Miwonša

Miwonša is an a priori conlang created in 2011 by Plusquamperfekt (member of http://www.unilang.org and http://www.incatena.org ). The grammar is essentially based on earlier conlang projects (for example Kĵatonša) that have not been finished. One unique feature of Miwonša is that it can be also written with Tarul, a conscript created in 2011.

PHONOLOGY
Green: phonemes / Blue: graphemes

''

Some additional comments:

(1) /b d g v/ occur only in loanwords and are not considered native phonemes.

(2) [tɕ tɕ'] are allophones of /c c'/ before /i, ĩ/

(3) When preceding /i, ĩ/, the graphemes  that usually represent alveolar consonants are used to indicate that the consonants are palatal:

 - [tɕi]; - [tɕ'i];  - [ɲi];  - [ɕi];  - [ʎi]

(4) /ɬ/ can be replaced by /ɕ/, /ɬw/ by /ʍ/. (5) /u/ cannot occur after /w, ʍ/, /i/ cannot occur after the phonemes /t, t', n, h, l, j/.

LATIN ORTHOGRAPHY
A (B) C Č (D) E F (G) H I J K L M N O P R S Š T U (V) W Y Z Ž

IMPORTANT: In this article accent marks (´) are used to help the reader to put the stress on the right position. In the official orthography, accents are not needed.

Basics
Tarul essentially resembles an alphasyllabary or abudiga, in that elements representing sounds are grouped together to complex symbols or blocks. The following example shows the words "Tarul" and "Miwonša" in Tarul and how the blocks are composed:



The classification of Tarul is difficult, though, as according to wikipedia vowels are "obligatory but secondary" in alphasyllabaries, which is not the case in Tarul. Another feature that militates against being an alphasyllabary is the complexity of blocks. A simple block can represent up to 6 phonemes, a splitted block (representing two syllables) can even contain 10 sounds. All in all, in Miwonša there is a 1 syllable :1 block ratio, if we consider splitted blocks to be ligatures of two simple blocks. The following figure shows how sounds in simple blocks are organized:



Basically all Tarul blocks are composed of two parts. The upper consonant half and the lower vowel half. The most complex syllable structure that Miwonša allows is [(S)C1(L)]V1(V2)(C2)]. "C" is a consonant, "V" is a vowel, "S" is either /s/ or /ʃ/ and "L" is /w/, /j/, /r/ or /l/.

Example 1:
 * "manskja" (school) → Syllable structure: (C1.V1.V2.C2)-(C1.L.V1) → Block 1 (2+6), Block 2 (3+5)

Explanation: As [mus] is a CVNC syllable, the upper half has to look like #2 and the lower half like #6 in the table. As [kja] is a CLV syllable, the upper half has to look like #3 and the lower half like #5 in figure "2".

Example 2:
 * "rušoi" (green) → Syllable structure: (C1.V1)-(C1.V1.V2) → Block 1 (1+5), Block 2 (1+6)

Explanation: As [lu] is a CV syllable, the upper half has to look like #1 and the lower half like #5 in the table. As [šoi] is a CVV syllable, the upper half has to look like #1 and the lower half like #6 in figure "2".

Vowels
Miwonša has 5 oral vowels, 5 nasal vowels and 8 diphthongs: . The next figure shows how to write them:



Note: The diphthong  is not used in modern Miwonša anymore.

Consonants
Apart from [ɕ] and [ʍ], which are represented by two elements (which could be described as "digraphs"), all phonemes have own graphemes in Tarul.

The following table shows all plosives, fricatives and affricates:



The next table shows the remaining consonant sounds and sound combinations with own graphemes:

(Please note: The grapheme <ł> was used for the alveolar lateral fricative <ɬ>, which is now represented by the digraph  in newer versions of the Latin orthography.)



Placeholders
The sections above show how to write words with regular syllable structures that do not violate the phonotactics of Miwonša. Three problems, however, have not been mentioned yet:
 * 1. What if a syllable starts with a vowel?
 * 2. How do we write loanwords with complex onsets and codas that cannot be displayed in one block?
 * 3. What if there is an abbrevation (for example: containing only consonants)?

The answer is that we need placeholders, which can be used to substitute vowels or consonants:

The first block shows the complete placemarker. To substitute a consonant, the upper half of the placeholder has to be used. To substitute a vowel, the lower half has to be used. Block 2 and 3 show the isolated sounds "a" and "p".

Splitted Consonants
In CVC syllables the upper half of the block gets splitted. This means that the upper half contains the left half of C1 (onset) and the right half of C2 (coda). Unfortunately two problems occur.
 * 1st problem: Some consonants look identical except for the vertical line in the middle (like "p" and "t")
 * 2nd problem: If C1=C2 (like "mam", "tat", "sas" etc.), then the CVC block looks exactly like the CV block

These difficulties can be removed by using diacritics which are explained in the following subsections:

What to do with the vertical line
Splitted consonants are always separated by vertical lines in the middle, whether the unsplitted consonants would have that line as well or not. Consonants that would not have it in a CV syllable can be distinguished from consonants that would by adding a little dot which indicates that the line in the middle would not be there in a CV syllable:



What to do if C1 is C2
Like in the first problem, we have consonants that would have a vertical line in the middle and consonants that would not. Therefore there are two diacritics to indicate that C1 is C2 - otherwise it would not be possible to distinguish [pap] from [tat]. Consonants which do not have this stroke in a simple syllable are indicated with two small horizontal lines. Consonants that always have that line get a little " fork ".

As only few consonants can be C2 in Miwonša, only few symbols with forks or horizontal lines are common. Of course, in loanwords other combinations are possible as well (for example "f_f or "n_n")

Important comment about [m]: When [m] is splitted, it looks differently than the simple [m] (see tables above). (Please note: The phoneme /r/ can only appear in the coda in loanwoards.)

Complex onsets
As mentioned above, onsets can contain up to three consonants (SCL), if [S] is /s, š/ and L is /r, l, w, j/. If the onset has the structure (S)C, block type #1 or #2 is needed, if there is (S)CG, #3 or #4 is needed.

To add "s" or "š" we have to add one [s] or two [š] two horizontal lines on each side in the lower half:

Splitted blocks
Splitted blocks can be easily distinguished from simple blocks, as the lower half is either asymmetrical or containing the same diacritics as splitted consonants. The rules for splitting consonants are the same as in simple blocks. Blocks can be splitted whenever there are two simple blocks containing no coda consonants.

A good example is the word "Miwonša":



SAMPLE TEXT IN TARUL


Please notice: The text sample was created in the first half of 2011. As both grammar and vocabulary have been changed a couple of times in the past twelve months, the text does not represent Modern Miwonša (May 2012) anymore.

GRAMMAR
==VOCABULARY==