(L&N) Q&A Thread - Quick questions go here [2010-2019]

A forum for discussing linguistics or just languages in general.
Locked
User avatar
Lambuzhao
korean
korean
Posts: 5405
Joined: 13 May 2012 02:57

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

Post by Lambuzhao »

Micamo,

"lion of strength" sounds very :heb: with the noun + noun.

Also, regarding verbs, I was looking for how to say "I like it" in Ancient :grc: and stumbled on an OT :heb: phrase which the Greek/Latin/English translates as "like": <good to (my/thy) eye>. That's an example of just the sort of oblique construction you mention, right?

Your mention of oblique constructions makes me think of Ancient Egyptian, where a number of verbless and verbal constructions using "hand", "eye" and "heart" were carried right over into Coptic.
User avatar
Micamo
MVP
MVP
Posts: 5671
Joined: 05 Sep 2010 19:48
Contact:

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

Post by Micamo »

Are there any natlangs with switch-reference markers for roles other than subjects, like objects?
My pronouns are <xe> [ziː] / <xym> [zɪm] / <xys> [zɪz]

My shitty twitter
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

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

Post by eldin raigmore »

Micamo wrote:Are there any natlangs with switch-reference markers for roles other than subjects, like objects?
Oh yes.
Read Switch-reference and universal grammar: proceedings of a Symposium on Switch Reference and Universal Grammar, Winnipeg, May 1981 John Haiman, Pamela Munro J. Benjamins Pub. Co., 1983 - 342 pages

Usually the reference-tracking tracks the marked clause's subject, but it may track whether it was the direct or indirect object of the referenced clause, as well as whether it was the subject.
Also, it may track proper containment, in at least one direction, as well as strict equality.

In case the language is morphosyntactically ergative, the switch-reference system often tracks the agents rather than the "subjects".

Adpihi's switch-reference system, which I've discussed before, contains no features not seen in natlangs. TTBOMK the combination occurs in no natlang, but each individual feature does.

In Adpihi there are two switch-reference-marking systems, operating simultaneously if the marked clause is transitive.
One system tracks the marked clause's Subject (if it's intransitive) or Actor (if it's transitive); the other tracks the marked clause's Undergoer (if it's transitive).
TTBOMK no natlang has two systems operating concurrently. But some ergative languages' switch-reference systems do track the marked clause's Subject or Undergoer instead of its Subject or Actor.

In Adpihi, the Subject-or-Actor marker tracks whether and how the marked clause's Subject-or-Actor is related to the referenced clause's Subject-or-Actor, or how it's related to the referenced clause's Undergoer-or-Exended-core-term. (Ditransitive clauses, and bivalent intransitive clauses, have Extended-core-terms.)
(You can read that as; one set of markings applies if the marked clause's Subject is related to the referenced clause's Subject; another applies if the marked clause's Subject or Agent is related to the reference clause's Direct Object or Indirect Object.)
That really happens in some natlang; in fact some of them are more detailed, marking relationship to the Direct Object differently than relationship to the Indirect Object.

In Adpihi, the Undergoer reference-tracker marks whether the marked clause's Undergoer is related, on the one hand, to the referenced clause's Subject or Undergoer, or, on the other hand, to the referenced clause's Actor or Extended core term.

In Adpihi, there's one set of markings for strict equality and a different set for proper containment.
In natlangs, if proper containment is marked, usually proper containment in one direction is marked the same as equality (edit: or is marked or unmarked the same as "no relation" /edit), while proper containment in the other direction is marked differently.
Edit: In Adpihi the direction of proper containment can usually be disambiguated by number or by gender. Adpihi has singular, dual, trial, paucal, and plural numbers. If two referents, one of which properly contains the other or vice-versa, have exactly the same grammatical number, then either they're both paucal or they're both plural; otherwise the one with the higher grammatical number properly contains the other one. Also, Adpihi has an epicene/common/mixed gender. If one of the two referents has common gender and the other does not, the common-gender one must properly contain the other one. If both referents have the same number and the same gender, there may be other ways to disambiguate, or it may be unnecessary to disambiguate.
Here's Adpihi's marks for switch-reference for the Subject-or-Actor marker-slot:
1. least marked; marked clause's Subject or Actor is same as referenced clause's Subject or Actor.
2. second-least marked; marked clause's Subject or Actor is same as referenced clause's Undergoer or Extended-Core-Term.
3. moderately marked; marked clause's Subject or Actor properly contains, or is properly contained in, referenced clause's Subject or Actor.
4. Most marked; marked clause's Subject or Actor properly contains, or is properly contained in, referenced clause's Undergoer or Extended-Core-Term.
0. Unmarked; marked clause's Subject or Actor is disjoint from any core participant of referenced clause. (Or perhaps they overlap without either one containing the other.)

Here's Adpihi's marks for switch-reference for the Undergoer marker-slot:
1. least marked; marked clause's Undergoer is same as referenced clause's Subject or Undergoer.
2. second-least marked; marked clause's Undergoer is same as referenced clause's Actor or Extended-Core-Term.
3. moderately marked; marked clause's Undergoer properly contains, or is properly contained in, referenced clause's Subject or Undergoer.
4. Most marked; marked clause's Subject or Actor properly contains, or is properly contained in, referenced clause's Actor or Extended-Core-Term.
0. Unmarked; marked clause's Undergoer is disjoint from any core participant of referenced clause. (Or perhaps they overlap without either one containing the other.)

Follow up on this search: http://www.dogpile.com/info.dogpl/searc ... -reference.
The first page of hits contains a link to a list of documents about switch-reference.
(Of course it also contains several links to individual documents.)

Because of the switch-reference system, Adpihi has "the principle of disjoint reference"; in any clause with more than one core term, the groups referred to by any two core terms cannot contain any members in common. (Strictly speaking I only need that for clauses that are in clause-chains, and could afford to violate it for stand-alone clauses that aren't in chains.)
Because of the principle of disjoint reference, Adpihi has a robust and complete and detailed reflexivization and reciprocalization system.
Edit: Also; because it's nearly impossible to work out the syntactic and morphosyntactic mechanics of combining morphosyntactic subordination with clause-chaining, Adpihi foregoes any embedding of clauses. Imagine a clause chain in which one of the clauses had a subordinate clause-chain and one of the subordinate clauses shared reference with a chainmate of its matrix. How would it be marked? It's a tangled mess, and as difficult as the excess complexity that might happen to the switch-reference system if the marked clause's referent-groups overlapped and so did the referenced-clause's.

(a female who loves an empty set?)
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Why in English do we call them "buildings" instead of "built

Post by eldin raigmore »

Seriously, why is the English word for an edificium "building" instead of "built"?
It should be a past perfective passive participle, not a present imperfective passiveactive participle; shouldn't it?

We don't call the pellets in a shotgun shell "shooting"; we call them "shot".
Last edited by eldin raigmore on 11 Apr 2013 06:49, edited 1 time in total.
ol bofosh
greek
greek
Posts: 668
Joined: 27 Aug 2012 14:59
Location: tʰæ.ɹʷˠə.ˈgɜʉ̯.nɜ kʰæ.tə.ˈlɜʉ̯.nʲɜ spɛ̝ɪ̯n ˈjʏː.ɹəʔp

Re: Why in English do we call them "buildings" instead of "b

Post by ol bofosh »

eldin raigmore wrote:Seriously, why is the English word for an edificium "building" instead of "built"?
It should be a past perfective passive participle, not a present imperfective passive participle; shouldn't it?

We don't call the pellets in a shotgun shell "shooting"; we call them "shot".
That sounds nice and logical. But maybe logic's the problem. [xD]
User avatar
Omzinesý
mongolian
mongolian
Posts: 4082
Joined: 27 Aug 2010 08:17
Location: nowhere [naʊhɪɚ]

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

Post by Omzinesý »

Building is an action nominal in Finnish, too.

rakennus 'building'

talo-n rakenn-us 'building of a house'
house-GEN build-DER


An action nominal in the meaning 'a product of X'.
My meta-thread: viewtopic.php?f=6&t=5760
User avatar
Xonen
moderator
moderator
Posts: 1080
Joined: 16 May 2010 00:25

Re: Why in English do we call them "buildings" instead of "b

Post by Xonen »

eldin raigmore wrote:Seriously, why is the English word for an edificium "building" instead of "built"?
It should be a past perfective passive participle, not a present imperfective passive participle; shouldn't it?
Why "should" it be either of those things? AFAIK, it's usually analyzed as a noun. Yes, it's derived from the verb by a suffix that in (many dialects of) Modern English happens to be identical with the one used for forming the present imperfective passive active participle, but that doesn't mean you have to analyze it as the same thing.

In fact, Wiktionary gives three distinct etymologies and even more uses for -ing. The kind of noun derivation that the noun building represents is probably no longer productive (at least not to any great extent), but obviously some nouns derived that way have remained in the language. Nothing terribly strange about that. [:)]
We don't call the pellets in a shotgun shell "shooting"; we call them "shot".
How's that any more logical, though? Now, I'm no expert on shotguns, but shouldn't the pellets actually be outside the shell once they've actually been, you know, shot?
Prinsessa
runic
runic
Posts: 2647
Joined: 07 Nov 2011 14:42

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

Post by Prinsessa »

The -ing in the noun is not the same as the one in the participle. They have different etymologies and I believe there are even dialects that still distinguish them. For comparison, the corresponding endings in Swedish are -[n]ing for the nouns and -ande for the participles.
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

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

Post by eldin raigmore »

All of you have good points.

Skógvur's point about the "-ing" ending on the noun not having the same etymology as the "-ing" ending on active or imperfective or present participles is the most informative one. Can Skógvur (or anyone, really) give me some links to authoritative references where I can get more detail?

Xonen's point "Wiktionary gives three distinct etymologies and even more uses for -ing. The kind of noun derivation that the noun building represents is probably no longer productive" naturally agrees with Skógvur's. It has less detail but more links. I haven't looked up the Wiktionary yet but I will.

Yeah, http://en.wiktionary.org/wiki/-ing#Etymology_1 gives three meanings for the suffix that has that etymology (from Old English "-ing" and "-ung" ). Meaning 1 is a gerund, meaning 3 is an action nominalization. The meaning that forms an uncountable collective noun -- meaning 2 -- is the one we want here, though of course "building-s" isn't uncountable the way "roofing" or "piping" are. Maybe "building-s" is an exception, or maybe Wiktionary needs to improve this entry. Actually, I don't know, but I think it possible "roofing-s" could also be pluralizable and hence a count-noun rather than a mass-or-measure (or at least uncountable) noun. "Piping" (the way they meant it in that article), however, is clearly a mass noun. But anyway, "writ-ing-s" and "read-ing-s" and "hear-ing-s" and "rul-ing-s" are all clearly count nouns, or at least pluralizable nouns, derived from the verbs "write", "read", "hear", and "rule" the same way "building" is derived from "build"; and I could have mistaken them for active participles, just as I did "build-ing"

Wiktionary says etymology 2 comes "from Middle English -inge, -ynge, alteration of earlier -inde, -ende, -and (see -and), from Old English -ende (present participle ending)". Have we already discussed here on this BBoard about how it's more often an imperfective participle than a present participle, and more often an active participle than either an imperfective or a present participle? I have that impression, but maybe I misremember. Anyway, English participles can't have voice unless the verb they're derived from is transitive, like "build" is.

So it looks like the "-ing" in "building" is not an active/imperfective/present-participle forming suffix. Rather it's a suffix that makes verbs into nouns. It's a coincidence that they are spelled the same now, and often pronounced the same. There are at least three different ways of making a noun from a verb by applying this "-ing"; or to put it another way all three of those noun-deriving suffixes seem to have already been spelled alike, and apparently sounded alike, in the earliest forms of Old English that don't have to be reconstructed and that we don't have to put an asterisk in front of, like they do when they tell us it comes from "Proto-Germanic *-andz (present participle ending)". I'd say Wiktionary got meaning 1 (gerund) and meaning 3 (action nominalization) right, though as Wiktionary acknowledges it isn't always clear that those will often be semantically distinct. Meaning 2 is clearly semantically distinct from the other two meanings (also it could be used on another part of speech than a verb; even another noun as in "housing" or "piping" or "roofing"). I don't think Wiktionary's description of it is perfect, but whatever; it's the one we want here. And I think it probably is still productive in English. And when you apply it to a noun to get a new noun I think it qualifies as inflection rather than derivation. Of course I could be wrong.

Etymology 3 is archaic (and/or poetic). I won't go into detail about it since my confusion was between Etymology 2 (the active imperfective present participle) and Etymology 1 (the derived uncountable(?) noun for a bunch of stuff considered as a collection).
Xonen also had a point when he wrote:How's that any more logical, though? Now, I'm no expert on shotguns, but shouldn't the pellets actually be outside the shell once they've actually been, you know, shot?
I'd already thought of that but hadn't decided what to do about it before reading his response.

ol bofosh's remark "maybe logic is the problem" is also to the point. If the pronunciations for two morphemes coalesce, it can be hard to tell that they are different endings; in fact it could be argued that they are no longer just homophones, but rather are now one morpheme with more uses. (Clearly I am not one of the people who would argue that way, though.) Also, if an inflectional affix becomes no-longer-productive and its meaning becomes no-longer-transparent (which means the affix becomes derivation rather than inflection), that's a second reason not to expect logic from language. In this case I think only the coalescence of spelling and the near-coalescence (or, depending on your 'lect, the complete coalescence) of the pronunciations, is the real invitation to make the error I made. Perhaps Etymology 1 has also lost some transparency, at least for Meaning 2 of Etymology 1; and that could also contribute to how easy it is to make the mistake I made. IMO, though, it's still productive; and probably quasi-transparent enough to still count as "inflection", at least when it doesn't change the part-of-speech.
Edit: My dog needs attention so it will be later (if ever) that I get the opportunity to respond to Omzinesy's response.
Last edited by eldin raigmore on 11 Apr 2013 06:42, edited 1 time in total.
User avatar
Micamo
MVP
MVP
Posts: 5671
Joined: 05 Sep 2010 19:48
Contact:

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

Post by Micamo »

eldin raigmore wrote:ol bofosh's remark "maybe logic is the problem" is also to the point. If the pronunciations for two morphemes coalesce, it can be hard to tell that they are different endings; in fact it could be argued that they are no longer just homophones, but rather are now one morpheme with more uses. (Clearly I am not one of the people who would argue that way, though.) Also, if an inflectional affix becomes no-longer-productive and its meaning becomes no-longer-transparent (which means the affix becomes derivation rather than inflection), that's a second reason not to expect logic from language. In this case I think only the coalescence of spelling and the near-coalescence (or, depending on your 'lect, the complete coalescence) of the pronunciations, is the real invitation to make the error I made. Perhaps Etymology 1 has also lost some transparency, at least for Meaning 2 of Etymology 1; and that could also contribute to how easy it is to make the mistake I made. IMO, though, it's still productive; and probably quasi-transparent enough to still count as "inflection", at least when it doesn't change the part-of-speech.
Analyzing them as the same morpheme, in my view, requires that the two affixes display similar syntactic behavior as well as similar morpho-phonetics. You'd have to be insane to claim that the saxon genitive -'s and the plural -s were the same morpheme (though they're written differently) just because they sound the same. The gerund -ing and the participle -ing are not only phonologically divergent in some dialects (moreso than -'s and -s) display very different morphosyntactic behaviors and this is more than enough to justify analyzing them as separate morphemes that happen to have become homophonous (in only some dialects).
My pronouns are <xe> [ziː] / <xym> [zɪm] / <xys> [zɪz]

My shitty twitter
User avatar
Xing
MVP
MVP
Posts: 4153
Joined: 22 Aug 2010 18:46

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

Post by Xing »

Does anyone know of any studies on vowel length in different English dialects?

I have noticed that those varieties of English that are commonly described as having a distinction between intrinsically long and intrinsically short vowels uphold a contrast between the three low back vowels - /ɑː/, /ɒ/ and /ɔː/ - while those varieties that are commonly described as lacking such a distinction merge two of them, or all three.

A hypothesis I've been pondering is that without clear length distinctions, /ɑ/ /ɒ/ and /ɔ/ become very difficult to tell apart from one another, and it's highly likely that at least two of them will merge.
Tmt97
cuneiform
cuneiform
Posts: 78
Joined: 07 Mar 2013 20:12

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

Post by Tmt97 »

Xing wrote:Does anyone know of any studies on vowel length in different English dialects?

I have noticed that those varieties of English that are commonly described as having a distinction between intrinsically long and intrinsically short vowels uphold a contrast between the three low back vowels - /ɑː/, /ɒ/ and /ɔː/ - while those varieties that are commonly described as lacking such a distinction merge two of them, or all three.

A hypothesis I've been pondering is that without clear length distinctions, /ɑ/ /ɒ/ and /ɔ/ become very difficult to tell apart from one another, and it's highly likely that at least two of them will merge.
This might help.'
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2829779/
User avatar
Stammalor
cuneiform
cuneiform
Posts: 143
Joined: 23 Apr 2012 16:14
Location: Scania

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

Post by Stammalor »

Anyone know how natlangs show cases on titles of things that aren't nouns? For example, take the movie Limitless (which I by the way I really like) and lets say that someone with a L1 with accusative case wants to say: I saw Limitless yesterday, it was really good, how would that person deal with the accusative? The movie is definetly a thing, but to the form it is a adjective. And unless the title of something is a NP (Like: I read Moby Dick-acc) then this should happend quite often.
User avatar
prettydragoon
sinic
sinic
Posts: 442
Joined: 29 Jan 2012 10:22
Location: Third star on the left, straight on till tiffin
Contact:

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

Post by prettydragoon »

:fin: Finnish simply declines everything. If it's an object, it declines in object cases, no matter which part of speech it may be.
Image
♀♥♀
What is this, how you say, Rireinutire?
User avatar
Xing
MVP
MVP
Posts: 4153
Joined: 22 Aug 2010 18:46

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

Post by Xing »

Icelandic declines proper names. Also foreign names may be declined (Sarah Palin > Söruh Palin), at least if the fit within common patterns of declination.

If a language has a regular way of forming cases - for example, a suffix that may be added to just about everything - this will probably apply on proper names of various kinds too.

If a language has more complex ways of forming cases - for example, if cases are formed through stem changes, or if there are several ways of forming plural - I suspect it's more likely that proper names - at least if they are foreign or in some way "unusual" - would take a zero case-marking.
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

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

Post by Creyeditor »

Xing wrote:Icelandic declines proper names. Also foreign names may be declined (Sarah Palin > Söruh Palin), at least if the fit within common patterns of declination.

If a language has a regular way of forming cases - for example, a suffix that may be added to just about everything - this will probably apply on proper names of various kinds too.

If a language has more complex ways of forming cases - for example, if cases are formed through stem changes, or if there are several ways of forming plural - I suspect it's more likely that proper names - at least if they are foreign or in some way "unusual" - would take a zero case-marking.
German usually uses zero-marking for proper names in accusative postion, except for ships. I really don't know why, but they always get a definite article which is declined.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
User avatar
Xing
MVP
MVP
Posts: 4153
Joined: 22 Aug 2010 18:46

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

Post by Xing »

Thanks.

It noticeable how long /æ/ is, according to the study. Historically, it would pattern with the short/lax/checked vowels - though some varieties of English have developed (or are developing) a phonological contrast between /æ/ and /æː/.

They didn't seem to have included the low back vowels, for some reason.
User avatar
Iron
sinic
sinic
Posts: 206
Joined: 09 Oct 2012 03:31
Location: Canaan
Contact:

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

Post by Iron »

What exactly specifies a syllable coda? Like say, in a word-final cluster, what consonants are part of it?

I'll give you some examples:

The /r/ in my current project is either [r] or [ɐ̭], the latter appearing the syllable coda. Would a word like börun be pronounced as ['pœ:ɐ̭yn] or ['pœ:ryn]? Or in the case of clusters, would núrt be [nʊɐ̭t] or [nʊrt]?
User avatar
Xing
MVP
MVP
Posts: 4153
Joined: 22 Aug 2010 18:46

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

Post by Xing »

Iron wrote:What exactly specifies a syllable coda? Like say, in a word-final cluster, what consonants are part of it?

I'll give you some examples:

The /r/ in my current project is either [r] or [ɐ̭], the latter appearing the syllable coda. Would a word like börun be pronounced as ['pœ:ɐ̭yn] or ['pœ:ryn]? Or in the case of clusters, would núrt be [nʊɐ̭t] or [nʊrt]?
In the latter example, /r/ is part of the coda - the part of a syllable that comes after its most sonorous segment - its nucleus. (Another possible analysis is to say that /r/ [ɐ̭] form a part of the nucleus, forming a diphthong [ʊɐ̭])

In the former example, it's a bit more tricky. Different languages syllabify words differently. In some languages, intervocalic consonants are best analysed as belonging to the coda of the previous syllable; in others to the onset of the following syllable.

Basically, if it's pronounced ['pœ:ɐ̭yn], /r/ would belong to the first syllable. If it's pronounced ['pœ:ryn], /r/ would belong to the onset of the second syllable.

How to syllabify polysyllabic words is one of the parts of phonology that's up to you. [:)]
User avatar
Iron
sinic
sinic
Posts: 206
Joined: 09 Oct 2012 03:31
Location: Canaan
Contact:

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

Post by Iron »

Xing wrote:In the latter example, /r/ is part of the coda - the part of a syllable that comes after its most sonorous segment - its nucleus. (Another possible analysis is to say that /r/ [ɐ̭] form a part of the nucleus, forming a diphthong [ʊɐ̭])

In the former example, it's a bit more tricky. Different languages syllabify words differently. In some languages, intervocalic consonants are best analysed as belonging to the coda of the previous syllable; in others to the onset of the following syllable.

Basically, if it's pronounced ['pœ:ɐ̭yn], /r/ would belong to the first syllable. If it's pronounced ['pœ:ryn], /r/ would belong to the onset of the second syllable.

How to syllabify polysyllabic words is one of the parts of phonology that's up to you. [:)]
Thanks. I think I'll stick with CVC being the "standard" syllable, seeing as it'll give me an excuse to use [ɐ̭].
Locked