Zompist Bboard Again

bradrn wrote: ↑Sun Mar 15, 2020 7:36 am

Richard W wrote: ↑Sun Mar 15, 2020 7:08 am

bradrn wrote: ↑Sun Mar 15, 2020 6:40 am
CO stands for `connector' which indicates that the morpheme has no longer a casual meaning but serves to connect two clauses.

Based on the first two quotes, I’m inclined to agree with Richard W in that it may mean ‘to do with cases’ — but in the last excerpt, ‘the morpheme’ seems to be a verb rather than a putative noun case. I think the only thing I can conclude for now is simply that the word is being used here in a very confusing way.

Have you some evidence you're not showing us? For example, a word meaning 'with' and functioning like a preposition can also mean 'and' (e.g. Thai กับ) for connecting noun phrases, and it doesn't seem much of a stretch to let it join clauses. Indeed, English 'with' can perform that semantic function, except that the second clause has to be nominalised, e.g. 'She washed the dishes with me drying them' meaning pretty much 'She washed the dishes and I dried them'.

Feel free to read the original paper yourself: https://minpaku.repo.nii.ac.jp/?action= ... &file_no=1. So no, I don’t have any evidence I haven’t showed you (unless there was something in the paper which I missed, which is certainly possible).

I don't see any reason to disagree with the author. These morphemes are acting as clause-joining conjunctions; it so happens that they are being suffixed to verbs, which is unsurprising given SOV word order. There is a claim that Tibetan clauses don't need to be transformed in order to apply adpositions to them. If that is valid, then I see nothing confusing in the construction. All that is noteworthy is the scope of these words, which seem to be more like adpositions than mere case markers.

bradrn wrote: ↑Sun Mar 15, 2020 4:44 amAs for case concord, I would be interested to know: do you have any information about how common it is? I used to be under the impression that it was pretty much restricted to European languages, but then I found a Dyirbal example with possible case concord, so now I’m not sure how common or uncommon it is.

Does Semitic count as European? Case concord also happens in Akkadian and Standard Arabic, in a similar way to the European pattern. (Curiously, neither Turkic nor Dravidian seem to have case concord...) I remember there is some degree of case concord in Warlpiri too. This example from Kayardild I posted a while ago has a possessive determiner agreeing in case too (although the less we say about agreement in Kayardild, the better).

Semitic seems quite European in many ways. Case concord is also to be seen in Hurrian, Georgian, NE Caucasian and Uralic, so it's a feature of Europe + Ancient Near East.

Richard W wrote: ↑Sun Mar 15, 2020 8:38 am

bradrn wrote: ↑Sun Mar 15, 2020 7:36 am

Richard W wrote: ↑Sun Mar 15, 2020 7:08 am
Have you some evidence you're not showing us? For example, a word meaning 'with' and functioning like a preposition can also mean 'and' (e.g. Thai กับ) for connecting noun phrases, and it doesn't seem much of a stretch to let it join clauses. Indeed, English 'with' can perform that semantic function, except that the second clause has to be nominalised, e.g. 'She washed the dishes with me drying them' meaning pretty much 'She washed the dishes and I dried them'.

Feel free to read the original paper yourself: https://minpaku.repo.nii.ac.jp/?action= ... &file_no=1. So no, I don’t have any evidence I haven’t showed you (unless there was something in the paper which I missed, which is certainly possible).

I don't see any reason to disagree with the author. These morphemes are acting as clause-joining conjunctions; it so happens that they are being suffixed to verbs, which is unsurprising given SOV word order. There is a claim that Tibetan clauses don't need to be transformed in order to apply adpositions to them. If that is valid, then I see nothing confusing in the construction. All that is noteworthy is the scope of these words, which seem to be more like adpositions than mere case markers.

After re-reading the paper, I think I do agree with you on this. But if Tibetan ‘cases’ are so clearly adpositions rather than usual cases, then why are they so commonly called cases? (cf. Hebrew lə-, which at first appears to be an ablative or dative case, but I’ve never heard of it being called anything other than a preposition.)

bradrn wrote: ↑Sun Mar 15, 2020 6:37 pm After re-reading the paper, I think I do agree with you on this. But if Tibetan ‘cases’ are so clearly adpositions rather than usual cases, then why are they so commonly called cases? (cf. Hebrew lə-, which at first appears to be an ablative or dative case, but I’ve never heard of it being called anything other than a preposition.)

I can think of several reasons.

1. Tibetan tradition. There is a traditional attempt to map them to Sanskrit cases.
2. They are clitics that assimilate to the previous consonant.
3. Discomfort with the idea of an adposition marking the subject. Some people aren't happy calling the Hebrew object marker a preposition.

I've seen an apparently cynical remark that when describing Sino-Tibetan (bello van Driem's terminological mutterings) languages, Indian linguists find cases and Chinese linguists find tones. Tibetan belongs to the Indosphere.

Richard W wrote: ↑Sun Mar 15, 2020 8:07 pm

bradrn wrote: ↑Sun Mar 15, 2020 6:37 pm After re-reading the paper, I think I do agree with you on this. But if Tibetan ‘cases’ are so clearly adpositions rather than usual cases, then why are they so commonly called cases? (cf. Hebrew lə-, which at first appears to be an ablative or dative case, but I’ve never heard of it being called anything other than a preposition.)

I can think of several reasons.

1. Tibetan tradition. There is a traditional attempt to map them to Sanskrit cases.
2. They are clitics that assimilate to the previous consonant.
3. Discomfort with the idea of an adposition marking the subject. Some people aren't happy calling the Hebrew object marker a preposition.

Those reasons are understandable, even if some of them make no sense linguistically — to me, (2) seems to be the only valid point.

(As for (1): I’m always unhappy when terminology is retained for purely historical reasons. I think I’ve proposed several times on this board that words like ‘infinitive’, ‘participle’ and ‘adverb’ are so specific to Latin and English linguistics that they should be scrapped as useless.)

I've seen an apparently cynical remark that when describing Sino-Tibetan (bello van Driem's terminological mutterings) languages, Indian linguists find cases and Chinese linguists find tones. Tibetan belongs to the Indosphere.

That sounds about right. I remember reading a grammar of Dongwang Tibetan by a linguist who was obviously familiar with Chinese languages, and which included a long, detailed discussion of the tonal system. (I remember this mainly because I then gave the phonology as a romanization challenge, only to discover that this apparently detailed description was in fact frustratingly incomplete.)

Glass Half Baked wrote: ↑Sun Mar 08, 2020 4:23 pm Qwynegold, I think I figured it out. The database was created by Mark Donohue at ANU ~2013, but in 2015-2016 budget cuts forced him out, and 2016 seems to be the last time anyone has accessed the database. Donohue now has a job in the private sector, with Language Intel, a linguistics consulting company (career goals, amirite?). Soon after taking this job the Cocos Islands domain he was using to host his personal email lapsed, presumably because he just uses his work email now, which is not listed on the company website. But they do have a company email, and a contact number that goes straight to Mark, so you can definitely ask him what happened to his database after he was forced to sell out and wear a suit.
http://languageintel.org/contact-us/

Oh, thanks for the detective work. But I have a little too much anxiety to e-mail some random linguist for the purpose of a conlanging project...

Nortaneous wrote: ↑Mon Mar 09, 2020 9:25 am

Glass Half Baked wrote: ↑Mon Mar 09, 2020 3:00 amThat's the site Qwynegold found earlier. The problem is, without column headings, it's just a random number generator.

phonotactics.csv-metadata.json has the headers

Now it says "closed access". -__-

I have another question. Does evidentiality have to be obligatory (in some circumstances) for it to be grammatical evidentiality? Wikipedia has this to say:

The elements in European languages indicating the information source are optional and usually do not indicate evidentiality as their primary function — thus they do not form a grammatical category.

I can't tell whether it's saying that optionality disqualifies it as being grammatical evidentiality or not. And I have no idea what WALS is doing. They count Finnish and Swedish as having grammatical evidentiality, but I have no idea what part of Finnish or Swedish grammar that is supposed to refer to.

Random language observation: I just noticed the importance of distinguishing [ˈpʰipʰi] (childish word for the penis) and [pʰiˈpʰi] (acronym of the term "prepositional phrase").

Richard W wrote: ↑Sun Mar 15, 2020 8:07 pmDiscomfort with the idea of an adposition marking the subject. Some people aren't happy calling the Hebrew object marker a preposition.

I am reminded of the couple times here on the ZBB, years ago, when I was scolded for glossing a clitic "particle" in a conlang of mine as "GEN", and also the Spanish personal "a" as "ACC" (vi a María saw.1S ACC Maria). I can only imagine the resistence linguists who call a preposition that marks subjects a "preposition" must get...

bradrn wrote: ↑Sun Mar 15, 2020 8:34 pm(As for (1): I’m always unhappy when terminology is retained for purely historical reasons. I think I’ve proposed several times on this board that words like ‘infinitive’, ‘participle’ and ‘adverb’ are so specific to Latin and English linguistics that they should be scrapped as useless.)

"Infinitive" feels pretty useless even in English, unless you use it to mean the basic invariable form ("be, have"), which would cover both the somewhat traditional "bare infinitive" and the traditional "present subjunctive".

I agree "adverb" is rather useless in many languages like e.g. Spanish and German, but I'd like to point out that even in these languages there is usually a small set of words indicating place/manner/time that seemingly behave like Latin/English adverbs by their nature...

Spanish aquí 'here' is sort of like the demonstrative/interrogative/personal pronouns in the way it can modify nouns straightforwardly with de (los libros de aquí 'the books that belong here'), and like the demonstrative pronouns it has an obvious deictic force in its meaning (and use), but when used to express a syntactic adjunct of place, it does so without the preposition en unlike nouns and pronouns (lo metí aquí 'I put it here', lo metí en esto 'I put it inside this'). Así 'this way, thus' is even more of a classic adverb, being unable to modify nouns with a PP beginning with de.

I've seen an apparently cynical remark that when describing Sino-Tibetan (bello van Driem's terminological mutterings) languages, Indian linguists find cases and Chinese linguists find tones. Tibetan belongs to the Indosphere.

I sometimes wonder whether the described simple vowel systems of most indigenous languages of Spanish Latin America (typically /a e i o u/ or /a e i o/, possibly with phonemic length) have been influenced by linguists being way too comfortable with Spanish and not enough with other phonologies (even their own native English if they have it, possibly). Although it may also be that Spanish has altered a lot of these languages as a superstratum anyway.

Qwynegold wrote: ↑Tue Mar 17, 2020 3:37 pmI have another question. Does evidentiality have to be obligatory (in some circumstances) for it to be grammatical evidentiality? Wikipedia has this to say:

The elements in European languages indicating the information source are optional and usually do not indicate evidentiality as their primary function — thus they do not form a grammatical category.

I can't tell whether it's saying that optionality disqualifies it as being grammatical evidentiality or not. And I have no idea what WALS is doing. They count Finnish and Swedish as having grammatical evidentiality, but I have no idea what part of Finnish or Swedish grammar that is supposed to refer to.

Traditionally, what Wikipedia says is correct, but it seems to me there is a small recent movement to identify the parts of the grammars of European languages where evidentiality plays a grammatical role (even if the are no dedicated evidential morphemes).

For example, you can come across papers about evidentiality in Classical Latin which talk about the use of the indicative vs. subjunctive to express first-hand witnessed knowledge vs. hypotheticals/hearsay (this happens in a few situations, e.g. after quod 'because'). Or the use of the indicative vs. subjunctive in Spanish relative clauses to express certainty vs. uncertainty (note that in Spanish the "certainty" may come from good inferences or reliable hearsay). Gabriele Diewald's Evidentiality in German (2010) is a 350-page-long argument for werden + inf. and scheinen + zu + inf. (and a couple others) as markers of inferential evidentiality some of the time (and there is a further cautious, tentative mention of wollen/sollen for mediated evidentiality / reported speech: Sie will nicht am Tatort gewesen sein 'She claims that she wasn't at the crime scene'...).

Presumably, the people who entered the WALS data for Finnish/Swedish have been influenced by this?

Cambridge is making about 58 textbooks freely available online through May:

https://www.cambridge.org/core/what-we- ... FB5AC93698

Looks like anyone can get at them! Though each chapter is a separate page.

zompist wrote: ↑Tue Mar 17, 2020 5:50 pm Cambridge is making about 58 textbooks freely available online through May:

Maybe especially topical, Thomas Payne's Describing Morphosyntax is one of the now-free textbooks.

I notice that they put up Dickins and Watson's Arabic: An Advanced Course, but not Schulz et al.'s Arabic: An Elementary-Intermediate Course...

Qwynegold wrote: ↑Tue Mar 17, 2020 3:00 pm

Nortaneous wrote: ↑Mon Mar 09, 2020 9:25 am

Glass Half Baked wrote: ↑Mon Mar 09, 2020 3:00 amThat's the site Qwynegold found earlier. The problem is, without column headings, it's just a random number generator.

phonotactics.csv-metadata.json has the headers

Now it says "closed access". -__-

the hell? that suggests to me that the project was for whatever reason actively axed; if it'd just lapsed into defunctness people wouldn't be actively restricting it. unless this is some weird Zenodo automation BS, but I don't think that exists

might be a good idea to just compile a newer, better one, although I'm not sure how that'd work. maybe for each language list all allowed clusters and have some kind of tool that can autogenerate WPD-style fields ("this language allows Cy, this language allows Cw...") and a relatively free-form search ("show me languages that allow #C[+nasal]C[+lateral -obstruent]")... on the other hand, I'm not sure what kinds of questions people would even want to use a phonotactics database to answer, and obv any database should be designed with "what questions will this be used to answer?" in mind

Are there any examples of sound changes or allomorphs in (phonetical) environment A or environment B, but not when both A and B are present? By environments I mean things like e.g. intervocally, after front vowels or word-finally

Ser wrote: ↑Tue Mar 17, 2020 4:07 pm

bradrn wrote: ↑Sun Mar 15, 2020 8:34 pm(As for (1): I’m always unhappy when terminology is retained for purely historical reasons. I think I’ve proposed several times on this board that words like ‘infinitive’, ‘participle’ and ‘adverb’ are so specific to Latin and English linguistics that they should be scrapped as useless.)

"Infinitive" feels pretty useless even in English, unless you use it to mean the basic invariable form ("be, have"), which would cover both the somewhat traditional "bare infinitive" and the traditional "present subjunctive".

I agree "adverb" is rather useless in many languages like e.g. Spanish and German, but I'd like to point out that even in these languages there is usually a small set of words indicating place/manner/time that seemingly behave like Latin/English adverbs by their nature...

I would dispute exactly how useful "adverb" really is even in English, given it covers a somewhat disparate and not necessarily very clearly defined set of elements. As for "participle", there's some ambiguity/overlap in English with "gerund" and you could make a case that we'd be better off inventing a new term entirely.

Curlyjimsam wrote: ↑Thu Mar 19, 2020 10:34 am

Ser wrote: ↑Tue Mar 17, 2020 4:07 pm

bradrn wrote: ↑Sun Mar 15, 2020 8:34 pm(As for (1): I’m always unhappy when terminology is retained for purely historical reasons. I think I’ve proposed several times on this board that words like ‘infinitive’, ‘participle’ and ‘adverb’ are so specific to Latin and English linguistics that they should be scrapped as useless.)

"Infinitive" feels pretty useless even in English, unless you use it to mean the basic invariable form ("be, have"), which would cover both the somewhat traditional "bare infinitive" and the traditional "present subjunctive".

I agree "adverb" is rather useless in many languages like e.g. Spanish and German, but I'd like to point out that even in these languages there is usually a small set of words indicating place/manner/time that seemingly behave like Latin/English adverbs by their nature...

I would dispute exactly how useful "adverb" really is even in English, given it covers a somewhat disparate and not necessarily very clearly defined set of elements. As for "participle", there's some ambiguity/overlap in English with "gerund" and you could make a case that we'd be better off inventing a new term entirely.

Yes, this is exactly my point!

Ser wrote: ↑Tue Mar 17, 2020 4:26 pm

Qwynegold wrote: ↑Tue Mar 17, 2020 3:37 pmI have another question. Does evidentiality have to be obligatory (in some circumstances) for it to be grammatical evidentiality? Wikipedia has this to say:

The elements in European languages indicating the information source are optional and usually do not indicate evidentiality as their primary function — thus they do not form a grammatical category.

I can't tell whether it's saying that optionality disqualifies it as being grammatical evidentiality or not. And I have no idea what WALS is doing. They count Finnish and Swedish as having grammatical evidentiality, but I have no idea what part of Finnish or Swedish grammar that is supposed to refer to.

Traditionally, what Wikipedia says is correct, but it seems to me there is a small recent movement to identify the parts of the grammars of European languages where evidentiality plays a grammatical role (even if the are no dedicated evidential morphemes).

Hmm, okay. But what about the question I had? Does grammatical evidentiality necessarily have to be obligatory?

Ser wrote: ↑Tue Mar 17, 2020 4:26 pmPresumably, the people who entered the WALS data for Finnish/Swedish have been influenced by this?

I have no idea. I would like to know what words or grammatical structures they have used as basis for claiming this.

akam chinjir wrote: ↑Tue Mar 17, 2020 10:20 pm

zompist wrote: ↑Tue Mar 17, 2020 5:50 pm Cambridge is making about 58 textbooks freely available online through May:

Maybe especially topical, Thomas Payne's Describing Morphosyntax is one of the now-free textbooks.

Now that I'm looking at this, it says:

Due to performance issues caused by unprecedented demand and reported misuse, we have had to temporarily remove the free access to textbooks. We apologise for the inconvenience caused and are working to address these concerns to reinstate free access as soon as possible.

I hate to write so many posts in a row here, because I'm afraid I will bury my own, more important questions in this thread. But I couldn't resist asking about this thing just for curiosity's sake.

Do you know of any examples, from various languages, of words for bodyparts (human or animal) that have a special look, and which contain a cranberry morpheme? I haven't checked the etymologies, so I'm not 100% sure that these are cranberry morphemes, but here are my examples:

English
bucktooth EDIT: Buck is not a cranberry morpheme.

Finnish
nököhammas - bucktooth (hammas = tooth)
hörökorva - a human ear that stands out (korva = ear)
luppakorva - an animal's ear that doesn't stand up, e.g. the ears of a beagle
hölösuu - a big mouth, maybe of a person who talks a lot (suu = mouth)
töpöhäntä - a cropped tail (häntä = tail)