Talking drums conlang: Chiranyi

[Linguist_Wannabe]

Recently I have become very interested in talking drums http://en.wikipedia.org/wiki/Talking_drum and have decided to make a conlang that would lend itself well to being played on them. It is to be spoken in my far-future Antarctican conworld (after runaway global warming has melted all of the ice). It is influenced by languages spoken near oceans that refugees could cross to get to Antarctica.

The language has a register tone system with 5 levels, and vertical vowel system, with only 3 phonemic vowel qualities /a/, /ə/ and /ɨ/. However these have many allophones depending on the surrounding consonants. For example Chiranyi is pronounced [tʃʰiɾæ̋ɲí], but phonemically it is /tʰjɨɾa̋njɨ́/. The two /j/ phonemes have fronted the surrounding /ɨ/ and /a/ phonemes to and [æ] respectively.

There are also two diphthong phonemes, /aɨ/ and /əɨ/.

Here are the consonant phonemes:

Nasals: /m/, /n/, /ŋ/
Aspirated Stops: /pʰ/, /tʰ/, /kʰ/
Plain Voiceless Stops: /p/, /t/, /k/, /ʔ/
Voiced Stops: /b/, /d/
Affricates: /tɬ/
Flaps: /ɾ/
Aspirated Fricatives: /sʰ/
Fricatives: /s/, /ɬ/, /h/
Approximants: /w/, /ɥ/, /ɹ/, /l/, /j/, /ʢ/

Chiranyi also has a chroneme http://en.wikipedia.org/wiki/Chroneme. It can lengthen vowels and non-initial /p/, /t/, /k/ (but not the aspirated versions). Examples include /ʔə̋ːkata̋/ - "the mass migration south to Antarctica from the northern continents", and /pʰə̀ŋkːwan/ - "penguin". However a consonant cannot be lengthened if the vowel preceding it was long e.g. */pʰə̀ːŋkːwan/ would not be a possible word. However it is permitted to have geminate consonants before long vowels e.g. /tʰápːəː/ - "despair".

The phonotactics are relatively simple, with the range of permitted syllable structures being C (C) V (C). The second element of a syllable initial cluster can only be /w/, /ɥ/, /j/ or /ʢ/. In non final syllables, the syllable coda can only be a nasal, which must be homorganic with the following consonant. So while /pʰə̀ŋkːwan/ - "penguin" is phonologically permitted, */pʰə̀ŋtːwan/ or */pʰə̀mkːwan/ are not.

Nasals are also permitted in final syllables. However they are realised simply as nasalisation of the preceding vowel e.g. /pʰə̀ŋkːwan/ - "penguin" is pronounced [pʰòŋkːwɔⁿ]. But the nasals are still present underlyingly, as we can see when we add the clitic /jɨ/ to form /pʰə̀ŋkːwanjɨ/ - "penguins (in general)", pronounced as [pʰòŋkːwœɲi].

In word final syllables, plain voiceless stops can also occur as coda consonants. However if they are not geminated, then they are debuccalised to [ʔ] e.g. /pʰaːt/ - "bird", is pronounced as [pʰaːʔ], and /njɨ̏k/ - "year" is pronounced as [ɲȉʔ]. Geminated stops at the end of words are pronounced as plain voiceless stops e.g. /sɥȁkː/ - "house", is pronounced as [sɥœ̏k]/

Likewise, when a process such as cliticisation means that such a coda consonant is no longer the final consonant of a word, then the above processes no longer occur e.g. /pʰaːtjɨ/ - "birds (in general)" -[pʰæːtji], /njɨ̏kji/ - "years (in general)" - [ɲȉci], /sɥȁkːjɨ/ - "houses (in general)" - [sɥœ̏cːi].

Chiranyi also has a Grassman`s law style restriction that prohibits more than a single aspirated consonant in a word. If a process such as cliticisation or compounding would result in a word with more than one aspirated, all of the aspirated consonants except the last are deaspirated e.g. when /pʰə̀ŋkːwan/ - "penguin" is combined with the definite article /tʰja/ (note that this comes after the noun), the result is /pə̀ŋkːwantʰja/ - "the penguin(s)", not */pʰə̀ŋkːwantʰja/.

One more thing I would like to point out is the tendency of certain consonants to fuse with /j/. /sʰj/ undergoes assimilation to [ʃ], and likewise /tʰj/ assimilates to [tʃʰ] (as in the pronunciation of Chiranyi). However for the purposes of the phonology, these are still aspirated consonants, as we can see when /tʰjɨɾa̋njɨ́/ - "Chiranyi" takes the definite article, become /tjɨɾanjɨtʰja/ - "the Chiranyi".

Similarly, /dj/ assimilates to [dʒ], and /nj/ and /ŋj/ both merge to [ɲ] (as in the pronounciation of the name "Chiranyi").

Please let me know of any thoughts you have on the phonology :)

[Ahzoh]

Wait... how can you have vowels and consonants, when the pagesaid talking drumscannot emulate such qualities...

[Man in Space]

Wait... how can you have vowels and consonants, when the pagesaid talking drumscannot emulate such qualities...

I'm assuming it's something like Pirahã or some African languages where either just using a few phonemes or manipulating the drum can produce tonal contours that can be understood by fluent speakers by using inference/context.

[Linguist_Wannabe]

Wait... how can you have vowels and consonants, when the pagesaid talking drumscannot emulate such qualities...

I'm assuming it's something like Pirahã or some African languages where either just using a few phonemes or manipulating the drum can produce tonal contours that can be understood by fluent speakers by using inference/context.

Yes that was the idea.

[Linguist_Wannabe]

A bit more on allophones and assimilation.

The cluster /kj/ assimilates to [c], e.g. /sɥȁkːjɨ/ - "houses (in general)" - [sɥœ̏cːi]. /kʰj/ is also sometimes heard as [cʰ], but more often as [tɬʰ] e.g. /kʰjàːn/ - "aluminium", is heard as [cʰæ̀ːⁿ] or [tɬʰæ̀ːⁿ].

/hj/ assimilates to [ç] e.g. /hjɨ̀nɘ̏tːɘ/ - "nomad" is pronounced [çìnɘ̏tːɘ].

/lj/ assimilates to [ʎ] e.g. /kʢàljɨtʰwɘ/ - "warm" is pronounced [qàʎʉtʰwo].

In the above example, we can also see that /kʢ/ assimilates to [q]. Likewise, /kʰʢ/ also assimilates to [qʰ] e.g. /nàkʰʢa̋ɨŋ/ - "nose" is pronounced [nɑ̀qʰɑ̋ɨⁿ].

Clusters of /ʔʢ/ (glottal stop + epiglottal approximant) assimilate to epiglottal stops [ʡ] e.g. /pʰɘ̏ɨʔʢa/ - "fruit farmer" is pronounced [pʰɘ̏ɯʡɑ].

Clusters of /hʢ/ assimilate to epiglottal fricatives [ʜ] e.g. /hʢa̋ɨŋwɘ/ - "old" is pronounced [ʜɑ̋uŋwo].

Clusters of /lʢ/ assimilate to [ɫ] e.g. /ɾwə̏lʢaɨŋ/ - "to burn" is pronounced [ɾwɔ̏ɫɑɨⁿ].

When followed by a glottal stop, diphthongs and long vowels are pronounced with a rise in pitch at the end (e.g. the vowel in /pʰaːt/ -> [pʰaːʔ] - "bird" starts with mid pitch but rises at the end).

If a word ends in a diphthong or a long vowel, then is pronounced with a fall in pitch at the end (e.g. the last vowel in /tʰápːəː/ - "despair" starts with mid pitch but falls at the end).

[xijlwya]

Could you describe how the language is "spoken" on the drum in more detail? I'd be most interested in it.

[Creyeditor]

I remember something similar, inspired by indian drums, IIRC. Too bad I forgot the name of that conlang ...

[Linguist_Wannabe]

Could you describe how the language is "spoken" on the drum in more detail? I'd be most interested in it.

Sure. One beat of the drum corresponds to one syllable. The pitch of the drum is modulated to reflect each of the 5 tones. Short monophthongs are played as "stopped" notes, where the drum is touched immediately after it is hit to stop it vibrating. If they are followed by a geminate /p/, /t/ or /k/, then there is a pause before the next syllable.

If a diphthong or a long vowel is followed by a glottal stop (pronounced with an allophonic rising pitch), then after the drum is struck, the pitch of the drum is adjusted while it is still vibrating, to produce a warbling rising pitch on the note. A similar thing is done for words that end in diphthongs or long vowels (which allophonically have falling pitch).

With 5 tones and 5 playing styles (normal, stopped, stopped + pause, rising, falling), then there are 25 possible patterns for each syllable. Hence there are 25*25 = 625 possible 2-syllable words, and 25*25*25 = 15,625 possible 3 syllable words that are distinct on the drums.