Zompist Bboard Again

akam chinjir wrote: ↑Tue Jun 02, 2020 10:05 pm

bradrn wrote: ↑Tue Jun 02, 2020 8:34 pm No, that wasn’t what I was talking about. Basically, if you have a language with pronominal clitics (for subjects, say), you can do either ‘I see you’ or ‘1s=see you’. If you have a language with pervasive zero anaphora, you can do either ‘I see you’ or ‘∅ see you’. I was asking whether I could combine those two strategies; that is, whether there are any languages which can do any of ‘I see you’ (with a full NP), ‘1s=see you’ (with a clitic), ‘∅ see you’ (with no clitic and zero anaphora).

Hmm, I think usually when you allow pro-drop, overt pronouns mostly only show up for contrast or focus or something, which you'd think would require an independent form. But it's also not hard to imagine having a three-way distinction here, like maybe a clitic pronoun (only) for a topic switch. Don't know any examples though.

Well, yes, this is why I was dubious that such languages exist: given that both clitic pronouns and dropped pronouns act as reduced forms of reference, there isn’t really much point in having both.

Further on this topic, I first became aware of zero anaphora in this post by chris_notts, citing Huang:

chris_notts wrote: ↑Fri May 01, 2020 4:09 pm

Huang wrote: Let me start with the pragmaticness of anaphora in a pragmatic language. If we compare a prototypical pragmatic language (such as Chinese, Japanese, and Korean) with a prototypical syntactic language (such as English, French and German), we will find that anaphora in the former behaves in a more pragmatic way than that in the latter (see e.g. Huang 1994 for a detailed analysis of Chinese). More specifically, a pragmatic language seems to have the following characteristics as far as anaphora goes.

Anaphora in a pragmatic language
(a) Massive occurrence of zero anaphora
(b) Existence of pragmatic zero anaphors or empty pragmatic categories
(c) Pragmatic obligatory control
(d) Long-distance reflexivization

The crucial point here for the previous discussion is that zeros are widely used, and not in a syntactically restricted way. Multiple zeros can occur in the same clause, with different antecedents, not restricted to be the marked topic of a previous or matrix clause … a zero in a subordinate clause is controlled by the matrix subject (and topic) in the first example, and by the matrix recipient (and non-topic) in the other. The choice is pragmatically driven, not syntactic rule driven, except for the fact that some restrictions come from the choice of matrix verb "promise" …

If languages with pronominal clitics don’t tend to have zero anaphora, does that mean that they tend to be ‘syntactic’ rather than ‘pragmatic’ in Huang’s dichotomy? Or can all of the above still apply in a language with pronominal clitics, if you replace ‘zeros’ with ‘pronominal clitics’?

(This was really my motivation for asking the question in the first place: I’d like my language to be more ‘pragmatic’ with lots of zeros and anaphora, but I also really like the idea of having a mostly-isolating language which nonetheless has lots of clitics, particularly pronominal clitics. I’m trying to figure out how much of each I can have, or if I can only have one and need to sacrifice the other.)

(Okay, one example: apparently my English allows all three options with first person singular, at least.)

Could you give an example? To my knowledge, English doesn’t have any clitic pronouns, and has only exceedingly rare pro-drop.

bradrn wrote: ↑Tue Jun 02, 2020 10:26 pm

(Okay, one example: apparently my English allows all three options with first person singular, at least.)

Could you give an example? To my knowledge, English doesn’t have any clitic pronouns, and has only exceedingly rare pro-drop.

I wrote that after noticing that in my previous sentence, I'd pro-dropped without really thinking about it; not sure what the rule is for me (but I'm pretty sure it's first-person-specific). I was taking it for granted that English pronouns often occur in phonologically/prosodically reduced forms, and don't constitute independent words. Like in "I wrote that..." I think for me the pronoun is usually just an unstressed centralised vowel stuck on the front of whatever follows it; it certainly doesn't get independent stress most of the time.

akam chinjir wrote: ↑Tue Jun 02, 2020 10:35 pm

bradrn wrote: ↑Tue Jun 02, 2020 10:26 pm

(Okay, one example: apparently my English allows all three options with first person singular, at least.)

Could you give an example? To my knowledge, English doesn’t have any clitic pronouns, and has only exceedingly rare pro-drop.

I wrote that after noticing that in my previous sentence, I'd pro-dropped without really thinking about it; not sure what the rule is for me (but I'm pretty sure it's first-person-specific).

‘Don't know any examples though.’ — is that the one you’re talking about? If so, that’s grammatical for me as well. Actually, another place I’ve noticed it is in emails, with constructions such as ‘Will tell you if anything changes’ or ‘Am running off to join the circus’. (Note: in case you were worrying, the second example has never occurred in any of my emails!)

I was taking it for granted that English pronouns often occur in phonologically/prosodically reduced forms, and don't constitute independent words. Like in "I wrote that..." I think for me the pronoun is usually just an unstressed centralised vowel stuck on the front of whatever follows it; it certainly doesn't get independent stress most of the time.

This is true of me as well. In fact, for me it can go even further, with the pronoun being reduced all the way to just a nasal: something like ‘I’m going to go now’ will often end up as [ŋ̍ˌɡɜnəˈɡɞ͡ʉ næ͡ʉ], and ‘I’m doing it’ becomes [n̩ˈdʉˑ.ɘŋˌɘt] (with two clitics!). But somehow I never made the connection that this means the pronoun has become a clitic.

bradrn wrote: ↑Tue Jun 02, 2020 10:26 pm To my knowledge, English ... has only exceedingly rare pro-drop.

doubt. not frequent, but dunno why you'd go from there to exceedingly rare

Nortaneous wrote: ↑Wed Jun 03, 2020 12:20 am

bradrn wrote: ↑Tue Jun 02, 2020 10:26 pm To my knowledge, English ... has only exceedingly rare pro-drop.

doubt. not frequent, but dunno why you'd go from there to exceedingly rare

When I wrote that, I hadn’t realised how common it was. But I changed my mind after akam chinjir pointed out that it’s more common than I thought — I do agree with you now.

I like the fact the vidya really is cognate with Vidya.

So Icelandic, Faroese, Greek, Latvian, and Lithuanian are the only modern IE languages to preserve the IE nominative singular -s, right? And the latter three are the only ones to preserve it as /s/?

Curious that these are basically on the fringes of Europe (or at least IE-speaking Europe, with the Baltics). Maybe a retention due to relative isolation from a core of more innovative IE varieties?

Nominative -s survived in Old French, so up to the 13th century or so.

It's preserved in some proper nouns (e.g. Jacques), and still pronounced in fils, though this is a spelling pronunciation.

missals wrote: ↑Sat Jun 06, 2020 5:05 pm So Icelandic, Faroese, Greek, Latvian, and Lithuanian are the only modern IE languages to preserve the IE nominative singular -s, right? And the latter three are the only ones to preserve it as /s/?

Curious that these are basically on the fringes of Europe (or at least IE-speaking Europe, with the Baltics). Maybe a retention due to relative isolation from a core of more innovative IE varieties?

In many I.E. languages the loss looks like a fairly natural outcome of phonological sound changes that got rid of word final /s/. The idea of isolated languages being more conservative is I think mainly an idea that they are less likely to analogically simplify complicated inherited parts of grammar, not that they are less likely to undergo regular sound changes (like loss of final consonants) that may obscure certain inherited parts of grammar. E.g. Greek had sound changes that got rid of PIE word-initial /s/ before vowels, but I think it's a coincidence that this happened not to be a morphologically significant sound change. (And the Greek use of nominative -s isn't always conservative: there are a number of cases where it has been analogically extended/restored to nouns that didn't end in /s/ in the nominative in Ancient Greek or PIE, such as Modern Greek άντρας/άνδρας < Ancient Greel ἀνήρ < PIE *h₂nḗr.)

(Edit: I'm also not sure whether Greek can really be characterized as isolated or positioned at the fringe of an IE-speaking language area. Isn't it fairly geographically close to Italic/Romance, Slavic, Armenian, Albanian, and Indo-Iranian languages?)

It looks like youre right. Its possible that some -s survives in a language in India, but looking through https://en.wikipedia.org/wiki/Sanskrit_nouns it seems that most of the inherited masculine -s was already gone by the time of classical Sanskrit, and what survives may not have been thought of as a suffix by that point. Notably the -s seems to be just as common on feminine and neuter nouns as it is on masculine nouns.

missals wrote: ↑Sat Jun 06, 2020 5:05 pmCurious that these are basically on the fringes of Europe (or at least IE-speaking Europe, with the Baltics). Maybe a retention due to relative isolation from a core of more innovative IE varieties?

Greek is not exactly on the fringe, but I'd say yes about the others.

zompist wrote: ↑Sat Jun 06, 2020 5:29 pmIt's preserved in some proper nouns (e.g. Jacques), and still pronounced in fils, though this is a spelling pronunciation.

The Trésor de la langue française at least thinks the -s was retained due to its frequent use as a vocative and to distinguish it from fīlum > fil 'thread'. Also, whoever wrote the English Wiktionary entry thinks the [s] survived via the Middle French pausal form.

I think there is enough dialectal evidence now that Middle French didn't uniformly drop all final consonants (many traditional dialects have instances of retained -r for example), even if it's true that a good lot of them in modern French were restored from spellings as knowledge of both standard French and literacy spread, notably -c and -f.

Spanish has some fossilized nominative -s too:
- In proper names: Dios, Jesús, Marcos, Carlos, Pilatos (besides the more common Pilato), Longinos (the centurion who said Jesus was divine by the end of John, also Longino), Roncesvalles (rumicis vallis, i.e. 'sorrel valley', rumicis is genitive but vallis nominative). Previously also Pablos, Domingos.
- Other: avestruz < avis strūthius (both nominative), piedra pómez (< petra pūmex, if accusative it'd be piedra *ponce < petra pūmicem).

I remember yous have discussed positive 'anymore' a few times last year. I just read The Scots Language Abroad by Michael Montgomery and it says that 'anymore' to mean 'nowadays' in Ulster Scots. Is this like the 'anymore' that you are familiar with?

Some Arabic fonts on Linux:



Notice how DejaVu Sans grabs the alif-waSla on the leftmost word from another font (FreeSans), because even though it's an essential letter to render Arabic diacritics, for some weird reason it's missing in that font. As it's usual with DejaVu fonts, it's so wide I had to reduce it to 20pt to make it look like the others.

Notice how FreeSans, in the rightmost word, puts the kasra (the little slanting line) incorrectly below the laam, unlike the other fonts which correctly put it under the shadda mark (the little <w>-like thing on top of the laam). Terrible. As it's usual with FreeSans, it's so thin and small that I had to increase its size to 25pt to make it look like the others. Sigh.

I'm seriously glad Google has published these Noto fonts because decent open-source, legally-unemcumbered good Arabic fonts have been otherwise basically non-existent...

Even on Windows, I have complaints about the bad choices in the design of the Arial/Times New Roman Arabic font (the two share exactly the same font glyphs). The letter shapes are particularly too small most of the time to render them alongside regular Latin script, although this can be addressed by simply increasing the size of any instance of Arabic text every time. But the thing is, the Tahoma font doesn't suffer from this, and is pretty decent for non-diacritic text, like DejaVu Sans and Noto Naskh Arabic / Noto Sans Arabic.

MacAnDàil wrote: ↑Tue Jun 09, 2020 1:31 pmI remember yous have discussed positive 'anymore' a few times last year. I just read The Scots Language Abroad by Michael Montgomery and it says that 'anymore' to mean 'nowadays' in Ulster Scots. Is this like the 'anymore' that you are familiar with?

Yes. Some people use it in North America, but it's pretty non-standard and possibly low-class.

Ser wrote: ↑Tue Jun 09, 2020 3:56 pm Notice how FreeSans, in the rightmost word, puts the kasra (the little slanting line) incorrectly below the laam, unlike the other fonts which correctly put it under the shadda mark (the little <w>-like thing on top of the laam). Terrible. As it's usual with FreeSans, it's so thin and small that I had to increase its size to 25pt to make it look like the others. Sigh.

My recollection, and this goes back forty years, is there are two different styles - one that writes kasra under shadda and another that always writes it below the consonant.

Richard W wrote: ↑Tue Jun 09, 2020 4:48 pmMy recollection, and this goes back forty years, is there are two different styles - one that writes kasra under shadda and another that always writes it below the consonant.

Oh, that's interesting! Could be the case. Nevertheless, that alternative style is a lot more uncommon nowadays, and FreeSans may be one of the very few fonts that use it.

EDIT: This user here in this thread from 2010, from the Arabeyes mailing list, has the opposite complaint than me, asking why the Arabeyes font puts the kasra under shadda!

Ser wrote: ↑Tue Jun 09, 2020 3:56 pm

MacAnDàil wrote: ↑Tue Jun 09, 2020 1:31 pmI remember yous have discussed positive 'anymore' a few times last year. I just read The Scots Language Abroad by Michael Montgomery and it says that 'anymore' to mean 'nowadays' in Ulster Scots. Is this like the 'anymore' that you are familiar with?

Yes. Some people use it in North America, but it's pretty non-standard and possibly low-class.

Non-standard, yes, but I don't think it's class-marked. So far as I can see, it's one of those stealth dialectal differences, where people who use it don't realize it's non-standard. (Another is expressions like "This code needs fixed", which is used in Ohio/Pennsylvania; users are often quite surprised when others 'correct' them.)

zompist wrote: ↑Tue Jun 09, 2020 5:02 pm

Ser wrote: ↑Tue Jun 09, 2020 3:56 pm

MacAnDàil wrote: ↑Tue Jun 09, 2020 1:31 pmI remember yous have discussed positive 'anymore' a few times last year. I just read The Scots Language Abroad by Michael Montgomery and it says that 'anymore' to mean 'nowadays' in Ulster Scots. Is this like the 'anymore' that you are familiar with?

Yes. Some people use it in North America, but it's pretty non-standard and possibly low-class.

Non-standard, yes, but I don't think it's class-marked. So far as I can see, it's one of those stealth dialectal differences, where people who use it don't realize it's non-standard. (Another is expressions like "This code needs fixed", which is used in Ohio/Pennsylvania; users are often quite surprised when others 'correct' them.)

I always thought positive "anymore" was a Midwestern thing. I grew up in Michigan and picked it up as a teenager (I think from my parents, who might have picked it up from others). I'm not aware of any class associations either.

Why aren't there languages with vowel systems like /y ø a ʌ ɯ/ or /y ø œ ɒ ʌ ɯ/? Is there any speculation as to why?

Ser wrote: ↑Tue Jun 09, 2020 11:13 pm Why aren't there languages with vowel systems like /y ø a ʌ ɯ/ or /y ø œ ɒ ʌ ɯ/? Is there any speculation as to why?

Not an expert on this at all, but I believe this is because front rounded and back unrounded vowels are less perceptually distinct than front unrounded and back rounded vowels. e.g. if you look at a formant chart like this:



Note that the front rounded and back unrounded vowels have a much closer F2 to each other than the corresponding front unrounded and back rounded vowels.

I think that's right. Most small to medium vowel systems can be seen as essentially two dimensional, described by F1 and F2 (unless they're vertical vowel systems, in which case they're actually one dimensional). Rounding and backing both reduce F2, so in the two dimensional view they're basically doing the same thing, and the vowels are usually positioned to make them more distinctive. [ø] and [ɤ] are closer in F2, and so sound less distinct from each other than [e] and [o].