phonemes:

these have their standard IPA values:

p, t, k, b, d, m, n, l, f, v, y (= /j/), s, h

a, e, i, o, u

these have IPA values assigned as follows:

ph /p^h/, th /t^h/, kh /k^h/, sh [english "sh" in "ship"], c [english "ch" in "church"], ng ["ng" of english "sing"], mg [velarized /m/], g /'/ (glottal stop).

this results in a language that has a tripartite distinction of voiced, unvoiced, and unvoiced aspirate. the only other detail to note is that an apostrophe is used to distinguish a digraph from two separate phonemes. eg:

kakhebel = ka-khe-bel

kak'hebel = kak-he-bel

this is only done when there is a possibility of confusion on the part of a native speaker. thus, suphebel 'you are ill' and sup'hebel 'you have endeared yourself' are never actually distinguished in writing.

as for phonotactics, tuvekkha permits 3 syllable types: V, CV, CVC. ie, a syllable can consist only of either a vowel, a consonant followed by a vowel, or a consonant plus vowel plus a second consonant (where "y" is treated as a consonant rather than a vowel off-glide). in addition, neither the initial "yi" nor the copula "iy" are permissible in tuvekkha. this results in a group of 5 + (13*5) - 1 + (13*5*13) - (13 + 13) = 5 + 65 - 1 + (169*5) - 26 = 69 - 26 + 845 = 43 + 845 = 888 syllables in the language.

obviously, this requires that there be a fair amount of homonymy in the language for it to be sufficiently productive, though that problem is simplified somewhat by the fact that there are only 2 classes of morpheme - we'll get to that next...

grammar

at its simplest, tuvekhkha is a SOV language. this means that the subject comes first, then the object, then the verb. the trick is that there are really only two classes of lexeme in tuvekha: roots (read: verbs) and suffixes. thus, the classic noun-verb distinction does not, strictly speaking, apply to tuvekha. this results in the phrase "mgus-mgus-mgus-mgus-hel-bet": 'i, who am hungry, eat food hungrily' (lit. 'the hungry one eats food hungrily'). (how the lexemes & suffixes are applied will be covered shortly.)

it might also be described as a 'quantum language' to some extent, in the sense that in relatively simple sentences, the role of each morpheme is not clear until the end of the utterance. this is the result of the fact that a root can be uttered long before its companion suffix ever "sees light of day". this will become clearer as we delve into the grammar.

so... without further ado:

to start with verbs (we will use standard morphosyntactic typology for convenience's sake); verbs are simply a root with no suffix. eg, "mgus" = `hungry, to be hungry'. so far, pretty easy, right?

nouns are formed by the selection of a lexeme and a noun-classifier suffix - "bet" being one of the most common. thus su + bet --> su-bet "you (singular, agent of phrase)", or (to use another suffix) su + bekh --> su-bekh "you (singular, patient of phrase)".

follow me, so far? okay... this is where it gets a little tricky. unlike most languages, which place the suffix immediately after its root, tuvekkha places the suffix at the end of the phrase. this results in a sentence that states: sumgusbet --> su-mgus-bet : `you are hungry'.

to further complicate matters, suffixes move out in a sequence that mirrors the original sequence leading to the verb. thus, 'you sicken me' is su-a-phe-bekh-bet --> suaphebekhbet. to break it down:

su   a   phe   bekh   bet

|   |    |      |     |

  |   |   verb    |     |  

|   |           |     |

|   |___________|     |

|        |            |

|      object         |

|                     |

|_____________________|

|

subject

obviously, this requires that both the speaker and listener not only keep track of morphemes, but be able to "trace back" the sequence of morphemes in order to assign them their correct grammatical role. this "nesting" structure is critical to understanding spoken tuvekkha.

Last updated 07.Dec.97, 10:31 CDT.