Conlang Random Thread

[Ares Land]

I tried recording some Simbri yesterday (I was procrastinating, really. Describing how the subjunctive works turned out to be a bit of a chore)

Here's a sentence I used earlier:

Ya xallo an etliqer am wânnin an tlebaqo somâ, taream.
Anyway, by following the tracks, we found out it was the beautiful panther that ate grandma.
Listen here.


Oddly enough, the glottalized consonant were a bit difficult... but the real problem turned out to be stress. My brain desperately wants to revert to French mode and stress last syllables.

[zompist]

Whimsical sentences are really zompist's trick which a lot of us stole; but there are other options.

Of course. Note that I borrowed this style from a Latin textbook, Latin for People, as well as from the syntax tradition.

Grammars (and syntax papers) tend to be dull, and it's a lot to ask of a reader to read one that's not even for a real language. So I think some amusement along the way is warranted.

You don't have to be funny; but if you're not, I suggest using stories or proverbs, or something else from your culture. That will probably be more interesting than "John washed himself".

If you look at sample texts in scholarly descriptions of natlangs, you will soon recognize that they are usually taken from actual speech acts such as conversations between native speakers or books written in the language discussed.

This is undoubtedly good practice, and one I followed for my sketches of Old Chinese and Sanskrit. But I also had another motive: I didn't trust myself to come up with accurate sentences in those languages, and unlike the modern languages, I had no native speakers to help out.

For ancient languages, texts might be all we have. But texts are not likely to be at all typical of how people actually speak.

The other problem with found texts is that sentences tend to be far more complex than is needed for the grammatical point at hand. To use an example at random, I could use

Gallia est omnis divisa in partes tres, quarum unam incolunt Belgae, aliam Aquitani, tertiam qui ipsorum lingua Celtae, nostra Galli appellantur.

to demonstrate how to use the pronoun "qui". But it's kinda overkill. Whether to teach a language, or merely to demonstrate something in the grammar, you should ideally use the simplest sentence that will do.

Zompist's Xurnese example about a sculptor hoping to govern a province one day is one of my favourite conlang text examples, as it tells us something about an important facet of Xurnese culture: the country is ruled by an art academy.

Thanks. Few examples can be that striking, but I do try for sentences that could at least be said in that culture. One reason, pace Jeffrey Henning, that I don't really like translations of the Babel Text or the North Wind and the Sun.

[WeepingElf]

[...]

Zompist's Xurnese example about a sculptor hoping to govern a province one day is one of my favourite conlang text examples, as it tells us something about an important facet of Xurnese culture: the country is ruled by an art academy.

Thanks. Few examples can be that striking, but I do try for sentences that could at least be said in that culture. One reason, pace Jeffrey Henning, that I don't really like translations of the Babel Text or the North Wind and the Sun.

It is often said in favour of such translations that they allow to "compare" the languages; but what is the point of such comparisons if the languages are not related to each other in any way? Translating the same text into a bunch of languages in order to compare them makes sense if they are closely related to each other, and even then one may use a text relevant to that particular family. Where such standard texts make sense IMHO is where you want to compute indexes of synthesis, agglutination etc., where you want to have texts which are of equal conceptual complexity.

[bradrn]

I tried recording some Simbri yesterday (I was procrastinating, really. Describing how the subjunctive works turned out to be a bit of a chore)

Here's a sentence I used earlier:

Ya xallo an etliqer am wânnin an tlebaqo somâ, taream.
Anyway, by following the tracks, we found out it was the beautiful panther that ate grandma.
Listen here.


Oddly enough, the glottalized consonant were a bit difficult... but the real problem turned out to be stress. My brain desperately wants to revert to French mode and stress last syllables.

IPA please?

(Also, is there any particular reason for the abrupt pitch changes? Are there unmarked tones, or was it just due to non-fluency in the language?)

[Nortaneous]

Many, if not most conlang grammars fall short at this: the examples clearly show that they are constructed. Of course, with a conlang, there are no "actual speech acts", but a good conlanger ought to be able to come up with something which gives the impression of one. (I have to admit that my grammar of Old Albic indeed contains examples that are obviously constructed, but these are meant to be replaced by ones drawn from texts I shall write in the language.)

cf. the h13post about how conlangs ought to have any thought put into pragmatics and information structure

[Kuchigakatai]

The other problem with found texts is that sentences tend to be far more complex than is needed for the grammatical point at hand. To use an example at random, I could use

Gallia est omnis divisa in partes tres, quarum unam incolunt Belgae, aliam Aquitani, tertiam qui ipsorum lingua Celtae, nostra Galli appellantur.

to demonstrate how to use the pronoun "qui". But it's kinda overkill. Whether to teach a language, or merely to demonstrate something in the grammar, you should ideally use the simplest sentence that will do.

It might amuse you to hear that, because most of the Classical-era Roman authors were indeed a bit too fond of long, ornate sentences, full of adverbial phrases, relative clauses and parenthetical remarks, what Latin reference grammars and dictionaries tend to do is take out most of the stuff, reducing quotes to the essential bits. For example, in section 568, Allen & Greenough's grammar shows:

Note— The expressions facere ut, committere ut, with the subjunctive, often form a periphrasis for the simple verb.

Fēcī ut Flāminium ē senātū ēicerem. (Cat. M. 42)
It was with reluctance that I expelled Flaminius from the senate.

And I'd like to note that the "with reluctance" bit seems uncalled for, not found in the sentence. But then, if you check Cicero's Catō Maior dē Senectūte, section 42, you find the following (and here I bold the words kept by A&G):

Invitus feci ut fortissimi viri T. Flaminini fratrem L. Flamininum e senatu eicerem septem annis post quam consul fuisset, sed notandam putavi libidinem.

Which shows the origin of the phrase "with reluctance", the adjective invītus! (They forgot to keep it while taking most words out.) "It was with reluctance that I expelled the brother of that very strong man Titus Flamininus, called Lucius Flamininus, from the Senate, seven years after he had been a consul, but I thought [his] licentiousness needed to be noted."

[Ares Land]

I tried recording some Simbri yesterday (I was procrastinating, really. Describing how the subjunctive works turned out to be a bit of a chore)

Here's a sentence I used earlier:

Ya xallo an etliqer am wânnin an tlebaqo somâ, taream.
Anyway, by following the tracks, we found out it was the beautiful panther that ate grandma.
Listen here.


Oddly enough, the glottalized consonant were a bit difficult... but the real problem turned out to be stress. My brain desperately wants to revert to French mode and stress last syllables.

IPA please?

(Also, is there any particular reason for the abrupt pitch changes? Are there unmarked tones, or was it just due to non-fluency in the language?)

Oh, sure:
[ˈja ˈçaʎ.ʎo an ɛˈtɬi.ʔjɛr am ˈwɒɲ.ɲin an ˈtɬe.ɓa.qo ˈso.mɒ ˈta.re.am]

(At least it's what I'm aiming at.)

Stressed syllables are supposed to have higher pitch, but it's very possible I overdid it.
[ɓ] and [q] generally tend to lower pitch, besides.

[Qwynegold]

Ah, yeah, they're a lot of work... A bit of originality helps I think. Sample sentences get a lot harder to make, but you get reusable vocabulary and constructions.
They're also less boring to write and they help sustain the reader's attention.

Whimsical sentences are really zompist's trick which a lot of us stole; but there are other options. For Old Hieratic Tarandim I use religious language, using both the Bible and the Popol Vuh as models, and it works really well.

Oh, so you steal sentences from elsewhere?! Is that okay? I have some sentences that may have originated from Wikipedia, and I've been worrying it might be bad somehow.

If you look at sample texts in scholarly descriptions of natlangs, you will soon recognize that they are usually taken from actual speech acts such as conversations between native speakers or books written in the language discussed. This is because elicited examples are considered second-class evidence (they are likely to elicit an ideal model of the language that exists in the mind of the informant but is not really held up in actual speech acts) while constructed examples are not considered evidence at all (because they only show the describing linguist's own ideas of the language).

Many, if not most conlang grammars fall short at this: the examples clearly show that they are constructed. Of course, with a conlang, there are no "actual speech acts", but a good conlanger ought to be able to come up with something which gives the impression of one. (I have to admit that my grammar of Old Albic indeed contains examples that are obviously constructed, but these are meant to be replaced by ones drawn from texts I shall write in the language.)

That said, Zompist's Xurnese example about a sculptor hoping to govern a province one day is one of my favourite conlang text examples, as it tells us something about an important facet of Xurnese culture: the country is ruled by an art academy. Almost everywhere else, it would be a non sequitur: why should a sculptor expect to be appointed a province? But in Xurno, it makes sense!

Sometimes I've been using sentences that sound like they might come from a conversation, but they are really boring and of course they lack the context of the conversation. I feel like I haven't done enough conworlding to go really indepth into the culture like Zompist has done. Here are some of my examples, with varying degree of "natural speechness":

That tree is very old.
I almost burned the food!
Has anyone ever seen the edge of the world?
Don't add anything more to the stew.
My kitten walked away.
This shark has come far from the sea.
What has made these holes?
I will be doing no more dancing with these feet anymore.
This net is broken. Is there any other?
Scuípìo and company <i>(all male)</i> are going to the river.
My child's in-laws live near the coast.
The weather is changing.
I want us to be like that couple.
I'll give you one of these two cherries. Ah! I dropped them both.
These nostrils are so full of sand!
Chruisola gave me these, but I don't know how to use them.

[Qwynegold]

Stressed syllables are supposed to have higher pitch, but it's very possible I overdid it.

Yeah, this is very common of conlangs getting spoken by their creators. Of course you'll not have a natural flow, because you aren't used to speaking this. People who have been speaking their conlangs out loud for a long time seem to get good at it, though.

I've been thinking that I should make recordings of me speaking my various conlangs, and use Praat to check how the stress, pitch accent or tones turned out. And then manipulate the sound file so that these things are more like what they should be. Then I can listen to those "fixed" files and try to copy that accent when I continue practicing.

[Ares Land]

Oh, so you steal sentences from elsewhere?! Is that okay? I have some sentences that may have originated from Wikipedia, and I've been worrying it might be bad somehow.

Ah, no, I don't. The sentences are, in-universe, from Tarandim scripture. But writing scripture is difficult, you have to fast in a cave for days and eat grasshopers, so to get the proper otherworldly toned, I stole bits from actual holy books and reworked them to fit the culture.

I really like the Popol Vuh a lot, and it was great for that purpose. The metaphors and episodes are truly alien, and so is the style. You really get the idea that it was intended as scripture, without it being biblical in the slightest. I have several translations of it, one of it is really bad, and the really bad one was most helpful. I got a lot of mileage from taking awkward turns of phrases or errors, and trying to figure out: 'what if that made sense for some culture'?

For instance:
And in those days the giant llamas will judge those who ate them.
That one's really a summary of an episode in the Popol Vuh, where Dogs and turkeys turned on failed humans. 'In those days' is Biblical.
The 'giant llamas' I got from figuring out what kind of animals they would raise. (They' actually existed: https://en.wikipedia.org/wiki/Palaeolama)

Lord Many Insects sent out the Child Screamer.
Again, that's from the Popol Vuh. 'Lord Many Insects' is inspired by the sinister names of the Lords of Xibalba. The 'Child Screamer' is an owl (the name is mine though, from the way owls screech at night).

It's a lot of work and I can't really do that for all sample sentences, but as you can tell, I had a lot of fun working those out.

For Simbri, I used a few bits from the New Testament because the translation exists and is fairly important in-universe.

[bradrn]

I tried recording some Simbri yesterday (I was procrastinating, really. Describing how the subjunctive works turned out to be a bit of a chore)

Here's a sentence I used earlier:

Ya xallo an etliqer am wânnin an tlebaqo somâ, taream.
Anyway, by following the tracks, we found out it was the beautiful panther that ate grandma.
Listen here.


Oddly enough, the glottalized consonant were a bit difficult... but the real problem turned out to be stress. My brain desperately wants to revert to French mode and stress last syllables.

IPA please?

(Also, is there any particular reason for the abrupt pitch changes? Are there unmarked tones, or was it just due to non-fluency in the language?)

Oh, sure:
[ˈja ˈçaʎ.ʎo an ɛˈtɬi.ʔjɛr am ˈwɒɲ.ɲin an ˈtɬe.ɓa.qo ˈso.mɒ ˈta.re.am]

(At least it's what I'm aiming at.)

Stressed syllables are supposed to have higher pitch, but it's very possible I overdid it.
[ɓ] and [q] generally tend to lower pitch, besides.

OK, so I tried recording this myself, and after many (many, many…) tries, I managed to get something I’m happy with. Listen here (if the Dropbox link works, that is). Some comments:

• I tried saying this one at a normal conversational pace, as opposed to the slowed-down pace of your recording. To me, this makes it sound slightly more ‘natural’ than your recording, but it also makes it a bit harder to distinguish all the sounds.
• The glottalised sounds in particular are a bit indistinct. The glottal stop has been reduced to creaky voice, as is usual for word-medial /ʔ/, and the implosive is also pretty weak. Sadly neither comes through very well in the recording, though the creaky voice is vaguely visible on a spectrogram.
• I also had problems with /ɒ/, especially in /ˈso.mɒ/ — which is somewhat odd, since my native language has that contrast!
• Bizarrely, I also regularly mispronounced /r/ as a uvular rhotic. I can only assume this is due to influence from Hebrew, the only other language I have even a vague command of (and even that is a huge overexaggeration).
• I’m not sure if I got the pitch quite right: stressed syllables are high as per your description, but sadly most of the non-stressed syllables (including /ɓa.qo/) also seem to be high as well.

For instance:
And in those days the giant llamas will judge those who ate them.
That one's really a summary of an episode in the Popol Vuh, where Dogs and turkeys turned on failed humans. 'In those days' is Biblical.
The 'giant llamas' I got from figuring out what kind of animals they would raise. (They' actually existed: https://en.wikipedia.org/wiki/Palaeolama)

I do remember being puzzled by this sample sentence, actually, so thanks for explaining where it came from! (It’s not the sort of sentence one forgets easily.)

[Ares Land]

Oh, that's really cool! Do you mind if I eventually use that sound sample in the grammar? (With all proper credit, of course).

The stressed syllables are all right, really. The implosive and uvular tend to lower the pitch of the following vowel, at least in my own speech, but it's not really a big deal if it doesn't in yours.

The intrusion of uvular r is funny, because (in-universe) it's the historical pronunciation of Simbri r and I think it's still in free variation with [r] in many dialects.

[bradrn]

Oh, that's really cool! Do you mind if I eventually use that sound sample in the grammar? (With all proper credit, of course).

Sure, of course!

(Though I can’t help but wonder about what format your grammar is in if you can embed an audio file into it… I just use LaTeX for mine.)

The stressed syllables are all right, really. The implosive and uvular tend to lower the pitch of the following vowel, at least in my own speech, but it's not really a big deal if it doesn't in yours.

Actually, I find it interesting that the implosive doesn’t do anything to pitch in my speech… historically, voiced stops tend to lower the pitch of the following vowel, and implosives are about as voiced as you can get. (Perhaps it’s because I pronounced it very weakly, though.)

The intrusion of uvular r is funny, because (in-universe) it's the historical pronunciation of Simbri r and I think it's still in free variation with [r] in many dialects.

Huh, how about that! I wonder what the explanation for that is… telepathy, perhaps?

[Ares Land]

(Though I can’t help but wonder about what format your grammar is in if you can embed an audio file into it… I just use LaTeX for mine.)

Oh, HTML with a few bits of CSS/JS trickery.
(I use Mediawiki for convenience, but I'll export an HTML version and polish that off when I have something presentable enough.)

[Travis B.]

For instance:
And in those days the giant llamas will judge those who ate them.
That one's really a summary of an episode in the Popol Vuh, where Dogs and turkeys turned on failed humans. 'In those days' is Biblical.
The 'giant llamas' I got from figuring out what kind of animals they would raise. (They' actually existed: https://en.wikipedia.org/wiki/Palaeolama)

Would the Mayans actually have had any memories of said Palaeolamas?

[Ares Land]

Would the Mayans actually have had any memories of said Palaeolamas?

Ah, sorry, that wasn't clear. The Palaeolamas are part of my conworld. The Popol Vuh has dogs and turkeys.

[Ares Land]

A short sample of written Simbri in the native script:

Tletlandis kâdas tlammas sôrsa Piyon.
tle-tland-is kâdas tlammas s-ôrsa Piyon
ANIM-IPRF\sell-PST donkey lame 1.EXCL-uncle Piyon

Uncle Piyon sold him a lame donkey.

Webp.net-compress-image.jpg (26.01 KiB) Viewed 15150 times

[bradrn]

A short sample of written Simbri in the native script:

Tletlandis kâdas tlammas sôrsa Piyon.
tle-tland-is kâdas tlammas s-ôrsa Piyon
ANIM-IPRF\sell-PST donkey lame 1.EXCL-uncle Piyon

Uncle Piyon sold him a lame donkey.
Webp.net-compress-image.jpg

The structure of this script looks quite interesting… the various shapes seem to be positioned mostly over, under and around each other in syllabic blocks. I assume it’s some sort of abugida, but I’d love to have a more detailed explanation! (Or even an annotated version of that image.)

Beyond that, what really sticks out to me is the diversity of shapes: you have right angles, acute angles, gentle curves, diagonal lines and even some more complex shapes. All of which makes me wonder what medium this is meant to be written in — modern pens can easily support such diversity of shape, but most others cannot. (It reminds me most strongly of Brāhmī, but even that uses a far more restricted set of shapes.)

[Ares Land]

A short sample of written Simbri in the native script:

Tletlandis kâdas tlammas sôrsa Piyon.
tle-tland-is kâdas tlammas s-ôrsa Piyon
ANIM-IPRF\sell-PST donkey lame 1.EXCL-uncle Piyon

Uncle Piyon sold him a lame donkey.
Webp.net-compress-image.jpg

The structure of this script looks quite interesting… the various shapes seem to be positioned mostly over, under and around each other in syllabic blocks. I assume it’s some sort of abugida, but I’d love to have a more detailed explanation! (Or even an annotated version of that image.)

Beyond that, what really sticks out to me is the diversity of shapes: you have right angles, acute angles, gentle curves, diagonal lines and even some more complex shapes. All of which makes me wonder what medium this is meant to be written in — modern pens can easily support such diversity of shape, but most others cannot. (It reminds me most strongly of Brāhmī, but even that uses a far more restricted set of shapes.)

Each line is one word, or three syllables, whichever comes first.
Each syllable is a rectangular block, read from top/left to bottom/right, in that order: feature(s) - vowel - feature.
Consonants are written with one or two features. (Some consonants get their own symbol, other get a combination of two). Under the vowel, there's usually just one symbol/feature.

Writing proceeded from ochre on textile to pigment on hemp paper. This is a modern hand, meant to be written with a pencil on paper. (Pencils don't use graphite but a mixture of wax, fat, or clay with pigments - ochre or charcoal) or printed (lithography, no moveable type).

There are other hands more suited to carving on rock or writing with ink, but I'm not satisfied yet with these.

[Richard W]

Each syllable is a rectangular block, read from top/left to bottom/right, in that order: feature(s) - vowel - feature.

I don't understand the writing of tlammas 'lame'. Does it have a different 's' to the other words, and why is it written as a superscript?

Do all consonants have features? How is the null initial consonant written?