Svagnaric (Phoenix) Language

[Pedant]

Brief idea, might work on it more a bit later on, but I've wondered for a while now how a bird language might work. Here are some preliminary notes:

• Speakers: The phoenixes (Classical Salvian svagnar, svagnō-; caladrius ardens, temporary species designation 4-7-7-6-2-1-1) are the oldest of the four physical sapient species on Ajjamah, with a basic civilization dating all the way back to twenty million years ago. A curious ornithan about the size of a large parrot, with wings and claws in the physical world and a third pair of limbs in the Dream, phoenixes have raven-like beaks and head structures but brightly-coloured plumage and a crest on their heads. Tool usage being eminently easy for their zygodactylic claws, the majority of their kind nevertheless remain at a premodern level, living in the jungles and windy places of the world. For some special few, there are the great star-nests high in the sky, reached by passing through the Six Levels of Heaven, and there they practice their magic and live free of want (and do some ridiculously complicated mathematical calculations). There is no point in pretending that the language of the birds above, high in the heavens, is the same as ever it was--but then they have magic to ease their communication, spell packages passed from one bird to another to stimulate thought processes and symbol receptivity. Down near the ground, while they are as quick-thinking as any creature can be, they lack what is understood by the heavenly phoenixes as high speech: concepts beyond the basics of life, such as predators and food and mating and flight, or even pronouns (though they do have names). They also have little variation in their language, as the species has remained global (although concentrated in Pelia) and interbreeding between "races" is common thanks to the magical birds being unable to reproduce with other magical birds of their kind.
• Phonology: The phoenix mouth is designed, not for croaking, but for warbling. Three rough tones have been recognized, high, mid, and low. The average "phoneme" is three beats long, and can be divided up in various ways: a long sustained note (marked as 3), three notes in succession (1:1:1), one note longer than the other (2:1 and 1:2), or with gaps in-between the note segments (1:1:0 or 1:0:1). Each note takes one of the three tones. The "phoneme" as a whole can be repeated as many times as is necessary, although a complete concept usually is only five or six "phonemes" long.
• Morphology and Syntax: The language is isolating to the extreme, as is necessary given the nature of their phonology. However, there is a set order to the modulations which cannot be deviated from. As an example, let us take a five-sequence system A-B-C-D-E. In theory, a valid phoenix sentence would be A-A-A, or A-A-B-C, or even A-A-C-C-C-E. A valid sentence would not be E-A-C, because that involves messing up the constituent order. There are in fact eleven such words in the phoenix language; remarkably, the syntax has remained relatively constant over what seems to have been hundreds of thousands of years.

Thoughts?

[Pabappa]

It looks interesting and well thought-out, but what do you mean by modulation? What do those letters stand for and how is A different from B through E?

[Pedant]

It looks interesting and well thought-out, but what do you mean by modulation? What do those letters stand for and how is A different from B through E?

Many thanks! In short:

1. By modulation I mean just that, a modulation in pitch in a specific manner. Seemed a more useful term than "phoneme" under the circumstances…
2. Ah, that was more just a demonstration. The letters represent specific modulations--for example A might be [high-low-mid], B might be [looooooow], C might be [high-high-high], etc. I have ten such variations decided upon, and seven with set meanings. (Two, by the way, are reserved for danger: a long mid-tone sound for danger in the air, and two piping high notes for danger on the ground or on a branch.)
   For more on the theory, do absolutely go and look up black-capped chickadees. The studies done on the syntax of their calls have been absolutely incredible, and were the main inspiration for this language.

Now, if only I knew how to link sound bites…

[Vardelm]

This is cool. REALLY cool.

I love the concept for the species and the variations. I already want to know more in terms of what cultures there might be, history, etc. etc. Lots of good details, like doing " ridiculously complicated mathematical calculations". Why? What are they calculating? I must know!

I'd like to see more info about how phonemes (modulations) form words (concepts), and how those form sentences. I don't quite see how it will work to have the ever "increasing" modulations through a sentence. I see see it as a phonotactic constraint for a word, though, which seems super cool. Definitely interested to see how many & what type of linguistic structures this can support.

[sasasha]

Really, really interesting - I absolutely love this idea.

I do wonder though, with my music hat on, if you are keeping the number of tones at a fairly safe level. If their physiology is deigned for warbling, I bet they can make pitches over a really wide range, and their ability to distinguish pitch will also probably be excellent - so I feel you might easily be able to fit in at least another couple of really safely distinct tones, which could quite vastly multiply your phonemes.

You know how a clarinet has different 'registers', which sound substantially different in tone as well as pitch? The same is true of the human voice (the reason that falsetto sounds so different to the modal voice at the same pitch - falsetto notes are actually harmonics of modal notes, and so are sort of like the 'overblown' notes on a wind instrument). I am no expert but I assume the syrinx of birds works the same way. It isn't just a pitch difference that you can exploit therefore - you could have (at least) a fundamental(=lower) register with several pitch distinctions, and a harmonic(=higher) register also with several pitch distinctions, and still be confident that your 'speakers' would always distinguish all the tones.

I mean, you could go WAY further than that - but I imagine you want to keep some limits on it. Just, only three pitch levels for a bird language seems maybe a bit safe, and a bit 'human'.

Apologies if you know all that already - and it's just a small question mark - I have nothing but admiration for all the rest! I particularly love the organising principle of each phoneme being based on three beats. Just trying it in my head it's very, very birdsong-like.

[zompist]

You've probably already noticed this, since you use "phoenix" as an adjective, but "Phoenician" refers to a human language and thus causes a little confusion!

(Both names of course go back to φοῖνιξ, which originally referred to the people, to the dye they sold, and to the date palm, and later was applied to the mythical bird.)

[mèþru]

Yeah that was confusing me.

[Pedant]

Okay, been a while, answering all these in order.

First of all, I am really very touched by all of your responses. I'll do my best not to make you proud.

Lots of good details, like doing " ridiculously complicated mathematical calculations". Why? What are they calculating? I must know!

Heh, heh…that would be telling.

I'd like to see more info about how phonemes (modulations) form words (concepts), and how those form sentences. I don't quite see how it will work to have the ever "increasing" modulations through a sentence. I see see it as a phonotactic constraint for a word, though, which seems super cool. Definitely interested to see how many & what type of linguistic structures this can support.

That I'm still working on, and somehow I reckon I'll have to make a change or two to the initial structure. Definitely to include more modulations…

I do wonder though, with my music hat on, if you are keeping the number of tones at a fairly safe level. If their physiology is deigned for warbling, I bet they can make pitches over a really wide range, and their ability to distinguish pitch will also probably be excellent - so I feel you might easily be able to fit in at least another couple of really safely distinct tones, which could quite vastly multiply your phonemes.
You know how a clarinet has different 'registers', which sound substantially different in tone as well as pitch? The same is true of the human voice (the reason that falsetto sounds so different to the modal voice at the same pitch - falsetto notes are actually harmonics of modal notes, and so are sort of like the 'overblown' notes on a wind instrument). I am no expert but I assume the syrinx of birds works the same way. It isn't just a pitch difference that you can exploit therefore - you could have (at least) a fundamental(=lower) register with several pitch distinctions, and a harmonic(=higher) register also with several pitch distinctions, and still be confident that your 'speakers' would always distinguish all the tones.
I mean, you could go WAY further than that - but I imagine you want to keep some limits on it. Just, only three pitch levels for a bird language seems maybe a bit safe, and a bit 'human'.
Apologies if you know all that already - and it's just a small question mark - I have nothing but admiration for all the rest! I particularly love the organising principle of each phoneme being based on three beats. Just trying it in my head it's very, very birdsong-like.

Oh, tempting, tempting…you've caught me, I actually worked out the initial elements of the language by whistling them myself, and seeing what I would recognize regardless of when I played it back to myself. Human brain making things more complicated (or too simple)…
So, yeah. Extending the range seems like the best bet. Maybe if I mark it down with some sort of key reference? Definitely going to see about including the syrinx element, even if that means I'll be unable to speak it properly (ah, well).

You've probably already noticed this, since you use "phoenix" as an adjective, but "Phoenician" refers to a human language and thus causes a little confusion!

(Both names of course go back to φοῖνιξ, which originally referred to the people, to the dye they sold, and to the date palm, and later was applied to the mythical bird.)

That I do know, Zompist, I just was very hard-pressed to find another adjective that worked. "Zarptacian" might have done, I suppose, using жар-пти́ца as a base, but that sounded a tad ugly to my ears. "Fenghuangian" might do, I guess. "Svagnaric" perhaps, from Classical Salvian svagnar (1m) "phoenix, firebird". I'm not sure, any suggestions?
Also, slightly unrelated note, might I ask for your assistance on something? Do you still have the formatting rules you created for the Iliu language (or one of them)? I know there was some basic detail in The Language Construction Kit, and I was wondering if that was all self-created or if there might be some guideline I could follow myself…

[zompist]

"Zarptacian" might have done, I suppose, using жар-пти́ца as a base, but that sounded a tad ugly to my ears. "Fenghuangian" might do, I guess. "Svagnaric" perhaps, from Classical Salvian svagnar (1m) "phoenix, firebird". I'm not sure, any suggestions?

I think Svagnaric is pretty good! And neighbors naming the language is a naturalistic touch.

Also, slightly unrelated note, might I ask for your assistance on something? Do you still have the formatting rules you created for the Iliu language (or one of them)? I know there was some basic detail in The Language Construction Kit, and I was wondering if that was all self-created or if there might be some guideline I could follow myself…

There's really not much behind it, except what's noted in the book. For what it's worth, I started with tones, and at least thought about what could be expressed underwater-- i.e, things like beat and intensity, but very little in the way of phonetic modulation. I wanted to avoid coming up with an alien speech that can be expressed easily in the Roman alphabet. (Clever as Adams's Lapine is, it's pretty unbelievable that rabbits could pronounce any of its words.)

[Vardelm]

Heh, heh…that would be telling.

Tease.

[sasasha]

Oh, tempting, tempting…you've caught me, I actually worked out the initial elements of the language by whistling them myself, and seeing what I would recognize regardless of when I played it back to myself. Human brain making things more complicated (or too simple)…
So, yeah. Extending the range seems like the best bet. Maybe if I mark it down with some sort of key reference? Definitely going to see about including the syrinx element, even if that means I'll be unable to speak it properly (ah, well).

Good thoughts! I'm looking forward to seeing how this develops.

I can see the tension between wanting realism and being-able-to-speak-it-myself. If it were me on this project I'd probably find it too frustrating to make it completely inaccessible for myself to use. Good for you if you are happy to transgress this, but if not...

Maybe it would help to imagine that the phoneme inventory you posted is like the Rotokas / Pirahã of bird languages. It's not that it's implausible, it just has a small phonemic inventory. Keep it if you are enjoying it -- just, if you make another Svagnaric language ( ), consider that it is likely to have more phonemes than this one.

If you want to extend the range a bit, but don't want to get too bogged down in what birds actually can and might do (something which varies massively between species, of course), and also want to keep your orthography looking like writing rather than music, I'd maybe go for something like either:

(a) 5 distinctions: AHMLB (taking A from 'altus' and B from 'bassus'). Probably still easy for you to distinguish these five with whistling: make your A roughly your highest note and B roughly your lowest and they will both be imbued with a sort of 'extreme' quality which will distinguish them from H and L.

(b) 6 distinctions: hmlHML (the system I described in my last post, where you are exploiting two registers, so there's a difference in tone as well as pitch between the two sets. This one is harder to whistle for a human -- there won't be the register change between the two sets with whistling, so it will just be 'a set of three in the lower range' and 'a set of three in the higher range', which might get confusing. You should be able to vocalise it, with some practice, but it may sound a bit like yodelling and not be the easiest thing to master.

Other thoughts that have occurred to me while thinking this over: croaks and clacks and the like are possible elements (I am sure you have considered this already); various bird-like composite tones could be models for other phonemes, i.e. a distinction between a high tone and a high tone that falls slightly at the end etc.; such features could be important suprasegmental phonetic features affecting prosody etc.; and finally, you could enter the territory you allude to when talking about "key reference" and think about whether your speakers use pitch relationships phonemically (i.e. a jump upwards of a 5th means something different to a jump upwards of a minor second, or even that all your speakers have perfect pitch and can pluck, say, 440Hz out of the air and it means something totally different to, say, 460Hz) but I think it would be very tricky ground to cover, and not necessarily how birds 'do' pitch.

[sasasha]

I did a little demo of something like (b) to test the practicality of what I was suggesting.

https://clyp.it/kyqx0llq?token=d5871e33 ... 889cafe807

It goes something like this:

mhml MHML mHM lmhLhlmHML

I found I wanted to set the pitch relationships, to aid distinguishing them, contrary to my previous skepticism. (I went for m as the tonic, h a 5th above tonic, l a minor 3rd below, and the pitch difference between hml and HML being an octave. The higher notes all use falsetto and the lower ones all modal voice).

This led me to think, if you did want to investigate that route, you could do some pretty interesting things. The selection of pitches was arbitrary. I could have selected very different sounding sets. This would be similar to selecting a mode. It made me think of Classical Indian raags: you probably know, different raags (sets of notes used in improvisation) are appropriate for different times of day and moods. Your speakers could select a different 'mode' depending on their mood, or other factors.

This even works just with the 3 distinctions of HML. And is something you could happily work on the core of the language without even considering, and then use as a feature affecting pragmatics.

I also found that (b) is pretty tricky to emulate in the human voice. Birds are good at swiftly changing between registers, and I (even as a professional singer) am not. On that level I wouldn't blame you for just sticking to your original system.

[bradrn]

I did a little demo of something like (b) to test the practicality of what I was suggesting.

https://clyp.it/kyqx0llq?token=d5871e33 ... 889cafe807

It goes something like this:

mhml MHML mHM lmhLhlmHML

I found I wanted to set the pitch relationships, to aid distinguishing them, contrary to my previous skepticism. (I went for m as the tonic, h a 5th above tonic, l a minor 3rd below, and the pitch difference between hml and HML being an octave. The higher notes all use falsetto and the lower ones all modal voice).

This led me to think, if you did want to investigate that route, you could do some pretty interesting things. The selection of pitches was arbitrary. I could have selected very different sounding sets. This would be similar to selecting a mode. It made me think of Classical Indian raags: you probably know, different raags (sets of notes used in improvisation) are appropriate for different times of day and moods. Your speakers could select a different 'mode' depending on their mood, or other factors.

This even works just with the 3 distinctions of HML. And is something you could happily work on the core of the language without even considering, and then use as a feature affecting pragmatics.

I also found that (b) is pretty tricky to emulate in the human voice. Birds are good at swiftly changing between registers, and I (even as a professional singer) am not. On that level I wouldn't blame you for just sticking to your original system.

Totally off-topic, but I just noticed that this appears to be the 10,000^th post in this subforum. (And this should be the 10,001^st.)

[Vardelm]

Totally off-topic, but I just noticed that this appears to be the 10,000^th post in this subforum. (And this should be the 10,001^st.)

Back on topic: what's the Svagnaric word for "palindrome"?

[sasasha]

Totally off-topic, but I just noticed that this appears to be the 10,000^th post in this subforum. (And this should be the 10,001^st.)

Cool! Makes me glad I wrote a second post rather than edited the two together!

[Bob]

Brief idea ...

Sounds cool. I don't have anything to add otherwise. I've done some work on comparative sign languages, though, and otherwise do have interest in the concept of languages as made with anatomy differing from that of humans.