Conlang Random Thread

[bradrn]

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

This is linguistics! Assume absolutely every terminological choice is subjective and willynilly, until proven otherwise. ‘Counterfactual’ is a sufficiently old term that I wouldn’t assume it refers to a single coherent thing.

[Ahzoh]

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

This is linguistics! Assume absolutely every terminological choice is subjective and willynilly, until proven otherwise. ‘Counterfactual’ is a sufficiently old term that I wouldn’t assume it refers to a single coherent thing.

I don't find this answer particularly satisfying. I am referring to a very specific definition/type of counterfactual, which I believe I have already defined.

It's like, sure, I can absolutely have a voiced uvular stop in my language, but there are clear reasons why they are rare: uvulars retract the tongue root, while voicing advances it, creating an inherent articulatory contradiction. And there's scientific reasons why certain combination of orders of noum, demonstrative, numeral, and adjective are unattested, such as iconicity principle, etc.

The abstract I read about (a download link to a pdf) suggests that using past tense (or nonfuture, in this case) morphology is pretty much a universal feature in languages with explicit tense marking.

This is tricky because I do not mark tense on the verb, or rather, tense is not the primary distinction. It is mood-prominent with a basic realis vs irrealis distinction, which is adjacent to a nonfuture vs future distinction.

I have a nonfuture/realis mood, and several irrealis moods, two of which are future-orientated and two of which are past-orientated. It is the counterfactuals that are past-orientated, conveying a sense of "lost past", a past that could have been but did not come about. I mark these with the -ya- prefix.

But because the distinction is primarily modal, that means there is: what the verb means when isolated and standing alone and what the verb means when combined in a main-subordinate clause conditional construction.

I think when the verb stands alone it has a meaning of being a counterfactual consequence, e.g. "I would have done my homework, but then my friends came over"
But then I must figure out if the counterfactual marks the counterfactual condition or the counterfactual consequence when there is a main-subordinate clause conditional construction.

So that is why I don't know whether the condition clause is supposed to be marked counterfactual or if it's the consequence clause, or both. I don't know the psycholinguostics or formal logic or semantic theories or whatever is involved.

[bradrn]

It's like, sure, I can absolutely have a voiced uvular stop in my language, but there are clear reasons why they are rare: uvulars retract the tongue root, while voicing advances it, creating an inherent articulatory contradiction.

It’s worth noting that phonetics is unusual, in that (a) it’s firmly linked to the underlying physics, and (b) the terminology is rigorously standardised (courtesy of the IPA). Most of linguistics is not like this.

And there's scientific reasons why certain combination of orders of noum, demonstrative, numeral, and adjective are unattested, such as iconicity principle, etc.

I have grave difficulties calling anything in linguistics ‘scientific’. Sure, people have suggested plenty of reasons for word order — but the reasons you get from Minimalists will be very different to the reasons you get from Cognitive Linguists, which in turn are very different from the reasons you get from Basic Linguistic Theorists. I sympathise with some of these reasons more than others, but I wouldn’t go so far as to call even my favourite theories ‘correct’ or ‘scientific’.

The abstract I read about (a download link to a pdf) suggests that using past tense (or nonfuture, in this case) morphology is pretty much a universal feature in languages with explicit tense marking.

Yes, it does. But I note that it tries to summarise the situation in just a couple of sentences. I’d go back to the references they cite and read those, to get a better idea of the full scope of cross-linguistic variation.

So that is why I don't know whether the condition clause is supposed to be marked counterfactual or if it's the consequence clause, or both. I don't know the psycholinguostics or formal logic or semantic theories or whatever is involved.

Yeah, don’t bother about the formal semantics. Just ignore it, is my advice. Trying to work that stuff out from first principles can make you go a bit crazy. The theories have their place, but I’ve never believed that they reveal any really fundamental facts about how language works.

(Only once have I tried to apply a formal theory to conlanging, when I used Optimality Theory to create a tone system. It really wasn’t worth it — took far too long, and gave unsatisfactory results. When I went back and redid that language, I changed it to use a more standard system of sequenced phonetic rules, which I’m far more satisfied with.)

[keenir]

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.

to my eyes, that looks like the same thing...the only difference is the tense of the had/left within the counterfactual.

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

This is linguistics! Assume absolutely every terminological choice is subjective and willynilly, until proven otherwise. ‘Counterfactual’ is a sufficiently old term that I wouldn’t assume it refers to a single coherent thing.

It's like, sure, I can absolutely have a voiced uvular stop in my language, but there are clear reasons why they are rare:

ah, but at the risk of me saying something stupid or obvious...things that are rare, are things that do exist, even if they are rare.

I mean, triconsonantal roots aren't exactly common, and yet this language family was born from your successes when you and I tried learning how to make triconsonantal roots; I'm glad you had so much good fortune with this over all this time, and I wish you much more.

whether or not its with things that are rare.

[Ahzoh]

to my eyes, that looks like the same thing...the only difference is the tense of the had/left within the counterfactual.

They are apparently completely different things, and the counterfactual is expressed with fake past tense.

ah, but at the risk of me saying something stupid or obvious...things that are rare, are things that do exist, even if they are rare.

That would be missing the point since rarity, or even attestation in general, is not the issue. The issue is logic. Is it more logical to do it this way? Or that way? Why or why not? Does it make logical sense?

I mean, triconsonantal roots aren't exactly common, and yet this language family was born from your successes when you and I tried learning how to make triconsonantal roots; I'm glad you had so much good fortune with this over all this time, and I wish you much more.

My success has involved a lot of demystification of the nature of such a concept.

[rotting bones]

Flavor conlang idea for my novel: Standard Intramural

(Let me know if this conlang idea already exists. I know it's terrible, but I think it's cool for an exotic SF setting. I have many pages of notes on this "language" that I didn't consult for this post. Last year, I had hopes of working this out is more detail, but I have been working pretty much nonstop since then.)

Intramural is natively spoken by the Partials of Betelgeuse.

The inhabited worlds of Betelgeuse are largely populated by non-sapient beasts called the Wholes. These creatures look brilliantly colored under infrared light, but look like black outlines on earth.

The Partials are a fungoid lifeform that infects the wholes. When a Whole is infected by a Partial, it gains sapience. Traditionally, Partials used this sapience to maximize their reproductive success. As the masters of their star empire, the Partials haven't had to stress about reproduction in millennia.

The early Partials communicated to each other by sending electrical signals. The atmosphere of their habitat conducts electricity under certain atmospheric conditions. Partials often moved from Whole to Whole in a pattern that enabled them to communicate clearly. They used this language to feed their hosts and congregate for support.

Since early times, the electrical language that Partials involuntarily emit is primarily meant to communicate information about weather and food. With time, it gathered prefixes and suffixes to communicate richer information content.

Once Partials had fully domesticated the Wholes and organized into communities, electrical emission became a source of shame to them. In polite society, one was expected to demonstrate one's control over the host by speaking a vocal language. Nevertheless, the vocal languages that emerged continued to be centered around weather and the speaker's mood.

Grammar idea

A sentence consists of a number of "root" words indicating weather and the speaker's mood. Sentences typically have 3 roots. Diverging from this pattern confers special significance. A number of prefixes and suffixes are added to these "roots" that communicate what humans think of as the relevant information in the sentence.

What is even more alien is that each substantive is often represented by affixes spread across multiple "roots". The word order in Intramuros is not the linear order of vocalization but rather which affixes are closer or further away from each "root".

The language has case "endings". The endings are attached to the outer end of the substantive affixes; that is, farther away from the root. A common use of the oblique case is as a dative.

Beyond the basics, the grammar is difficult for humans to follow. Refined speech requires harmony in choosing affixes across clauses, sentences and occasionally paragraphs. The easiest of these rules is that a sequence of syllables with more than one interpreation are perceived as more refined.

Example:

M-zai-ko y-p-ku mal-tok-de.
(Cold, pressured, formal) Welcome to the Intramuros.

Phonetic notes

y: schwa
x: velar fricative
r: syllabic consonant

Vocabulary

Roots
<na>: hot
<ku>: pressure
<zai>: cold
<lo>: dilation
<gav>: positive charge
<mesone>: negative charge
<rois>: moisture
<ak>: dry
<xas>: food
<pon>: bones
<ruk>: emergency
<sr>: fun
<tok>: formal

Particles
<pa>: subordination

Case Endings
y-: oblique case

Substantives
m-, --, -de: welcome
-ko, p-, mal-: The Intramuros

[keenir]

to my eyes, that looks like the same thing...the only difference is the tense of the had/left within the counterfactual.

They are apparently completely different things, and the counterfactual is expressed with fake past tense.

fake past tense? I'm suspecting that the two example statements were splitting an extremely fine hair, and "fake past tense" makes me think i was right.

ah, but at the risk of me saying something stupid or obvious...things that are rare, are things that do exist, even if they are rare.

That would be missing the point since rarity, or even attestation in general, is not the issue. The issue is logic. Is it more logical to do it this way? Or that way? Why or why not? Does it make logical sense?

I think you may be venturing far enough into the weeds tall grass that you're about to find Pokemon.

You are performing what JRRT himself called "subcreation"...in these matters, logic bends to your will, for you are its creator.

*thinks*
Okay, I think I have a better answer for you: is this particular conlang emerging from a top-down conference that has made modifications to the speakers' language? (think Ataturk with Turkish, the various attempts in and outside of China to get Mandarin to fit into a typewriter, etc)...if the answer is Yes, then logic can apply: the logic and reasoning of that conferencemembers.

if the answer is No, then logic need not apply.

I mean, triconsonantal roots aren't exactly common, and yet this language family was born from your successes when you and I tried learning how to make triconsonantal roots; I'm glad you had so much good fortune with this over all this time, and I wish you much more.

My success has involved a lot of demystification of the nature of such a concept.

demystification? do the D'ni know you're doing that?

sorry; seriously, I wasn't aware the triconsonantal roots were mystical to anyone outside of kabalah(sp) and "The Bible Code".

[Ahzoh]

sorry; seriously, I wasn't aware the triconsonantal roots were mystical to anyone outside of kabalah(sp) and "The Bible Code".

You look on quora and non-linguistic subreddits and you'll find many people who a) think speakers of Semitic languages don't use vowels at all a la Salishan or b) think words are actually composed of roots and templates and that the vowels between the consonants are completely without rhyme or reason, instead of the actual reality that there is some kind of original stem that undergoes various transformations.

[bradrn]

A sentence consists of a number of "root" words indicating weather and the speaker's mood. Sentences typically have 3 roots. Diverging from this pattern confers special significance. A number of prefixes and suffixes are added to these "roots" that communicate what humans think of as the relevant information in the sentence.

Intriguing… if done right, this could be one of the best alien conlangs I’ve seen.

What is even more alien is that each substantive is often represented by affixes spread across multiple "roots". The word order in Intramuros is not the linear order of vocalization but rather which affixes are closer or further away from each "root".

The language has case "endings". The endings are attached to the outer end of the substantive affixes; that is, farther away from the root. A common use of the oblique case is as a dative.

I’ll note, however, that this is not particularly alien. From your description, it sounds very similar to some polysynthetic languages like Greenlandic and Nuu-chah-nulth.

[Ares Land]

Flavor conlang idea for my novel: Standard Intramural

(Let me know if this conlang idea already exists. I know it's terrible, but I think it's cool for an exotic SF setting. I have many pages of notes on this "language" that I didn't consult for this post. Last year, I had hopes of working this out is more detail, but I have been working pretty much nonstop since then.)

No, no, I don't think it's terrible -- and I don't think I encountered the idea before. I'd be glad to see this developped further when you get enough time, but this is really interesting.

[jal]

So that is why I don't know whether the condition clause is supposed to be marked counterfactual or if it's the consequence clause, or both. I don't know the psycholinguostics or formal logic or semantic theories or whatever is involved.

"Condition clause" and "consequence clause" are properly called "protasis" and "apodosis", or "antecedent" and "consequent" - I prefer the former terminology. What to mark the verb in what clause is up to you I'd say. I think the world's languages have all the varieties you could want: marking none, marking one or the other, or marking both (either the same or differently). It might also depend on how probable the speaker thinks the protasis is, if it's 100% certain, you have a "when" clause, and the apodosis is also no longer deemed counterfactual or uncertain.

As for your question, if in "I would have done my homework, but then my friends came over", "to do" is marked with the counterfactual, I would pose that in "If my friends hadn't come over, I would have done my homework", you would use the same counterfactual in the aposdosis no? Semantically these are almost the same. You could mark the If-clause with the counterfactual as well, but I don't think that's common cross-linguistically, especially since it's not really counterfactual when the condition is not in the past:

"If my friends don't come over, I will do my homework"
"If my friends hadn't come over, I would've done my homework"

These sentences are sementically very close, the first being set in the future, the second in the past. And though the condition itself is still hypothetical in the first, and counterfactual (apparently) in the second, relative to the apodosis they are equal.


JAL

[Ahzoh]

"Condition clause" and "consequence clause" are properly called "protasis" and "apodosis", or "antecedent" and "consequent" - I prefer the former terminology.

I don't like the former since they don't help me remember which clause means what. The fromer is basically the same as my terminology except the clause conveying the conditions is called "antecedent"

As for your question, if in "I would have done my homework, but then my friends came over", "to do" is marked with the counterfactual, I would pose that in "If my friends hadn't come over, I would have done my homework", you would use the same counterfactual in the aposdosis no?

Well yes, but I could just as much mark it in the irrealis mood since the irrealis deals with unrealized or nonactual events. The clause would make just as much sense.
So it can be "if [counterfactual mood] then [counterfactual mood]" or "if [counterfactual mood] then [irrealis mood]"

You could mark the If-clause with the counterfactual as well, but I don't think that's common cross-linguistically

A language that marks mood obligatorily on the verb but not tense or aspect is already uncommon.

...especially since it's not really counterfactual when the condition is not in the past:

"If my friends don't come over, I will do my homework"
"If my friends hadn't come over, I would've done my homework"

But I wouldn't consider the first sentence to be a counterfactual at all, but a hypothetical, and thus I would properly mark both clauses in the irrealis mood. The counterfactual in Vrkhazhian explicitly conveys "the past has already happened, we're talking about an alternate past". It is essentially a modal past tense.

If used with future adverbs or time reference the counterfactual would probably take on a desirative/optative meaning, such as "I wanted to go grocery shopping tomorrow, but my car broke down"

[jal]

Well yes, but I could just as much mark it in the irrealis mood since the irrealis deals with unrealized or nonactual events.

So what for you is the difference between "counterfactual" and "unrealized/nonactual"? Seems to me that semantically these are very close, if not the same. It seems you insist on a mood that's also a tense, which is fine, but then call that tense part a tense, and you have a "past irrealis", which is a counterfactual, and a "non-past irrealis", which covers all other cases.


JAL

[Ahzoh]

Well yes, but I could just as much mark it in the irrealis mood since the irrealis deals with unrealized or nonactual events.

So what for you is the difference between "counterfactual" and "unrealized/nonactual"? Seems to me that semantically these are very close, if not the same. It seems you insist on a mood that's also a tense, which is fine, but then call that tense part a tense, and you have a "past irrealis", which is a counterfactual, and a "non-past irrealis", which covers all other cases.


JAL

The irrealis is typically future-orientated while the counterfactual is nonfuture-orientated. So the distinction in the protasis makes sense. Yet, they (at least the irrealis) don't have to refer to time at all. Or, rather, the irrealis is far less tied to temporality than the counterfactual.
And when they each stand alone they seem to have an apodotic meaning. That is, at least, how counterfactuals behave in languages that have them as explicit moods (who also have a distinct general irrealis mood)

So then it's a matter if the protasis and apodosis must agree in mood markers or not. And if they need not, then I reason that one of the clauses should govern the temporality or primary modality.

[jal]

The irrealis is typically future-orientated while the counterfactual is nonfuture-orientated. So the distinction in the protasis makes sense. Yet, they (at least the irrealis) don't have to refer to time at all. Or, rather, the irrealis is far less tied to temporality than the counterfactual.

So in that case I'd say always use the irrealis in the protasis, as it is/was "undecided" at the time of the apodosis.

And when they each stand alone they seem to have an apodotic meaning. That is, at least, how counterfactuals behave in languages that have them as explicit moods (who also have distinct irrealis and realis moods)

That makes sense, as the apodosis is the main clause, the protasis a subclause.

So then it's a matter if the protasis and apodosis must agree in mood markers or not. And if they need not, then I reason that one of the clauses should govern the temporality or primary modality.

The apodosis, I'd say, as that's the main clause. Also, the protasis and apodosis can govern different temporality:

"If he had come, she wouldn't be still here tomorrow"

So it seems as if your irrealis can be used with seperate temporal information if used in the protasis, and the choice for the apodosis is either counterfactual (which is +past) or irrealis (especially for non-past references, though could be used for past references where the outcome is unsure)?


JAL

[Ahzoh]

Yea, it logically follows that the main clause should be the main pivot point.

So it seems as if your irrealis can be used with seperate temporal information if used in the protasis, and the choice for the apodosis is either counterfactual (which is +past) or irrealis (especially for non-past references, though could be used for past references where the outcome is unsure)?

I must emphasize that if there is to be any temporal references, it is between future and non-future, not past and non-past.

The realis can be marked in the apodosis, for matters involving implication (there are hoofprints here, so there must be a deer) or logical consequence (there is thunder if there is lightning).

[jal]

I must emphasize that if there is to be any temporal references, it is between future and non-future, not past and non-past.

Sure, but your counterfactual seems decidedly +past.

The realis can be marked in the apodosis, for matters involving implication (there are hoofprints here, so there must be a deer) or logical consequence (there is thunder if there is lightning).

Or certainty, I assume ("When I get home, I'll call you")?

Anyway, I think it's just a matter of coming up with all the constructions you need marking, and figuring out what would be the most logical marking, or whether you'll just have the speakers decide based on semantic and/or pragmatic considerations.


JAL

[Ahzoh]

As I work on Vrkhazhian more and more I decide to make it an almost rigidly head-initial language. And more to that, I think I will make it a verb-initial language, with the behaviour that animate nouns are placed before inanimate nouns, even if the animate noun is the syntactic object.

Thus:
eat\REAL-3sg Sam-NOM orange-ABS (VSO)
fall_on\REAL-3sg mouse-ACC pillar-ERG (VOS)

Almost like a direct-indirect but without explicit marking on the verb.

[jal]

almost rigidly head-initial ... animate nouns are placed before inanimate nouns, even if the animate noun is the syntactic object.

That kinda contradict each other :D.

eat\REAL-3sg Sam-NOM orange-ABS (VSO)
fall_on\REAL-3sg mouse-ACC pillar-ERG (VOS)

I have some difficulty with an animate-inanimate triggered pattern to have nom/abs and acc/erg. In fact, I don't think there's any language that has all four like this?


JAL

[bradrn]

almost rigidly head-initial ... animate nouns are placed before inanimate nouns, even if the animate noun is the syntactic object.

That kinda contradict each other .

No, it makes perfect sense to me. The verb is the head.

eat\REAL-3sg Sam-NOM orange-ABS (VSO)
fall_on\REAL-3sg mouse-ACC pillar-ERG (VOS)

I have some difficulty with an animate-inanimate triggered pattern to have nom/abs and acc/erg. In fact, I don't think there's any language that has all four like this?

I’m not aware of any with this specific combination, but Sanzhi Dargwa combines ergative–absolutive case-marking with hierarchical verbal agreement. So I don’t find the pattern here to be implausible.