Random Conlang Grammar Ideas Thread

[Ælfwine]

Speaking of romance...

I am tempted to go back to my old romlang idea, which I named either Pelsodian (Lacus Pelsodis, or Lake Balaton in Hungry), or Castellese (Named after Keszthely, or the city it survives in).

I envisioned the language being halfway between Rhaetoromance and Romanian, with certain features of both...however research has led me to conclude that it would be more Western than Eastern. I may waive part of this due to artistic liberty, but I'll see.

Anyway, this language has an interesting (and necessarily early, which may work for its location) sound change: Vulgar Latin -u (from -um) retained its nasalization and lowered to -a from -õ. On the other hand -s becomes vocalized to -j as it had done in Italian and Romanian, leading to endings like -us > -os > -oj > -e.

This gives the language two interesting grammatical properties (relative to most other romance languages) 1.) the neuter gender is preserved while the masculine gender merges with the feminine and 2.) The nominative case is kept distinct from the accusative (but not in all declensions), although the genitive and dative still merge. This is a sort of halfway compromise between the old French system (keeps the nominative distinct but merges the accusative, dative, and genitive) and Romanian (merges the nominative with the accusative and dative with the genitive.)

This paper also suggests a similar merger of the cases, but unfortunately I cannot read Hungarian to see why that is.

[Pabappa]

http://verduria.org/viewtopic.php?p=29134#p29134

An all-instinct language such that babies are born fully fluent, but cannot learn even a single new word or grammatical construct during their entire lives. This could lead to a species that is highly intelligent but cannot efficiently make tools, since they would only be able to ever describe their tools indirectly, and could not use tools to make other tools. If I use this idea it will be for a non-human culture, but it would be cute to see this idea explored for a human culture.

[Vilike]

That reminds me of the Archer Marmousets from Worldbuilding Notes' video: River Basin | Archer Marmouset

[Pabappa]

Oh, very nice idea. Both the language and the idea of an animal using tool-making and bodily weapons interdependently. I have to admit, I dont think that idea would have ever occurred to me ... all my animals are either the stereotypical sharp-teeth/sharp-claws natural predator type, or the paws-in-the-air tool using type. For a while I was using wasps that could shoot their stingers, but I straight-up stole that from a science fiction story (where the victim then became pregnant with the wasp larvae), and I suspect even he wasnt the first one to think of it. Even that still isnt the same, though, because those wasps never used tools to enhance or increase their aim.

[Kuchigakatai]

That reminds me of the Archer Marmousets from Worldbuilding Notes' video: River Basin | Archer Marmouset

The voice synthesizer sounds so weird yet amusing, particularly the part of needing to use a falling tone in every sentence-final word stressed on the last or only syllable, even extending consonants to accomplish it ("adults." [əˈdɐɫɫts˥˨], "for the string." [stɹɪŋŋŋ˥˨]), and otherwise lengthening the syllable in the same position ("or shooters." [ˈʃʊtəəəz], "for this video." ["vɪdioooʊ], "colonies." [ˈkʰɒləniiiz], but not in "and an illustrator." [ˈɪləstɹeɪtə]).

[Pabappa]

Thats not her voice??? I thought she just had a strange accent.

[Vilike]

Thats not her voice??? I thought she just had a strange accent.

She's Polish. I find her voice soothing.

[mèþru]

That is her voice. I used to share a Discord server with her and we talked on mic.

[Bob]

I read a lot of these and they're a lot of good ideas.

[Kuchigakatai]

A form of future English in which verbs have been regularized further quite a bit:

take took taken
shake shook shaken
make mook maken
break brook braken
wake up, wook up, waken up

speak spoke spoken
sneak snoke snoken
seek soke soken

write wrote written
fight fote fitten
light up, lote up, litten up

[Moose-tache]

Some of these seem more likely than others.

People already innovate past participial forms (my grandma always said "boughten"). But I think many native speakers are less aware of the patterns in simple past forms. For example, how many speakers realize that woke "should be" wook by analogy? And how is made going to analogically become "mook" when "maked" is sitting right there?

In my opinion, the most likely subjects of analogy are not the ones that are etymological strong verbs, but weak verbs that sound like strong verbs despite having no historical relationship. For example, I frequently have to stop myself from saying "succame" instead of succumbed. But conlangers usually overlook these cases because they're not obvious in writing.

[Pabappa]

My mother consistently said "skun" for the past form of skin. Not that it came up often, but she's never used made-up forms like that in other situaitons, so to her the word must be real.

it would make sense to just use the most common verb in the lexicon and build the rhymes based on that. if there is no obvious choice, it might not work. e.g. "take" is probably just barel more common than "make", so while it oculd work, i think thsi group would probably split into two. on the other hand, "speak" has a clear dominance over its rhymes, so even though "snuck" is established colloquially it could cede its paradigm to the wider pattern if the whole language was changing.

we could look into kids' speech too .... do kids sometimes say "maded" instead of "maked"? if they are seeing the past tense form as basic, then maybe that would spread to othger verbs as well.

[Travis B.]

we could look into kids' speech too .... do kids sometimes say "maded" instead of "maked"? if they are seeing the past tense form as basic, then maybe that would spread to othger verbs as well.

I have heard my daughter say "hadded" intead of "had" on occasion, and she's eleven.

[Richard W]

We could look into kids' speech too .... do kids sometimes say "maded" instead of "maked"? if they are seeing the past tense form as basic, then maybe that would spread to othger verbs as well.

That doesn't make the simple past the basic form. It's just a rule that the past (or simple past?) ends in ///ed///. My daughter went through a phase of applying that rule, but I can't remember how much it was applied to irregular weak verbs. Another striking feature is that the simple past is sometimes used for the past participle, and that's not just in children's speech.

[Pedant]

A long time ago I had a language called Kroldistani which marked verbs not for past, present or future but for different times of the day (morning, noon, afternoon, night, and a neutral period). I’ve tried looking for equivalents, but so far nothing. Has anyone ever tried this out before, or would it seem silly?

[Kuchigakatai]

A long time ago I had a language called Kroldistani which marked verbs not for past, present or future but for different times of the day (morning, noon, afternoon, night, and a neutral period). I’ve tried looking for equivalents, but so far nothing. Has anyone ever tried this out before, or would it seem silly?

Standard Arabic has أصبح ʔasˤbaħa 'become by the morning', أمسى ʔamsaa 'become by the evening' and بات baata 'become during the night' (think of how useful they are to say e.g. 'I woke up with a headache / sick', 'by the end of the day I was exhausted'), besides the more neutral صار sˤaara 'become'.

These are the only examples I'm aware of such a thing in Arabic though, and note these verbs are related to صباح sˤabaaħ 'morning', مساء masaaʔ 'evening' and مبيت mabiit 'overnight stay' (also بيت bait 'house'). In the case of ʔasˤbaħa and ʔamsaa, these have been derived via the ʔaCCaCa transfix, which usually derives causative verbs from basic verbs (علم ʕalima 'know sth' > أعلم ʔaʕlama 'inform sb, let sb know'), but as in this case also intransitive verbs from nouns and adjectives (ذنب‎ ðanb 'sin' > أذنب‎ ʔaðnaba 'to sin', حسن ħasan 'good, nice; good-looking' > أحسن ʔaħsana 'to act wisely, do good'). On the other hand, baata 'become during the night' is a basic type of verb though.

Also, as far as modern Standard Arabic goes, أصبح ʔasˤbaħa has largely lost its morning time connotation, and has now even become the default verb to express 'become', with صار sˤaara still around as a very common synonym.

[dɮ the phoneme]

A language with SOV word order and prepositions, where prepositional phrases are frequently post-verbal (à la Persian). There are also lots of phrasal verbs consisting of verb + preposition. Eventually the prepositions fuse with the verb, perhaps even eroding phonologically to some degree, but the word order is maintained. Thus you're left with two classes of verb, one that selects for SOV order and one that selects for SVO order.

[missals]

For a little bit now, after reading about allocutive agreement in Beja ("a morphological feature in which the gender of an addressee is marked overtly in an utterance using fully grammaticalized markers even if the addressee is not referred to in the utterance"), I've speculated about a language that marks gender of the speaker, addressee, and referent, ideally on every single word in the language.

All three of those things exist irl, though I'm not certain if they all exist at once in the same language, or if so, if they'd ever all be marked on the same word at the same time (let alone all words). But it seems plausible.

Since speaker/referent and addressee/referent frequently overlap, this might be tricky to implement. Most realistically, it might be through parallel and partially overlapping systems of marking. But I tried my hand at a tightly-integrated paradigmatic system. These are verbal suffixes, and nouns would be marked in a similar way. Number is distinguished for referent only. In classic west Eurasian manner, mixed-sex groups are always referred to with the masculine plural and the masculine singular serves as the generic form.



In this language, stress is penultimate (ignoring epenthetic vowels), and the only possible word-final vowels are /i/ and /ə/ (spelled a). All words that end in /i/ have a feminine referent, and all words that end in /ə/ have a masculine referent. Word-internal coda consonants are permitted, but no consonants may occur word-finally. Thus, is it phonologically impossible for a word to not to be explicitly masculine or feminine in form, save perhaps for some particles or clitics that undergo contraction.

Since that huge chart possibly doesn't make any sense (I confused myself several times while creating that table) I'll demonstrate it with a made-up verb root, hab- 'walk'. Epenthetic /i/ breaking up impermissible consonant clusters is colored red.

habma 'I (a man) am walking' (addressed to a man)
habimta 'I (a man) am walking' (addressed to a woman)
habmi 'I (a woman) am walking' (addressed to a man)
habimti 'I (a woman) am walking' (addressed to a woman)

habuma 'We (a group of men or a mixed-sex group) are walking' (addressed to a man)
habumta 'We (a group of men or a mixed-sex group) are walking' (addressed to a woman)
habumi 'We (a group of women) are walking' (addressed to a man)
habumti 'We (a group of women) are walking' (addressed to a woman)

habsa 'you (a man) are walking' (spoken by a man)
habsti 'you (a woman) are walking' (spoken by a man)
habinsa 'you (a man) are walking' (spoken by a woman)
habinsti 'you (a woman) are walking' (spoken by a woman)

habusa 'you all (a group of men or a mixed-sex group) are walking' (spoken by a man)
habusti 'you all (a group of women) are walking' (spoken by a man)
habunsa 'you all (a group of men or a mixed-sex group) are walking' (spoken by a woman)
habunsti 'you all (a group of women) are walking' (spoken by a woman)

haba 'he is walking' (spoken by a man, addressed to a man)
habta 'he is walking' (spoken by a man, addressed to a woman)
habna 'he is walking' (spoken by a woman, addressed to a man)
habinta 'he is walking' (spoken by a woman, addressed to a woman)

habi 'she is walking' (spoken by a man, addressed to a man)
habti 'she is walking' (spoken by a man, addressed to a woman)
habni 'she is walking' (spoken by a woman, addressed to a man)
habinti 'she is walking' (spoken by a woman, addressed to a woman)

habwa 'they (a group of men or a mixed-sex group) are walking' (spoken by a man, addressed to a man)
habuta 'they (a group of men or a mixed-sex group) are walking' (spoken by a man, addressed to a woman)
habuna 'they (a group of men or a mixed-sex group) are walking' (spoken by a woman, addressed to a man)
habunta 'they (a group of men or a mixed-sex group) are walking' (spoken by a woman, addressed to a woman)

habwi 'they (a group of women) are walking' (spoken by a man, addressed to a man)
habuti 'they (a group of women) are walking' (spoken by a man, addressed to a woman)
habuni 'they (a group of women) are walking' (spoken by a woman, addressed to a man)
habunti 'they (a group of women) are walking' (spoken by a woman, addressed to a woman)

[Neonnaut]

My mother consistently said "skun" for the past form of skin. Not that it came up often, but she's never used made-up forms like that in other situaitons, so to her the word must be real.

It is real to me, and others in my region too, but only to living things and injuries, like "I skun my knee".

[TomHChappell]

All three of those things exist irl, though I'm not certain if they all exist at once in the same language, or if so, if they'd ever all be marked on the same word at the same time (let alone all words). But it seems plausible.

Do each pair of them co-exist in some natlang, possibly a different natlang for each pair?