f=6&t=5760
User avatar
sangi39
moderator
moderator
Posts: 2786
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

Re: (L&N) Q&A Thread - Quick questions go here

Post by sangi39 »

Not a quick question really, but a friend came to me with this:

Image

... asking if I might know anyone who might be able to translate it (assuming they could read the handwriting).

We're both fairly sure it's some form of Italian (probably Neopolitan?) from the late 16th Century (given the date at the top of the page), but beyond "seems to be a financial agreement between some number of parties". including payment for things lake damages (maybe?) we couldn't get very far.

I'm pretty sure we have some Italian speakers on the Board, so I thought "eh, let's ask the Board". So, if anyone might be able to help, that'd be a big thank you from me [:)]


Interesting side note, I didn't realise that, at one point, some Italian writers employed an umlauted <y> for <ij> (hell, I wasn't even sure <j> was even common enough a thing to warrant that), given that the text included "sijno" which was apparently replaced by "siano" in modern standard Italian (thank you Reddit for that nugget).
You can tell the same lie a thousand times,
But it never gets any more true,
So close your eyes once more and once more believe
That they all still believe in you.
Just one time.
User avatar
Omzinesý
runic
runic
Posts: 3055
Joined: 27 Aug 2010 08:17
Location: nowhere [naʊhɪɚ]

Re: (L&N) Q&A Thread - Quick questions go here

Post by Omzinesý »

Can Swedish have the sentence: Stenen såg pojken. be interpreted 'The boy saw the stone.'? My understanding is that Swedish word order is somewhat "freer" than that of English. Or must it be understood in the unnatural sense that the stone saw the boy?
My meta-thread: viewtopic.php?f=6&t=5760
User avatar
k1234567890y
mayan
mayan
Posts: 2368
Joined: 04 Jan 2014 04:47
Contact:

Re: (L&N) Q&A Thread - Quick questions go here

Post by k1234567890y »

I just found that the Lithuanian word for "boa" is smauglys, and I suspect that it is a cognate with Proto-Slavic *smokъ "dragon" and Old Norse smjúga "to creep through".

Is there information about the etymology of this word?
I prefer to not be referred to with masculine pronouns and nouns such as “he/him/his”.
User avatar
sangi39
moderator
moderator
Posts: 2786
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

Re: (L&N) Q&A Thread - Quick questions go here

Post by sangi39 »

k1234567890y wrote: 10 Apr 2021 13:59 I just found that the Lithuanian word for "boa" is smauglys, and I suspect that it is a cognate with Proto-Slavic *smokъ "dragon" and Old Norse smjúga "to creep through".

Is there information about the etymology of this word?
The only thing I've been able to find so far is that it might come from smaugti or smaugyti, which both mean "choke" or "strangle", with some suggestion that it might have originally means "to choke with smoke" (etymonline lists it as a cognate of English "smoke"), and some Lithuanian documents I've find containing smauglas (as a nouns) end up with Google translating this as "choke" as well (-smaug- appears in verbs like pasmaugti, "strangle" as well"). The Lithuanian word for "horse", arklys comes from arklas, "plough" and an agentive suffix, -ys, so it might be that smauglys comes from the same place, i.e. smauglas+ys which would make it something like "choker" (which, interestingly, is also where Russian удав, "boa constrictor" comes from, i.e. удавить, "to strangle")
You can tell the same lie a thousand times,
But it never gets any more true,
So close your eyes once more and once more believe
That they all still believe in you.
Just one time.
User avatar
k1234567890y
mayan
mayan
Posts: 2368
Joined: 04 Jan 2014 04:47
Contact:

Re: (L&N) Q&A Thread - Quick questions go here

Post by k1234567890y »

sangi39 wrote: 10 Apr 2021 16:06
k1234567890y wrote: 10 Apr 2021 13:59 I just found that the Lithuanian word for "boa" is smauglys, and I suspect that it is a cognate with Proto-Slavic *smokъ "dragon" and Old Norse smjúga "to creep through".

Is there information about the etymology of this word?
The only thing I've been able to find so far is that it might come from smaugti or smaugyti, which both mean "choke" or "strangle", with some suggestion that it might have originally means "to choke with smoke" (etymonline lists it as a cognate of English "smoke"), and some Lithuanian documents I've find containing smauglas (as a nouns) end up with Google translating this as "choke" as well (-smaug- appears in verbs like pasmaugti, "strangle" as well"). The Lithuanian word for "horse", arklys comes from arklas, "plough" and an agentive suffix, -ys, so it might be that smauglys comes from the same place, i.e. smauglas+ys which would make it something like "choker" (which, interestingly, is also where Russian удав, "boa constrictor" comes from, i.e. удавить, "to strangle")
ok thanks for help (:
I prefer to not be referred to with masculine pronouns and nouns such as “he/him/his”.
User avatar
KaiTheHomoSapien
greek
greek
Posts: 555
Joined: 15 Feb 2016 06:10
Location: Northern California

Re: (L&N) Q&A Thread - Quick questions go here

Post by KaiTheHomoSapien »

I know that I have asked this question before but I wanted to ask it again in light of a recent episode of the podcast Lingthusiasm:

How would you characterize a "rhotic" sound? I'm not asking for a scholarly definition, but your opinion or sense of what it is that rhotics all have in common, if anything. Is the sense of being "r-like" intuitively real but not phonetically real?

One definition on Wikipedia defines rhotics in terms of sonority: rhotics are less sonorant than vowels, but more sonorant than laterals. Would you agree with that? Is there another definition that works?
User avatar
eldin raigmore
korean
korean
Posts: 5842
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: (L&N) Q&A Thread - Quick questions go here

Post by eldin raigmore »

KaiTheHomoSapien wrote: 18 Apr 2021 19:48 I know that I have asked this question before but I wanted to ask it again in light of a recent episode of the podcast Lingthusiasm:

How would you characterize a "rhotic" sound? I'm not asking for a scholarly definition, but your opinion or sense of what it is that rhotics all have in common, if anything. Is the sense of being "r-like" intuitively real but not phonetically real?

One definition on Wikipedia defines rhotics in terms of sonority: rhotics are less sonorant than vowels, but more sonorant than laterals. Would you agree with that? Is there another definition that works?
The following notwithstanding; each language basically has its own “definition” of “rhotic”. (If it has a liquid which isn’t a lateral!)

Notwithstanding the above, there are several strong tendencies: to wit:

Rhotics raise the first formant and lower the second formant to be nearly equal to each other, while simultaneously raising the third formant and lowering the fourth formant to be nearly equal to each other.

They narrow one end or the other of the tract thru which the airstream passes. Retroflexes narrow the front end; uvulars narrow the back end.

They are similar in sonority to other liquids (including lateral liquids) and to glides or semivowels.
I know there’s a sonority order with respect to glides and with respect to lateral liquids but I think your statement about it might not be quite right?
There’s also a sonority order wrt nasals. I think all liquids are more sonorant than nasals but less sonorant than glides. I could be wrong, I guess.

Rhotic’s PoA tends to be either retroflex or uvular, OR, their MoA tends to be either approximant or trill.
IIANM there are lateral rhotics.
A trill might be classed as a rhotic even if it’s neither retroflex nor uvular.
A retroflex or uvular might be classed as a rhotic even if it’s neither a trill nor an approximant.
An approximant which isn’t a semivowel and isn’t a lateral is likely to be classed as a rhotic.
Any liquid which isn’t a lateral might be a rhotic.

.....

That’s what I think.

I suppose I could be wrong.
User avatar
KaiTheHomoSapien
greek
greek
Posts: 555
Joined: 15 Feb 2016 06:10
Location: Northern California

Re: (L&N) Q&A Thread - Quick questions go here

Post by KaiTheHomoSapien »

^Thanks for the answer. Although I admit I'm a little confused on what a "formant" is :wat:
Salmoneus
MVP
MVP
Posts: 2320
Joined: 19 Sep 2011 19:37

Re: (L&N) Q&A Thread - Quick questions go here

Post by Salmoneus »

Formants:

Sounds are waves of pressure in the air. But sounds in nature are not simply pure sine waves - they are very complicated waves. Which is to say, the pressure over time follows a complicated curve. But mathematically, complicated curves are equivalent to the combination of many simple curves. These curves are known as "partials" of the sound. Each partial has its own frequency (or, conversely, given a constant medium, each partial has its own wavelength). The frequencies of partials can be almost random, in the case of very 'noise-like' sounds. But some things - vibrating strings, and columns of air inside tubes open at at least one end, for instance, including the human mouth/trachea/etc - produce "harmonic" sounds, in which the strongest partials are actuall simple multiples of the lowest-frequency partial.

The obvious way to alter sound would be to change its frequency. But, probably because humans naturally have a wide range of 'natural' frequencies, because we differ in size, and hence the length of our air-tubes, frequency changes are actually only a minor part of human language ('tone'). Instead, the primary way we distinguish sounds is to use our air-tubes to 'resonate' in such a way as to amplify some partials, while diminishing others (we can also bend the partials away from where they mathematically ought to be, since our air-tubes are not perfectly uniform tubes, and are slightly flexible).

In linguistics, a "formant" is a strong partial. When we speak, our sounds can be mostly analysed (simplistically speaking) as a combination of two, three or four loud partials, called 'formants'. Although the absolute frequency of each partial varies from person to person, we can recognise the pattern of how the formants relate to each other. In [ i ], for instance, the first and second formants are very wide apart, while in [ u ] they are very close together. In general, vowels can be distinguished purely by the first two formants - "vowel height" is the frequency of the first (lowest) formant, while "frontness" is the frequency of the second formant. Consonants can then either alter the first two formants of the surrounding vowels, and/or do interesting things with the other two formants. Acoustically, most phonetic features are actually heard as patterns of formants.


---------

On rhotics:

Rhoticity is an example of what Wittgenstein termed 'family resemblance'. You know it when you hear it. Specifically, there is no one characteristic shared by all rhotics and no non-rhotics, just as there is no single feature that makes family members look alike. John and his brother perhaps share a distinctive nose, while John's daughter doesn't have his nose, but does have his chin, which John's brother doesn't have. John's brother and daughter thus both look like John, but there isn't a specific feature that makes them look like one another. And the family isn't defined by whether it looks like John - maybe John's brother looks more like their father than John does, while John looks more like their mother. So there's not necessarily a single defining feature of the family at all... but nonetheless, when you see the whole family in one photograph, it looks like a family - you say "oh yeah, they look like they're related!" - even though there may be two outlying members who actually don't objectively look much like each other, even though each sort of looks like a member of the family as a whole.

Similarly, there are some sounds that are almost definitely rhotics, and then other sounds that sound like those rhotics may also be considered rhotics, even if they don't have much in common. It's not an objective classification, but depends on context. So, for instance, because of the usual English /r/, a sound like a labiodental approximant may also be treated as a rhotic, as it sounds similar; but labiodental approximants are found in other European languages where they are not considered rhotics, because the archetypal rhotic in that language is quite distinct (a labiodental approximant can sound like a postalveolar approximant, but doesn't sound much like an alveolar trill). So rather than there being "rhotics" and "non-rhotics", there are more and less rhotic sounds, in the sense that some sounds are very likely to be considered rhotic (alveolar trills) while other sounds (labiodental approximants, uvular fricatives) are less likely to be considered rhotic in a given language, and then again others (bilabial stops) are almost certain not to be considered rhotic in any language.
User avatar
KaiTheHomoSapien
greek
greek
Posts: 555
Joined: 15 Feb 2016 06:10
Location: Northern California

Re: (L&N) Q&A Thread - Quick questions go here

Post by KaiTheHomoSapien »

Thank you [:D]

I can understand the "family resemblance" analogy. There are some sounds that are always considered rhotics (certain non-lateral liquids) and there are some sounds that are similar to those sounds (and maybe only differ by one feature) that can also be considered rhotics in some languages (even though they might not have much in common with each other). As they pointed out on Lingthusiasm there are some sounds that are considered rhotics in some languages but not in others, like the voiced uvular fricative which is an allophone of the voiced velar fricative in Arabic and is transcribed as <gh> and not thought of as "r-like" in that language.
User avatar
Omzinesý
runic
runic
Posts: 3055
Joined: 27 Aug 2010 08:17
Location: nowhere [naʊhɪɚ]

Re: (L&N) Q&A Thread - Quick questions go here

Post by Omzinesý »

<cc> seems to be pronounced [ks] in English and French, at least in some words like vaccine. It that a sound change between Latin and French?
My meta-thread: viewtopic.php?f=6&t=5760
Salmoneus
MVP
MVP
Posts: 2320
Joined: 19 Sep 2011 19:37

Re: (L&N) Q&A Thread - Quick questions go here

Post by Salmoneus »

Perhaps you'd like to look up some Romance sound changes? They're actually pretty well documented online!

In this case, yes, Latin /k/ was palatalised before front vowels, becoming either /ts/ (in the Gallic sphere) or /tS/ (elsewhere). The former often softened further to a fricative like /s/ or /T/. This meant that in a /kk/ cluster, the first /k/ wouldn't palatalise (no following front vowel) but the second could before front vowels, yielding /kts/, /ks/, /kT/, etc. The first /k/ could then in some languages either be dropped to simplify the cluster (Spanish and Portuguese have simply /s/ (or /T/)), or else assimilated in POA (Italian has /ttS/). In French, however, the simplification of /kts/ to /ks/ happened early enough that there was no stop to assimilate /k/ to, and the cluster was simple enough to avoid further simplification.

English is not actually directly descended from Latin, but often it has borrowed Latinate words from Latin, Anglo-Norman, or French; this has happened in this case, giving French-style /ks/ in this word, and in the general Anglo-Latinate pronunciation.

Irish, interestingly, borrowed the Latin word early enough that it didn't undergo palatalisation. [well, not the Romance assibilating palatalisation, it's a palatalised (or palatal) stop]
User avatar
Pabappa
greek
greek
Posts: 486
Joined: 18 Nov 2017 02:41
Contact:

Re: (L&N) Q&A Thread - Quick questions go here

Post by Pabappa »

are we sure that the words with /ks/ arent just all reborrowings? i would expect them to both palatalize and then degeminate, just as how as you say italian doesnt have any words with /ktʃ/.
I'll take the theses, and you can have the thoses.
Salmoneus
MVP
MVP
Posts: 2320
Joined: 19 Sep 2011 19:37

Re: (L&N) Q&A Thread - Quick questions go here

Post by Salmoneus »

Pabappa wrote: 26 Apr 2021 23:25 are we sure that the words with /ks/ arent just all reborrowings? i would expect them to both palatalize and then degeminate, just as how as you say italian doesnt have any words with /ktʃ/.
Why would you 'expect' that?

Give examples of cases where Latin /kk/ becomes /s/. If the /ks/ examples are only reborrowings, it should be easy to find a horde of 'real' reflexes with /s/, or even doublets!

And in any case, as I say, Italian doesn't degeminate, it just assimilates: /kk/ > /ktS/ > /ttS/.
Post Reply