Conlang Random Thread

[Rounin Ryuuji]

Anyone ever wondered (in conlanging terms) what English might have looked like if the Norman conquest had exerted more influence on English phonology and grammar but less on vocabulary?

I think it might be fun to produce an Anglic language matching that description. As a quick thought, for a few changes, I might go with these:

1. The dialect of Middle English that became dominant had /ew iw/ > [øː yː], giving French /ø y/ sounds to which they could be readily mapped when they were borrowed. Consequently, Middle English has forms like boef, boeuf, beuf ([bøːf]);
2. "Th-fronting" is universalised, eventually merging with existing /t d/ to [t̪ d̪];
3. [x] > [f] is universalised, often with the preceding vowel becoming long.

So, by the time of the Great Vowel Shift, we have the following vowel inventory:

1. High Front Vowel Merger: /iː yː/ merge to [iɥ]. resulting in an early modern vie-vue [viɥ] (vie-view) merger.
2. Front rounded /øː/ diphthongisies to [øj], resulting in early modern beuif [bøjf].
3. To fill the gaps they leave behind, existing /oː ou əu/ become /øː/, and /uː/ becomes /yː/, resulting in Early Modern schoe, schoo, scheu ("shoe"), gloe, gloo, gleu [gløː] ("glow"), schaddeu ("shadow"), and rume ([ɾyːm]) "room" and fluer [flyːəɾ] "flower"
4. [ɔː au] raise to [oː ɔː], then to [uː oː], producing Early Modern roust [ɾuːst] "roast", laue [loː] "law".
5. [eː ei] merge to [iː] to fill the space left by the high front vowel merger, yielding modern seen, sine, sein [siːn] "seen", leif [liːf] "lea, meadow".
6. [aː ɛː] become [ɛː eː], yielding Early Modern state [stɛːt]. hete [heːt] "heat"
7. To fill the gap left by [aː], [ɔi] unrounds to [ai], and then smooths to [aː], yielding Early Modern cane [kaːn] "coin", taal [taːl] "toil"
8. [ui] becomes [ɥiː], yielding early modern buile [bɥiːl] "boil".

This would, in the end, be mostly just me changing the great vowel shift to make a vowel system more to my liking, but if you want to make an Anglic language that feels more French, then there's no reason not to.

[jal]

No, because that's not how language contact works. Vocabulary is in most situations the easiest thing to change.

I'm going to disagree by proxy. I've read that there's two situations in which languages might influence each other. The first is contact where speakers of language A have contact with speakers of language B, but do not speak language B very well (at most at pidgin level). In this scenario, language A is likely to borrow only vocabulary (predominently nouns) from language B. The second is contact where speakers of language A are also near-fluent in language B. In this scenario, language A is more likely to borrow grammatical structures, though not so much vocabulary. It has been hypothesized that this is what happened between Anglo-Saxon and Celtic, the former borrowing the do-questions and negations.

I'm not sure whether the above is more than a hypothesis of someone, or whether there's more support for it.


JAL

[bradrn]

No, because that's not how language contact works. Vocabulary is in most situations the easiest thing to change.

I'm going to disagree by proxy. I've read that there's two situations in which languages might influence each other. The first is contact where speakers of language A have contact with speakers of language B, but do not speak language B very well (at most at pidgin level). In this scenario, language A is likely to borrow only vocabulary (predominently nouns) from language B. The second is contact where speakers of language A are also near-fluent in language B. In this scenario, language A is more likely to borrow grammatical structures, though not so much vocabulary. It has been hypothesized that this is what happened between Anglo-Saxon and Celtic, the former borrowing the do-questions and negations.

I feel this is oversimplified — there are more interesting situations than just ‘A is fluent in B’ vs ‘A is not fluent in B’.

For instance, I do know of one circumstance where grammar is borrowed but vocabulary is not: situations where languages are in intense contact, but people feel a strong sense of linguistic pride. (This famously happens in e.g. the Vaupés River basin, where the various communities identify with their own languages, but also practice linguistic endogamy. I suppose Kupwar qualifies too.) Under such circumstances, people try to avoid ‘contaminating’ their languages with others’ vocabulary… but grammars tend to converge anyway, because this is less obvious to the speakers themselves.

[Otto Kretschmer]

Didn't Yola and Fingalian also have Irish influence in phonology due to long standing contact with Irish?

https://youtu.be/NpKbY3P860o?si=7u0hWFp71Ra9zzYH

[jal]

but grammars tend to converge anyway, because this is less obvious to the speakers themselves.

Iirc there's an African(?) city where two languages are spoken that have almost no shared vocabulary, but almost identical grammar. Can't recall which city, unfortunately, and a quick google doesn't show anything useful.


JAL

[bradrn]

but grammars tend to converge anyway, because this is less obvious to the speakers themselves.

Iirc there's an African(?) city where two languages are spoken that have almost no shared vocabulary, but almost identical grammar. Can't recall which city, unfortunately, and a quick google doesn't show anything useful.

I did mention Kupwar already, but that’s in Asia. (Apparently the name for this is ‘metatypy’.)

[malloc]

Despite my best efforts, the accent system in my conlang refuses to work. Every attempt to make it interesting and aesthetically pleasing simply results in contrivance and contradiction. I really want to avoid the plodding thud of stress while also avoiding the complication of tone. Pitch accent seems appealing as an alternative, but the polysynthetic grammar means that most morphemes will surface without accent, undermining its ability to distinguish morphemes.

My current proposal is assigning accent regularly like many stress systems but declaring by fiat that it will surface as pitch. Developing interesting rules has proven difficult, however, since the principles involved keep resulting in contradictions. I initially planned on something like the Latin stress rule but with final superheavy syllables taking accent. This makes sense if understood as accenting the antepenultimate mora, with some adjustments for syllable structure. Yet when applying optimality theory (which seems the industry standard in phonology research) this system falls apart. No matter how I rank the constraints involved, they keep resulting in some forms having counterintuitive accent.

For instance, one ranking of constraints implies that taatataa should receive penultimate rather than antepenultimate accent. Other rankings present serious problems for short words, implying that they cannot be footed or accenting them on the wrong syllable.

[Creyeditor]

This might be more of a problem for Optimality Theory than a problem for your conlang. OT has some emprical problems anyway and 'stress the antepenultimate mora' sounds like a good description of a conlang.

[bradrn]

My current proposal is assigning accent regularly like many stress systems but declaring by fiat that it will surface as pitch. Developing interesting rules has proven difficult, however, since the principles involved keep resulting in contradictions. I initially planned on something like the Latin stress rule but with final superheavy syllables taking accent. This makes sense if understood as accenting the antepenultimate mora, with some adjustments for syllable structure. Yet when applying optimality theory (which seems the industry standard in phonology research) this system falls apart. No matter how I rank the constraints involved, they keep resulting in some forms having counterintuitive accent.

Antepenultimate stress is something which a lot of theories have trouble with. I believe the usual (non-Optimality-Theoretic) analysis of Latin takes the last mora as being extrametrical, which neatly gives the attested stress pattern.

[Ahzoh]

Apparently Akkadian stress is also like Latin's: stress the third-to-last mora of a word, which results in the last syllable never being pronounced unless it's superheavy. My conlang, Vrkhazhian, also possesses this stress system.

[jal]

I did mention Kupwar already, but that’s in Asia. (Apparently the name for this is ‘metatypy’.)

Ah, yes, I think it is Kupwar that I was thinking about, thanks!


JAL

[AwfullyAmateur]

Not metatypy, but I've been having fun putting Sodemeresh loanwords into Estreri. Which is how the word sirtuke ('to sail') came into Estreri, even though in its original language sirtuke is infinitive and Estreri lacks infinitives. My explanation for this is that Soduar is coastal, Estrer is landlocked, and the two countries are right next to each other, so Estrer borrows a lot of oceanic/sailing-related words from there. Sorry if this is too rambling lol.

[masako]

just a Klingon script I'm working up

[bradrn]

just a Klingon script I'm working up

Reminds me of the Ithkuil script! (The name of which I’ve sadly forgotten.) How does this one work?

[WeepingElf]

I spot a Nineteen-Eighty-Four reference in the words listed.

[malloc]

Antepenultimate stress is something which a lot of theories have trouble with.

Yeah, there's a reason they call this phenomenon "stress" I suppose.

[Ahzoh]

I still can't figure out which clauses, the condition or the consequent, is supposed to be marked counterfactual and which is not. Or if they're both supposed to.

[condition] [consequence]
1.a) [irrealis] [irrealis]
1.b) [irrealis] [counterfactual]
2.a) [irrealis] [irrealis]
2.b) [counterfactual] [irrealis]

To me it makes sense to mark the condition clause as counterfactual, but I read this pdf on counterfactuals in nonfuture vs future languages and it appears the consequent clause is the clause marked counterfactual.

[masako]

Reminds me of the Ithkuil script! (The name of which I’ve sadly forgotten.) How does this one work?

I'm still working on it, but this is what I have so far.

I spot a Nineteen-Eighty-Four reference in the words listed.

Marc has included quite a few Easter eggs in the vocabulary, so you might be right.

[bradrn]

I still can't figure out which clauses, the condition or the consequent, is supposed to be marked counterfactual and which is not. Or if they're both supposed to.

[condition] [consequence]
1.a) [irrealis] [irrealis]
1.b) [irrealis] [counterfactual]
2.a) [irrealis] [irrealis]
2.b) [counterfactual] [irrealis]

To me it makes sense to mark the condition clause as counterfactual, but I read this pdf on counterfactuals in nonfuture vs future languages and it appears the consequent clause is the clause marked counterfactual.

One thing to keep in mind: not all categories named ‘counterfactual’ necessarily behave in the same way! It’s often a good idea to have a look at multiple sources, to get an idea of the cross-linguistic variation.

(Or, you know, just do it the way you want — it’s your conlang, after all…)

[Ahzoh]

I still can't figure out which clauses, the condition or the consequent, is supposed to be marked counterfactual and which is not. Or if they're both supposed to.

[condition] [consequence]
1.a) [irrealis] [irrealis]
1.b) [irrealis] [counterfactual]
2.a) [irrealis] [irrealis]
2.b) [counterfactual] [irrealis]

To me it makes sense to mark the condition clause as counterfactual, but I read this pdf on counterfactuals in nonfuture vs future languages and it appears the consequent clause is the clause marked counterfactual.

One thing to keep in mind: not all categories named ‘counterfactual’ necessarily behave in the same way! It’s often a good idea to have a look at multiple sources, to get an idea of the cross-linguistic variation.

(Or, you know, just do it the way you want — it’s your conlang, after all…)

I was certain a counterfactual is simply "an alternate past state that did not occur in reality". That "if I had money, we would be drinking beer", asserts that having money is contrary to reality. But the drinking of beer is also contrary to reality.

Though for this same reason counterfactual futures confuse the fuck out of me like "if she left tomorrow, she would arrive on time", which is somehow distinct from indicative conditional "if she leaves tomorrow, she will arrive on time"

I don't think this matter is all that subjective and willynilly.