Grammaticalization Quickie Thread

[Rounin Ryuuji]

Back-formation and reanalysis are wonderful things.

[StrangerCoug]

So I've redone the plain pronouns as follows, now with proper plural forms based on the suggested ideas:

Person	Singular	Plural
1st	ze /t͡se/	zid /t͡sit/
2nd	re /re/	rid /rit/
3rd animate	ne /ne/	nid /nit/
3rd inanimate	na /na/	nad /nat/

At least ridce is still around as the second person polite pronoun, probably for both singular and plural. Zidce I'm no longer likely to keep as a first-person polite pronoun based on feedback, but it may survive as having gone semantic shift from being an arrogant way to refer to oneself in the first person to no longer being in serious use as a pronoun (though still a rude term).

Now, as regards the earlier discussion about demonstratives, there's supposed to be a pattern: the proximal ones all begin with z, the medial ones all begin with r, and the distal ones all begin with n. I went with the idea of having na/ne doing double duty as demonstrative determiners. For the demonstrative adjectives, in the nominative inanimate singular I currently have zotłʼa /ˈt͡so.t͡ɬʼa/ for the proximal, rotłʼa /ˈro.t͡ɬʼa/ for the medial, and notłʼa /ˈno.t͡ɬʼa/ for the distal. (Change the final –a in all three to –ad for the inanimate plural, to –e for the animate singular, and to –id for the animate plural.) Given that, would you expect inanimate demonstrative determiners za and ra to exist, and if so, would it be confusing to have ze and re also be animate demonstrative determiners?

[Rounin Ryuuji]

Given that, would you expect inanimate demonstrative determiners za and ra to exist, and if so, would it be confusing to have ze and re also be animate demonstrative determiners?

I think they very well could. You could even use "za" or "ra" as impolite first- and second-person pronouns, respectively.

[hwhatting]

Given that, would you expect inanimate demonstrative determiners za and ra to exist, and if so, would it be confusing to have ze and re also be animate demonstrative determiners?

If the French can live with le as male and la as female article / object pronoun, your conpeople will survive having a ze/re "animate" vs. za / ra inanimate contrast.

[Ahzoh]

I have a proto language where animacy is marked by stress where inanimate nouns are penultimate or word-initial stress while animate nouns are word-final stress (presumably because of adding affixes to nouns to make them animate). Said language lacks case-marking and prototypically marks the plural with the suffix /i/.

So I want to create a declension class system for these two word classes but I don't have a clue where the affixes could come from to create this difference in declension. The descending language is split ergative where animate nouns are nom-acc (doing is less marked) and inanimate nouns are erg-abs (being done to is less marked).

My end goal is that there is a declension class for inanimate nouns where the absolutive singular is -0 (null) and the plural is -ь /ʲ/ (сьол "lake" / сьоль "lakes") and there is a declension class for animate nouns where the nominative singular is -а /a/ and the plural is -е /e/ (сьола "river" / сьоле "rivers") and another for animate nouns where the endings are -у /u/ and -оі /oj/ (тьху "human" / тьхоі "humans").
But I hope to come up with more classes, but I just don't know where the rest of the affixes can come from.

I do have sound changes for how some classes came into existence:

More: show

Code: Select all

Unstressed:
-æ > -ʲə > -ʲ
-e > -ʲə > -ʲ
-ø > -ʲə > -ʲ
-i > -ʲə > -ʲ
-y > -ʲə > -ʲ
-ɒ > -ə  > -0
-ɤ > -ə  > -0
-o > -ə  > -0
-ɯ > -ə  > -0
-u > -ə  > -0
Stressed:
-æ > -ʲa
-e > -ʲe
-ø > -ʲo
-i > -ʲi
-y > -ʲu
-ɒ > -a
-ɤ > -e
-o > -o
-ɯ > -i
-u > -u
Long (stressed and unstressed):
-æː > -ʲa
-eː > -ʲe
-øː > -ʲo
-iː > -ʲi
-yː > -ʲu
-ɒː > -a
-ɤː > -e
-oː > -o
-ɯː > -i
-uː > -u

Code: Select all

Inanimate:
/ˈsøː.lɒ/ "lake"   > /ˈsʲo.lə/  > /ˈsʲol/
/ˈsøː.lɒj/ "lakes" > /ˈsʲo.ləj/ > /ˈsʲolʲ/

Animate:
/søː.ˈlɒ/ "river"   > /sʲo.ˈla/  > /sʲo.ˈla/
/søː.ˈlɒj/ "rivers" > /sʲo.ˈlaj/ > /sʲo.ˈle/

[Rounin Ryuuji]

I think you're on your way to figuring it out. The route will be to invent adpositions and cause them to coalesce into either the beginnings or ends of nouns. This also works with verbs (a verb noting location can become a genitive or locative particle, one noting direction can become a dative or lative, and so on and so on). After this, apply a healthy dose of sound change and analogical levelling, and you ought to have cases!

[Ahzoh]

Hmm, so if I want declension classes where some nouns take different affixes than nouns of other classes for a similar grammatical category I'd have to come up with situations where one noun uses one adposition and another noun uses another adposition.

As it is, the difference between the -а ending class and the -у ending class is that the latter is used primarily for nouns involving humans.

[cedh]

Hmm, so if I want declension classes where some nouns take different affixes than nouns of other classes for a similar grammatical category I'd have to come up with situations where one noun uses one adposition and another noun uses another adposition.

As it is, the difference between the -а ending class and the -у ending class is that the latter is used primarily for nouns involving humans.

This sounds like a good start to me. I don't know what kinds of cases you want to have, but animacy and human/non-human are distinctions that can easily be associated with using different adpositions (or even using adpositions at all, cf. the Spanish use of a as an accusative marker for human direct objects only). Just to sketch out a general idea, you could, for instance, grammaticalize a roughly genitive/ablative/instrumental case from something like "by" for humans, something like "of" for non-human animates, and something like "with" or "from" for inanimates. And you could grammaticalize a roughly dative/allative/benefactive case from something like "for" for humans and something like "to" for non-humans.

[Oxygenman]

I am hoping to diachronically develop a “light verb” system where there is a small, closed class of inflecting verbs that combine with some other element to form a complex predicate, similar to the indigenous languages of Australia. Assuming this was not the system used in the protolanguage, what exactly happens to the proto-verbs that no longer inflect? For example if “speak” is now “speaking do” does the infinitive “speak” fall out of use and functionally cease to exist?

[bradrn]

I am hoping to diachronically develop a “light verb” system where there is a small, closed class of inflecting verbs that combine with some other element to form a complex predicate, similar to the indigenous languages of Australia. Assuming this was not the system used in the protolanguage, what exactly happens to the proto-verbs that no longer inflect? For example if “speak” is now “speaking do” does the infinitive “speak” fall out of use and functionally cease to exist?

Ah, light verbs — my favourite topic! Unfortunately, off the top of my head I can’t recall seeing anything about their evolution. Prima facie I’d expect the verbs to nominalise and turn into nouns, but that’s just a guess, and then again I don’t recall anyone mentioning the presence of any fossilised morphology in Australian noninflecting verbs. Pawley 2006 also suggests that this isn’t the case, though admittedly his focus is more on generic verbs than noninflecting verbs. I’ll do some more thorough reading once I get home and have access to my resources again.

[Zju]

Couldn't just periphrastic tenses become more common and eventually displace simple tenses? For all but a handful of verbs, that is.

[Travis B.]

Couldn't just periphrastic tenses become more common and eventually displace simple tenses? For all but a handful of verbs, that is.

We've already seen this with colloquial German, where in many varieties the preterite has largely been lost except for a small set of verbs, and has been replaced with the perfect outside of such verbs.

[Zju]

Couldn't just periphrastic tenses become more common and eventually displace simple tenses? For all but a handful of verbs, that is.

We've already seen this with colloquial German, where in many varieties the preterite has largely been lost except for a small set of verbs, and has been replaced with the perfect outside of such verbs.

But does the same hold true for present tenses as well?

[Travis B.]

Couldn't just periphrastic tenses become more common and eventually displace simple tenses? For all but a handful of verbs, that is.

We've already seen this with colloquial German, where in many varieties the preterite has largely been lost except for a small set of verbs, and has been replaced with the perfect outside of such verbs.

But does the same hold true for present tenses as well?

A good example there is Japanese, where new verbs are formed from nouns with する suru and the set of verbs is rather closed. All one needs then is for verbs to start being nominalized and then used with such a verbifier, which to me is very plausible.

[Rounin Ryuuji]

The Japanese example that came to mind for a present tense was -ている/-てる (te iru), the present progressive form of the verb (which seems to be used more broadly than the English form) — "I love you" is often 愛してる (aishiteru), literally "I am loving you".

As far as nominalisation goes, the i-stem of a godan verb is already very frequently a noun. I could easily see this fossilising and getting -ru, -suru, or -te(i)ru attached as a verbaliser (with regular -teru possibly replacing irregular -suru as the productive form).

[StrangerCoug]

Is it attested for there to be distinct words for "to be" and "to be so"? I don't want to give my language dedicated words for "yes" or "no," and I want to give them pro-verbs that they can use to respond to polar questions instead. (They can also employ echo responses.)

[Kuchigakatai]

Is it attested for there to be distinct words for "to be" and "to be so"? I don't want to give my language dedicated words for "yes" or "no," and I want to give them pro-verbs that they can use to respond to polar questions instead. (They can also employ echo responses.)

Classical Chinese 為 wéi 'to be' vs. 然 rán 'to be so' could maybe count. Relatedly, the common Mandarin word for 'of course, certainly' is 當然 dāngrán, etymologically "when it is so".

You could say your conlang's equivalents of yes/no are simply verbs meaning "to affirm, say yes" and "to deny, say no".

[Oxygenman]

TW: N00b struggling through first conlang.

I'm trying to work out a grammatical change where an old genitive case becomes a fossilized ending on nouns that become a closed class of "adjectives" and then the old possessive sense of the genitive is replaced by an old ablative case.

I wanted the adjectives to be the "marked" word in the phrase and to have the marking be suffixes, but am thinking that this conflicts with my intent for this language to be SOV and dependent-marking. If I wanted suffixes, wouldn't it need to be head-marking? Would it change if I derived the marking from a noun vs. a verb? I started trying to gloss a sentence but don't know if I'm on the right track here.

Dog old-GEN man-ABL young-GEN cat black-GEN chased
The young man's old dog chased the black cat

[Travis B.]

TW: N00b struggling through first conlang.

I'm trying to work out a grammatical change where an old genitive case becomes a fossilized ending on nouns that become a closed class of "adjectives" and then the old possessive sense of the genitive is replaced by an old ablative case.

I wanted the adjectives to be the "marked" word in the phrase and to have the marking be suffixes, but am thinking that this conflicts with my intent for this language to be SOV and dependent-marking. If I wanted suffixes, wouldn't it need to be head-marking? Would it change if I derived the marking from a noun vs. a verb? I started trying to gloss a sentence but don't know if I'm on the right track here.

Dog old-GEN man-ABL young-GEN cat black-GEN chased
The young man's old dog chased the black cat

There are languages with an underlying SOV order but with Noun-Genitive order - take Standard German for instance (yes, at the surface StG is V2, but that is a minor detail here).

[Rounin Ryuuji]

I'm trying to work out a grammatical change where an old genitive case becomes a fossilized ending on nouns that become a closed class of "adjectives" and then the old possessive sense of the genitive is replaced by an old ablative case.

I don't see why this couldn't happen. Genitive forms are frequently used as adjectives, and positional markers can be used to note both description and possession, especially if you had your old accusative shift to an oblique with directional markers accompanying it, with the ablative case having already had some overlap with the genitive and so coming to take the place of it (especially if it were formerly used in conjunction with a possessive particle that was later dropped or cliticised and eroded away).

I wanted the adjectives to be the "marked" word in the phrase and to have the marking be suffixes, but am thinking that this conflicts with my intent for this language to be SOV and dependent-marking. If I wanted suffixes, wouldn't it need to be head-marking? Would it change if I derived the marking from a noun vs. a verb?

I'm not so sure on this point.

I started trying to gloss a sentence but don't know if I'm on the right track here.

Dog old-GEN man-ABL young-GEN cat black-GEN chased
The young man's old dog chased the black cat.

Looks fine to me.