Linguistic Miscellany Thread

Natural languages and linguistics
bradrn
Posts: 6257
Joined: Fri Oct 19, 2018 1:25 am

Re: Linguistic Miscellany Thread

Post by bradrn »

Zju wrote: Wed May 22, 2024 3:20 pm Got some inspiration: reflexive markers. They only refer to the subject. So if the word order in a reflexive clause is fixed to S REFL V (and reflexion is doubly marked), that could over time be interpreted as S=NOM V, after other case suffixes have appeared. Finally, the nominative enclitic would be grammaticalised to a suffix and spread to all subjects.
I don’t find this enormously plausible. You’d need a step in the middle where the reflexive becomes generalised to non-reflexive sentences, and I’m not aware of a single case of that happening — it can easily generalise to other kinds of valence reduction, but there’s no motivation for it to go beyond that.
zompist wrote: Wed May 22, 2024 5:22 pm Another option might be to have a topicalization marker, which then gets reinterpreted as a subject marker.
To me, this seems like the most plausible origin for a nominative marker (an overtly concatenative one, that is). Topics and subjects are famously related.
Travis B. wrote: Wed May 22, 2024 5:36 pm (Note how most of IE has since then lost the original IE nominative markers, aside from a few outliers like Greek, insular NGmc, and Baltic.)
It’s worth noting here that North Germanic, at least, has rather odd case-marking in places. Look at Icelandic’s ‘quirky subjects’, for instance. Even English is arguably marked-nominative.
Conlangs: Scratchpad | Texts | antilanguage
Software: See http://bradrn.com/projects.html
Other: Ergativity for Novices

(Why does phpBB not let me add >5 links here?)
zompist
Site Admin
Posts: 2944
Joined: Sun Jul 08, 2018 5:46 am
Location: Right here, probably
Contact:

Re: Linguistic Miscellany Thread

Post by zompist »

bradrn wrote: Wed May 22, 2024 5:53 pmEven English is arguably marked-nominative.
How so?
User avatar
Glass Half Baked
Posts: 104
Joined: Fri Mar 06, 2020 6:16 am

Re: Linguistic Miscellany Thread

Post by Glass Half Baked »

They're probably referring to the fact that English pronouns are (approximately) "nominative," "possessive," and "everything else," with the everything else being all kinds of objects, plus isolated uses like clefting and exclamations. English pidgins always, AFAIK, analyze the objective form as the "main" form, so it's possible that the nominative pronouns are the more marked series.
vlad
Posts: 94
Joined: Fri Jan 17, 2020 11:24 pm

Re: Linguistic Miscellany Thread

Post by vlad »

Glass Half Baked wrote: Wed May 22, 2024 10:49 pmEnglish pidgins always, AFAIK, analyze the objective form as the "main" form
Chinese Pidgin English used my as the general first person pronoun. (Similarly, Pidgin Portuguese used minha as the nominative, for some reason.)
bradrn
Posts: 6257
Joined: Fri Oct 19, 2018 1:25 am

Re: Linguistic Miscellany Thread

Post by bradrn »

zompist wrote: Wed May 22, 2024 8:00 pm
bradrn wrote: Wed May 22, 2024 5:53 pmEven English is arguably marked-nominative.
How so?
Basically, what Glass Half Baked said:
  • Citation form for the pronouns is the object series: me, us, them, etc.
  • The object series is used after prepositions
  • The object series is used for emphatic subjects and topics
  • The object series is used after be (and by extension in clefts), unlike most nominative-accusative languages (e.g. Latin) where the copula takes two nominative nouns — but like some other marked-nominative languages
  • The two sorts of possessive pronouns are obviously more closely related to the object series than the subject series (me, my, mine vs I; us, our, ours vs we)
Conlangs: Scratchpad | Texts | antilanguage
Software: See http://bradrn.com/projects.html
Other: Ergativity for Novices

(Why does phpBB not let me add >5 links here?)
zompist
Site Admin
Posts: 2944
Joined: Sun Jul 08, 2018 5:46 am
Location: Right here, probably
Contact:

Re: Linguistic Miscellany Thread

Post by zompist »

bradrn wrote: Thu May 23, 2024 3:58 am
bradrn wrote: Wed May 22, 2024 5:53 pmEven English is arguably marked-nominative.
  • Citation form for the pronouns is the object series: me, us, them, etc.
  • The object series is used after prepositions
  • The object series is used for emphatic subjects and topics
  • The object series is used after be (and by extension in clefts), unlike most nominative-accusative languages (e.g. Latin) where the copula takes two nominative nouns — but like some other marked-nominative languages
  • The two sorts of possessive pronouns are obviously more closely related to the object series than the subject series (me, my, mine vs I; us, our, ours vs we)
Thanks for answering. I think this is suggestive (thus, arguable), but not convincing.

Citation form is just lexicography. The last point is just inherited from IE, so to be relevant you'd have to show that the argument holds for all of IE.

The others all apply to the disjunctive pronouns in French:

2. Il vient après moi.
3. Moi je le vois.
4. C'est moi.

Does this mean that these forms are marked in French? Not really; it means that they're the only ones that can be independent words. Je and me are both cliticized to the verb and can neither be separated out nor emphasized.

English is not the same, but perhaps it's on the same path: "I" used to be used in more situations, especially after be. It can still be stressed, but it's notable that we're more likely to answer "Who did it?" with "Me!" not "I!"

You'd probably have to explain what notion of markedness you're using. To me the form used in clefting and emphasis would be more rather than less marked. (Note that in French the clefted/emphatic form is different from both nom. and acc. pronouns.)
bradrn
Posts: 6257
Joined: Fri Oct 19, 2018 1:25 am

Re: Linguistic Miscellany Thread

Post by bradrn »

Starting with the most important point first:
zompist wrote: Thu May 23, 2024 4:22 am You'd probably have to explain what notion of markedness you're using. To me the form used in clefting and emphasis would be more rather than less marked. (Note that in French the clefted/emphatic form is different from both nom. and acc. pronouns.)
Essentially, by ‘less marked’, I mean ‘has a wider range of usage’. In a nominative-accusative or ergative-absolutive system, there are two core cases, both with their core function, but one of the cases will generally be the ‘default’ for usages outside that range too. Usually, people call call that the ‘less marked’ case. To justify this, there’s the very important fact that this almost always coincides with formal markedness: the null-marked case generally has the widest usage.

(Haspelmath has a paper where he makes a good case for simply saying, ‘has a wider range of usage’, when that’s what we mean. ‘Markedness’ is a horribly overloaded term which can mean all kinds of different things. But I think this is one place where it’s not unjustified, because like I said, both formal and functional markedness criteria point in the same direction here.)
Citation form is just lexicography. The last point is just inherited from IE, so to be relevant you'd have to show that the argument holds for all of IE.
Indeed: those two are the weakest arguments.

(And, besides, if PIE was active-stative like many people claim, then perhaps it would have been marked-S anyway?)
The others all apply to the disjunctive pronouns in French:

2. Il vient après moi.
3. Moi je le vois.
4. C'est moi.
The key difference here is that the disjunctive pronouns are, in general, different from both subject and object series (moi vs je/me): they are neither nominative nor accusative. (Indeed, in the one case where they overlap with the other series, elle is the nominative form!) Therefore, this gives no grounds to call French ‘marked-nominative’.
Does this mean that these forms are marked in French?
I think this notion of ‘markedness’ is just not a very useful concept in French. As I said, markedness in case-marking is generally related to which of the core cases is generalised the most beyond its core function. But in French, neither nominative nor accusative pronouns are generalised in any way at all. (There’s some syncretism with the dative pronouns, but that’s about it.) So I think I’d call the French bound pronouns equal in terms of markedness: the less marked forms would be the free forms, if any. (It’s very weird to think of emphatic pronouns as being ‘less marked’, but it falls out naturally from the definition as ‘wider usage range’.)
English is not the same, but perhaps it's on the same path: "I" used to be used in more situations, especially after be.
Wasn’t this mostly prescriptivists trying to make English work like Latin? I’m not sure if anyone ever actually talked like this naturally.
It can still be stressed, but it's notable that we're more likely to answer "Who did it?" with "Me!" not "I!"
Indeed, which was what I meant by them being the citation form (and also the emphatic form, for that matter).
Conlangs: Scratchpad | Texts | antilanguage
Software: See http://bradrn.com/projects.html
Other: Ergativity for Novices

(Why does phpBB not let me add >5 links here?)
zompist
Site Admin
Posts: 2944
Joined: Sun Jul 08, 2018 5:46 am
Location: Right here, probably
Contact:

Re: Linguistic Miscellany Thread

Post by zompist »

bradrn wrote: Thu May 23, 2024 5:13 am
zompist wrote: Thu May 23, 2024 4:22 am You'd probably have to explain what notion of markedness you're using. To me the form used in clefting and emphasis would be more rather than less marked. (Note that in French the clefted/emphatic form is different from both nom. and acc. pronouns.)
Essentially, by ‘less marked’, I mean ‘has a wider range of usage’. In a nominative-accusative or ergative-absolutive system, there are two core cases, both with their core function, but one of the cases will generally be the ‘default’ for usages outside that range too. Usually, people call call that the ‘less marked’ case. To justify this, there’s the very important fact that this almost always coincides with formal markedness: the null-marked case generally has the widest usage.

(Haspelmath has a paper where he makes a good case for simply saying, ‘has a wider range of usage’, when that’s what we mean. ‘Markedness’ is a horribly overloaded term which can mean all kinds of different things. But I think this is one place where it’s not unjustified, because like I said, both formal and functional markedness criteria point in the same direction here.)
I was confused because this conflicts with how "marked" is used in generative grammar.

"Default" might be a better word here... I do agree that "me" etc. is the default form in English (and "moi" in French)!
(And, besides, if PIE was active-stative like many people claim, then perhaps it would have been marked-S anyway?)
Sure, but then most IE languages went through a nom/acc phase... none of them have simply retained an active-stative system.
The key difference here is that the disjunctive pronouns are, in general, different from both subject and object series (moi vs je/me): they are neither nominative nor accusative. (Indeed, in the one case where they overlap with the other series, elle is the nominative form!) Therefore, this gives no grounds to call French ‘marked-nominative’.
Yes, but that wasn't what I was suggesting; I suggested that the English facts do resemble French. It wouldn't surprise me if there's some borrowing going on here. Note that you still say "Soy yo" in Spanish.
English is not the same, but perhaps it's on the same path: "I" used to be used in more situations, especially after be.
Wasn’t this mostly prescriptivists trying to make English work like Latin? I’m not sure if anyone ever actually talked like this naturally.
A little Googling suggests that "It is me" is not attested before 1500. (But a literature search would be needed to know how common it was.)
bradrn
Posts: 6257
Joined: Fri Oct 19, 2018 1:25 am

Re: Linguistic Miscellany Thread

Post by bradrn »

zompist wrote: Thu May 23, 2024 6:19 am "Default" might be a better word here... I do agree that "me" etc. is the default form in English (and "moi" in French)!
OK, ‘default’ is a far better word here. Then ‘marked-nominative languages’ might be better termed ‘default-accusative languages’, and so on.
The key difference here is that the disjunctive pronouns are, in general, different from both subject and object series (moi vs je/me): they are neither nominative nor accusative. (Indeed, in the one case where they overlap with the other series, elle is the nominative form!) Therefore, this gives no grounds to call French ‘marked-nominative’.
Yes, but that wasn't what I was suggesting; I suggested that the English facts do resemble French.
They resemble French only in that they have a default series of pronouns! (Which is probably a universal for all languages with pronouns.) The key part of my argument is that the default series of pronouns is in fact the object series, rather being the subject series (as in e.g. Latin) or a separate series (as in French).
It wouldn't surprise me if there's some borrowing going on here. Note that you still say "Soy yo" in Spanish.
I know nothing of Spanish; what does this mean?
English is not the same, but perhaps it's on the same path: "I" used to be used in more situations, especially after be.
Wasn’t this mostly prescriptivists trying to make English work like Latin? I’m not sure if anyone ever actually talked like this naturally.
A little Googling suggests that "It is me" is not attested before 1500. (But a literature search would be needed to know how common it was.)
Interesting, thanks! (But then, ’before 1500’ is already early Middle English… it’s debatable how comparable that is with Modern English.)
Conlangs: Scratchpad | Texts | antilanguage
Software: See http://bradrn.com/projects.html
Other: Ergativity for Novices

(Why does phpBB not let me add >5 links here?)
User avatar
jal
Posts: 939
Joined: Fri Aug 31, 2018 3:13 pm

Re: Linguistic Miscellany Thread

Post by jal »

zompist wrote: Wed May 22, 2024 5:22 pm(Well, really you have suppletive roots. But if you choose the sound change right, it can look like something is added to the nominative— e.g. if the sound change is that a final consonant is lost.)
That's indeed a cool way to do that, though you'd probably need a limited set of word-final sounds or else you'd have a gazillion different forms.


JAL
Zju
Posts: 911
Joined: Fri Aug 03, 2018 4:05 pm

Re: Linguistic Miscellany Thread

Post by Zju »

bradrn wrote: Wed May 22, 2024 5:53 pm
Zju wrote: Wed May 22, 2024 3:20 pm Got some inspiration: reflexive markers. They only refer to the subject. So if the word order in a reflexive clause is fixed to S REFL V (and reflexion is doubly marked), that could over time be interpreted as S=NOM V, after other case suffixes have appeared. Finally, the nominative enclitic would be grammaticalised to a suffix and spread to all subjects.
I don’t find this enormously plausible. You’d need a step in the middle where the reflexive becomes generalised to non-reflexive sentences [...]
I was thinking that the middle step would be that the reflexive marker would be interpreted as nominative marker in reflexive sentences first, and then spreads out elsewhere.

But I gave up for another, finicky reason: I want only animates to be marked for nominative, and even then maybe not in all cases. Maybe I could do something with an auxiliary verb or intensifier that had a similar distribution to begin with.
accusatives
The resulting morphosyntactical alignment doesn't have a proper accusative to speak of.
/j/ <j>

Ɂaləɂahina asəkipaɂə ileku omkiroro salka.
Loɂ ɂerleku asəɂulŋusikraɂə seləɂahina əɂətlahɂun əiŋɂiɂŋa.
Hərlaɂ. Hərlaɂ. Hərlaɂ. Hərlaɂ. Hərlaɂ. Hərlaɂ. Hərlaɂ.
User avatar
WeepingElf
Posts: 1510
Joined: Sun Jul 15, 2018 12:39 pm
Location: Braunschweig, Germany
Contact:

Re: Linguistic Miscellany Thread

Post by WeepingElf »

bradrn wrote: Thu May 23, 2024 7:24 am
zompist wrote: Thu May 23, 2024 6:19 am "Default" might be a better word here... I do agree that "me" etc. is the default form in English (and "moi" in French)!
OK, ‘default’ is a far better word here. Then ‘marked-nominative languages’ might be better termed ‘default-accusative languages’, and so on.
But marked-nominative language is the established term linguists use for this. Also, default in the meaning of 'fallback value if no other value is specified' is computerese. Only IT folks use it that way. The usual meaning is 'failure to do something', e.g. to meet a payment.
... brought to you by the Weeping Elf
My conlang pages
bradrn
Posts: 6257
Joined: Fri Oct 19, 2018 1:25 am

Re: Linguistic Miscellany Thread

Post by bradrn »

Zju wrote: Thu May 23, 2024 10:39 am
bradrn wrote: Wed May 22, 2024 5:53 pm
Zju wrote: Wed May 22, 2024 3:20 pm Got some inspiration: reflexive markers. They only refer to the subject. So if the word order in a reflexive clause is fixed to S REFL V (and reflexion is doubly marked), that could over time be interpreted as S=NOM V, after other case suffixes have appeared. Finally, the nominative enclitic would be grammaticalised to a suffix and spread to all subjects.
I don’t find this enormously plausible. You’d need a step in the middle where the reflexive becomes generalised to non-reflexive sentences [...]
I was thinking that the middle step would be that the reflexive marker would be interpreted as nominative marker in reflexive sentences first, and then spreads out elsewhere.
But what’s the motivation for it to be interpreted as a nominative marker? I suppose I could see it happening if there’s lots of ambitransitive verbs, but not otherwise.
accusatives
The resulting morphosyntactical alignment doesn't have a proper accusative to speak of.
Well, it does: it’s the unmarked case which is used for objects and other things. It’s as much of an accusative as the unmarked nominative of many languages is a nominative.
WeepingElf wrote: Thu May 23, 2024 11:13 am But marked-nominative language is the established term linguists use for this.
There’s lots of established terms that linguists use for things. Quite a lot of them are very bad. This one is better than many, but it’s still not great. (Haspelmath on so-called ‘markedness’ is worth a read.)
Also, default in the meaning of 'fallback value if no other value is specified' is computerese. Only IT folks use it that way.
I’m quite sure I’ve seen linguists use it this way too, at least on occasion.
Conlangs: Scratchpad | Texts | antilanguage
Software: See http://bradrn.com/projects.html
Other: Ergativity for Novices

(Why does phpBB not let me add >5 links here?)
Estav
Posts: 196
Joined: Sun Aug 12, 2018 10:22 am

Re: Linguistic Miscellany Thread

Post by Estav »

Zju wrote: Tue May 21, 2024 2:22 pm From what can a nominative case suffix or marker develop? Esp. if there were no nominative markers so far and there's few cases to begin with. Maybe a demonstrative or an intensifier?
Japanese ga (currently used as a nominative/subject marker, although there's a distinct topic marker wa) seems to have been used as a genitive marker (apparently mostly for animate specific nouns, including pronouns) in Old Japanese. How exactly these uses relate to each other isn't clear to me, but I found this explanation on reddit outlining the hypothesis that the nominative use was generalized from using the genitive to mark the subject of subordinate clauses whose head verb was treated similarly in some ways to a noun (compare the "subjective genitive" in English nominalizations, e.g. "the artist's creation of these works").
zompist
Site Admin
Posts: 2944
Joined: Sun Jul 08, 2018 5:46 am
Location: Right here, probably
Contact:

Re: Linguistic Miscellany Thread

Post by zompist »

bradrn wrote: Thu May 23, 2024 7:24 am The key part of my argument is that the default series of pronouns is in fact the object series, rather being the subject series (as in e.g. Latin) or a separate series (as in French).
I don't disagree with this observation, but my point is that both English and French are generalizing the default pronoun— making it more default, as it were. Which is why I compared their behavior to Spanish.

You could probably argue that the default pronoun in Spanish is the nominative (e.g. yo). Some examples:

Soy yo = It's me.
Fui yo quien hice la cena. = It was me who made dinner.
—¿Quien hice esto? —Yo. = "Who did this?" "Me."

Nothing is ever quite parallel... you use a disjunctive pronoun after a preposition (para mí = for me), and so far as I know there's no Constituent Dislocation ("Me, I won't go").
Travis B.
Posts: 6853
Joined: Sun Jul 15, 2018 8:52 pm

Re: Linguistic Miscellany Thread

Post by Travis B. »

Estav wrote: Thu May 23, 2024 1:23 pm
Zju wrote: Tue May 21, 2024 2:22 pm From what can a nominative case suffix or marker develop? Esp. if there were no nominative markers so far and there's few cases to begin with. Maybe a demonstrative or an intensifier?
Japanese ga (currently used as a nominative/subject marker, although there's a distinct topic marker wa) seems to have been used as a genitive marker (apparently mostly for animate specific nouns, including pronouns) in Old Japanese. How exactly these uses relate to each other isn't clear to me, but I found this explanation on reddit outlining the hypothesis that the nominative use was generalized from using the genitive to mark the subject of subordinate clauses whose head verb was treated similarly in some ways to a noun (compare the "subjective genitive" in English nominalizations, e.g. "the artist's creation of these works").
In the case of English, we see in everyday English a "subjective accusative" in the place of the "subjective genitive" (e.g. "If it weren't for me setting up the firewall this Windows box'd be full of worms by now.")
Yaaludinuya siima d'at yiseka wohadetafa gaare.
Ennadinut'a gaare d'ate eetatadi siiman.
T'awraa t'awraa t'awraa t'awraa t'awraa t'awraa t'awraa.
Travis B.
Posts: 6853
Joined: Sun Jul 15, 2018 8:52 pm

Re: Linguistic Miscellany Thread

Post by Travis B. »

We also see the subjective accusative in things like "me and my friends rode around town in my pickup truck removing stop signs".
Yaaludinuya siima d'at yiseka wohadetafa gaare.
Ennadinut'a gaare d'ate eetatadi siiman.
T'awraa t'awraa t'awraa t'awraa t'awraa t'awraa t'awraa.
Richard W
Posts: 1471
Joined: Sat Aug 11, 2018 12:53 pm

Re: Linguistic Miscellany Thread

Post by Richard W »

Travis B. wrote: Thu May 23, 2024 4:52 pm We also see the subjective accusative in things like "me and my friends rode around town in my pickup truck removing stop signs".
I think that's just an example of the nominative being triggered by an adjacent verb, with fallback to the unmarked accusative when separated.
Travis B.
Posts: 6853
Joined: Sun Jul 15, 2018 8:52 pm

Re: Linguistic Miscellany Thread

Post by Travis B. »

Richard W wrote: Thu May 23, 2024 5:20 pm
Travis B. wrote: Thu May 23, 2024 4:52 pm We also see the subjective accusative in things like "me and my friends rode around town in my pickup truck removing stop signs".
I think that's just an example of the nominative being triggered by an adjacent verb, with fallback to the unmarked accusative when separated.
Which only goes to show that English is a marked-nominative language.
Yaaludinuya siima d'at yiseka wohadetafa gaare.
Ennadinut'a gaare d'ate eetatadi siiman.
T'awraa t'awraa t'awraa t'awraa t'awraa t'awraa t'awraa.
bradrn
Posts: 6257
Joined: Fri Oct 19, 2018 1:25 am

Re: Linguistic Miscellany Thread

Post by bradrn »

zompist wrote: Thu May 23, 2024 4:16 pm […] my point is that both English and French are generalizing the default pronoun— making it more default, as it were.
Well, my definition of ‘default pronoun’ was precisely that it is generalised. So wouldn’t this just be a circular argument?
You could probably argue that the default pronoun in Spanish is the nominative (e.g. yo). Some examples:

Soy yo = It's me.
Fui yo quien hice la cena. = It was me who made dinner.
—¿Quien hice esto? —Yo. = "Who did this?" "Me."

Nothing is ever quite parallel... you use a disjunctive pronoun after a preposition (para mí = for me)
Wait, Spanish has both disjunctive pronouns and nominative pronouns?
[…] and so far as I know there's no Constituent Dislocation ("Me, I won't go").
Now this surprises me a lot. I seem to recall a suggestion (from Payne, I think) that this sort of topicalisation is a universal.
Travis B. wrote: Thu May 23, 2024 4:30 pm In the case of English, we see in everyday English a "subjective accusative" in the place of the "subjective genitive" (e.g. "If it weren't for me setting up the firewall this Windows box'd be full of worms by now.")
Travis B. wrote: Thu May 23, 2024 4:52 pm We also see the subjective accusative in things like "me and my friends rode around town in my pickup truck removing stop signs".
Yes, this ‘subjective accusative’ is precisely what I mean by ‘English being marked-nominative / default-accusative’. The nominative is restricted specifically to pronominal subjects of finite verbs, and the accusative is used in all other situations, including these.

(Also, now I feel a bit stupid for forgetting about conjunctions in my list a few posts ago. They’ve long been a bugbear of prescriptivists…)
Travis B. wrote: Thu May 23, 2024 5:22 pm
Richard W wrote: Thu May 23, 2024 5:20 pm I think that's just an example of the nominative being triggered by an adjacent verb, with fallback to the unmarked accusative when separated.
Which only goes to show that English is a marked-nominative language.
Exactly!
Conlangs: Scratchpad | Texts | antilanguage
Software: See http://bradrn.com/projects.html
Other: Ergativity for Novices

(Why does phpBB not let me add >5 links here?)
Post Reply