Omni-kan syntax

[KathTheDragon]

Relative clauses
I'm looking through ways one could form relative clauses. To check one idea, I would like to ask everyone to tell how they interpret the following sentences. Just rewrite the sentences in normal English please.

1) snake jump cat fight
2) cat jump hiss
3) I see snake cat fight
4) cat jump fight snake
5) I see cat fight snake

I agree with Creyeditor’s interpretation. To me, these seem more like SVCs than relative clauses. (And this is fine! SVCs are a perfectly natural thing to have in a language — and they’re readily interpretable by English-speakers, as Creyeditor showed.)

I also agree, but I also think this tells you far more about English speakers than it does about any of these as potentially being relative clauses.

[Creyeditor]

My native language is German. The problem with the same sequences of words in German is that the bare verb stem is usually interpreted as an imperative verb. For 3) and 5) I get the results I got with English.

[Travis B.]

Relative clauses
I'm looking through ways one could form relative clauses. To check one idea, I would like to ask everyone to tell how they interpret the following sentences. Just rewrite the sentences in normal English please.

1) snake jump cat fight
2) cat jump hiss
3) I see snake cat fight
4) cat jump fight snake
5) I see cat fight snake

I agree with Creyeditor’s interpretation. To me, these seem more like SVCs than relative clauses. (And this is fine! SVCs are a perfectly natural thing to have in a language — and they’re readily interpretable by English-speakers, as Creyeditor showed.)

I also agree with Creyeditor's assessment. These seem to me to be a mixture of SVC's and complement clauses. SVC's should not be a problem for native English speakers IMO; after all English has forms like "go V" and "come V" and has forms that are almost SVC-like aside from having separating "and"s.

[Travis B.]

My native language is German. The problem with the same sequences of words in German is that the bare verb stem is usually interpreted as an imperative verb. For 3) and 5) I get the results I got with English.

The thing is that I imagine the prototypical IAL as lacking verb inflection, with tense and aspect being marked with auxiliaries or particles.

[Ares Land]

Relative clauses
I'm looking through ways one could form relative clauses. To check one idea, I would like to ask everyone to tell how they interpret the following sentences. Just rewrite the sentences in normal English please.

1) snake jump cat fight
2) cat jump hiss
3) I see snake cat fight
4) cat jump fight snake
5) I see cat fight snake

The snake jumps and the cat fights (it)
The cat jumps and hisses
I see the snake and the cat fight
The cat jumps and fights the snake
I see the cat fighting the snake.

[Qwynegold]

I was gonna post my answers now, but now I'm on my phone and my notes are on my computer. So I'll do that tomorrow. In the meanwhile, what is SVC?

[Qwynegold]

Why not just go with ‘all nouns, always optional’? Yes, it’s less common and (slightly) less expressive, but it has the advantage that both people whose native language has number and people whose native language doesn’t will make grammatical sentences from the start.

Because us westerners will interpret everything unmarked as singular. So there will be a lot of misinterpretations, which are worse than grammar mistakes that the receiver might still understand correctly.

[Richard W]

and I recall there are languages (in India? I don't remember) where the singular noun form is used with all numbers,

I know. Finnish does this. So that's why I chose to make it optional when used together with numerals; the meaning will remain unambiguous.

Using the singular after numbers is also to be found in Norfolk, England; it's far from just a Turkish phenomenon. It also comes in attributive phrases in standard English, such as a five ton elephant.

[Travis B.]

I was gonna post my answers now, but now I'm on my phone and my notes are on my computer. So I'll do that tomorrow. In the meanwhile, what is SVC?

SVC's are serial verb constructions.

[Qwynegold]

Alright, now I've got my notes. I didn't include any tense or aspect stuff in the sentences I wrote earlier, because I haven't completed that stuff yet. My clear sentences have that, but you can just as well imagine everything being in simple present tense.

The snake that the cat is fighting jumped.
1) snake jump cat fight _
CE: A snake jumps and a cat fights.
AL: The snake jumps and the cat fights (it)

Although no one used a relative clause for this sentence, the intended meaning is not really that far off, in my opinion.


The cat that hisses jumped.
2) cat jump _ hiss
CE: A cat jumps and hisses.
AL: The cat jumps and hisses

Same here, though here the order of the verbs is reversed.


I saw the snake that the cat fought.
3) I see snake cat fight _
CE: I see a snake and a cat fight.
AL: I see the snake and the cat fight

If we take this to be present tense, the meaning is in practice the same, since if you see the snake that means you are also seeing the cat.


The cat that is fighting a snake jumped.
4) cat jump _ fight snake
CE: A cat jumps and fights a snake.
AL: The cat jumps and fights the snake

I saw the cat that fought a snake.
5) I see cat fight snake
CE: I see a cat fighting a snake.
AL: I see the cat fighting the snake.

The guessed meanings and the intended meanings for these last two sentences are also very similar.


What I tried here is making relative clauses that are not marked in any way. If I've understood my own old notes correctly, relative clauses should follow the noun they modify. And the SVO order was followed in both the relative clause and the matrix clause. The relativized noun appears only in the matrix clause and has been omitted from the relative clause. This spot has been marked with _. The underlined word is one that occurs in the right place of the matrix and relative clauses simultaneously.

If I put the omissive pronoun I talked about earlier in this thread, or just a regular old pronoun, where the _ is, what do you think? Would these still be serial verb constructions? Would they be easy to understand and produce then, or not? I'm thinking about the place where I worked about a year ago, where almost no one spoke good Swedish. Those guys would never use relative clauses, so I'm trying to think of how one could form relative clauses, or some substitution for them, that even they could use.

[bradrn]

I suspect the reason everyone was misinterpreting these was because the relative clauses are for the most part not adjacent to the head noun:

1) snake jump [cat fight _]
2) cat jump [_ hiss]
3) I see snake [cat fight _]
4) cat jump [_ fight snake]
5) I see cat [_ fight snake]

I suspect the most minimalistic way to fix this would be to use a correlative strategy, like so:

1) cat fight snake, that snake jump
2) cat hiss, that cat jump
3) cat fight snake, I see that snake
4) cat fight snake, that cat jump
5) cat fight snake, I see that cat

Though you can also explore other strategies like using resumptive pronouns or adding an explicit relativiser.

[Rounin Ryuuji]

Relative clauses
I'm looking through ways one could form relative clauses. To check one idea, I would like to ask everyone to tell how they interpret the following sentences. Just rewrite the sentences in normal English please.

1) snake jump cat fight
2) cat jump hiss
3) I see snake cat fight
4) cat jump fight snake
5) I see cat fight snake

1.) Snakes jump and cats fight.
2.) Cats jump and hiss.
3.) I see the snake-cat (whatever that is) fight.
4.) Cat, go jump and fight the snake!
5.) I see the cat fighting the snake.

I would probably give up the first one for lost after a bit, but then decide the third, after thinking about it, meant "I see a snake and cat fighting" rather than that some sort of Last Airbender-like snake cat generally tending to fight.

[Qwynegold]

I would probably give up the first one for lost after a bit, but then decide the third, after thinking about it, meant "I see a snake and cat fighting" rather than that some sort of Last Airbender-like snake cat generally tending to fight.

Lol. I definitely need some way of separating the relative clauses from the rest.

[Qwynegold]

I suspect the reason everyone was misinterpreting these was because the relative clauses are for the most part not adjacent to the head noun:

Aha. I didn't want to insert the relative clause in the middle of it all, like we do in English.

I suspect the most minimalistic way to fix this would be to use a correlative strategy, like so:

1) cat fight snake, that snake jump
2) cat hiss, that cat jump
3) cat fight snake, I see that snake
4) cat fight snake, that cat jump
5) cat fight snake, I see that cat

Though you can also explore other strategies like using resumptive pronouns or adding an explicit relativiser.

Ah, that's what I was thinking first, except instead of "that N" I would use a 3rd person pronoun. An explicit relativizer would be neat, but I'm afraid a lot of people would not bother to learn how to use it.

[Qwynegold]

Hmm, if I'm reading WALS correctly, it's called a correlative clause if the relative clause contains something to mark it as such, but a paratactic relative clause if it doesn't.

[bradrn]

I suspect the reason everyone was misinterpreting these was because the relative clauses are for the most part not adjacent to the head noun:

Aha. I didn't want to insert the relative clause in the middle of it all, like we do in English.

I don’t see why that should be a problem… it’s what pretty much every other language does.

I suspect the most minimalistic way to fix this would be to use a correlative strategy, like so:

1) cat fight snake, that snake jump
2) cat hiss, that cat jump
3) cat fight snake, I see that snake
4) cat fight snake, that cat jump
5) cat fight snake, I see that cat

Though you can also explore other strategies like using resumptive pronouns or adding an explicit relativiser.

Ah, that's what I was thinking first, except instead of "that N" I would use a 3rd person pronoun. An explicit relativizer would be neat, but I'm afraid a lot of people would not bother to learn how to use it.

Except that introduces a lot of ambiguity: what jumps in cat fight snake, it jump?

Hmm, if I'm reading WALS correctly, it's called a correlative clause if the relative clause contains something to mark it as such, but a paratactic relative clause if it doesn't.

True; that’s what I should have called it.

[Qwynegold]

Aha. I didn't want to insert the relative clause in the middle of it all, like we do in English.

I don’t see why that should be a problem… it’s what pretty much every other language does.

Because it gets word-salady.

Except that introduces a lot of ambiguity: what jumps in cat fight snake, it jump?

You're right. I have to think about this...

[Qwynegold]

The labial continuant
Omni-kan has a phoneme that can be realized as any of [v], [ʋ] or [w]. I had decided to spell it with <w>, though <v> is very tempting.

Disadvantages of <v>

• When Spanish speakers and East Asians see <v>, they will likely pronounce it wrong, even if they are told to use /w/

Disadvantages of <w>

• It takes twice as long to write than <v>, in handwriting
• It takes about twice as much horizontal space than <v>, depending on your handwriting style or font
• It just seems dumb to use <w> for a phoneme that can be realized as [v], when you're not using <v> for anything

More about word order
Omni-kan has the following orders:

• possessor GEN possessee*
• Noun - adjective
• Noun - demonstrative
• Noun - numeral
• Preposition - noun
• Noun - relative clause
• Head last in compound words

The ones marked with green are the most common order, both in SVO languages and all languages. Prepositions are not the most common adposition in all languages, but they are dominant in SVO langs. For order in compounds I have no data.

*I'm probably biased because I really want this order. Reasons for and against this order are below. So the question is, does the one merit outweigh the demerits? Can anyone think of any other points that are relevant?

Merits

• It's the most common order among all languages (685 vs. 468)

Demerits

• The opposite order is more common in SVO langs (249 vs. 106)
• The opposite order is way, way more common among my focus languages (languages that I base the vocabulary on)

[Rounin Ryuuji]

The labial continuant
Omni-kan has a phoneme that can be realized as any of [v], [ʋ] or [w]. I had decided to spell it with <w>, though <v> is very tempting.

Disadvantages of <v>

• When Spanish speakers and East Asians see <v>, they will likely pronounce it wrong, even if they are told to use /w/

Disadvantages of <w>

• It takes twice as long to write than <v>, in handwriting
• It takes about twice as much horizontal space than <v>, depending on your handwriting style or font
• It just seems dumb to use <w> for a phoneme that can be realized as [v], when you're not using <v> for anything

The disadvantages of writing w strike me as really non-issues, while Spanish, being the second most frequent native language on Earth, and the most common one using the Roman alphabet as its orthography, conditioning the pronunciation of v is quite worthy of consideration. Orthographically, w is also used in English (third most common native language, most widely-used International language) and Hanyu Pinyin (used to Romanise the most common language). This considered, I really think orthographic w the more sensible choice.

More about word order
Omni-kan has the following orders:

• possessor GEN possessee*
• Noun - adjective
• Noun - demonstrative
• Noun - numeral
• Preposition - noun
• Noun - relative clause
• Head last in compound words

The ones marked with green are the most common order, both in SVO languages and all languages. Prepositions are not the most common adposition in all languages, but they are dominant in SVO langs. For order in compounds I have no data.

*I'm probably biased because I really want this order. Reasons for and against this order are below. So the question is, does the one merit outweigh the demerits? Can anyone think of any other points that are relevant?

Merits

• It's the most common order among all languages (685 vs. 468)

Demerits

• The opposite order is more common in SVO langs (249 vs. 106)
• The opposite order is way, way more common among my focus languages (languages that I base the vocabulary on)

Is all of this adjusted for number of speakers?

Edit: Addendum —

By family, Indo-European (including both Spanish and English, both extremely widespread, and also French, the Western Classical Languages Greek and Latin, the Eastern Classical Languages Sanskrit and various varieties of Persian, and many of the contemporary languages of India) is most common, followed by Sinitic. I think an auxiliary language which is tailored for a broad majority of humanity would reference these families most strongly, since quite a number of languages have very small bases of speakers.

[KathTheDragon]

It just seems dumb to use <w> for a phoneme that can be realized as [v], when you're not using <v> for anything

Given that the phoneme can also be realised as [w] and you wouldn't be using <w> for anything, this "disadvantage" can be applied to <v> as well. And as Ryuuji said, the other two "disadvantages" are non-issues. To me, they look like you're grasping for anything to make <w> look less appealing because you'd rather have <v>.