|
|
(368 intermediate revisions by 5 users not shown) |
Line 1: |
Line 1: |
| | ===Introduction=== |
|
| |
|
| | Thulean is the sole surviving member of the Insular branch of the Hercynian language family. It has an estimated 33 million speakers. The language is agglutinative and polysynthetic with an active-stative morphosyntax of the fluid-S subtype. It is written using the Latin alphabet as is universal amongst Western European languages. |
|
| |
|
| <!--
| | ===Phonology=== |
|
| |
|
| This is a short reminder of the language format policy.
| | A) PHONEME INVENTORY |
|
| |
|
| I. Write a short piece stating your intents and purposes when creating the language (Design goal, inspiration, ideas, and so on).
| | 1) CONSONANTS |
| II. Write a short introduction to your language. (Who speaks it? When was it created? By whom? or what? are some example questions that can be answered here)
| |
| III. Once done, try making sure everything is properly spelt so as to avoid unnecessary reader fatigue.
| |
|
| |
|
| -->
| | NASAL: |
|
| |
|
| ==Introduction==
| | -fortis: /mb, nd, Ng/ |
|
| |
|
| ===Description===
| | -lenis: /m, n, N/ |
| The Thulean language (endonym: Minegkaksi) has an agglutinative and polysynthetic morphology. Its morphosyntactic alignment is split intransitive with the fluid-S subtype. It is left-branching and double-marking. Thulean is a language isolate and has not been proven to belong to any established language family, although speculative suggestions of links between it and the Uralic or Uralo-Siberian families have been made in the past. It is spoken by the Selhat or Thuleans and has a total number of speakers exceeding 30 million.
| |
|
| |
|
| ===The Thurse===
| | LIQUID: |
|
| |
|
| Thurse is the collective name in English for the European pygmy phenotype. Thurse males and females are similar in stature and have an average height of between 135 to 140 cm. They are pale-skinned with a distinctive pattern of dark stripes for which their clade is best known and have straight or wavy hair. Red hair is more common among the Thurse than any other phenotypical group. High androgyny among Thurse males and high neoteny generally are also notable traits. The Thurse belong to several ethnic groups, the largest of which is the Selhat or Thuleans who occupy Thule (endonym: Telku) the northernmost of the British Isles. The Thurse phenotype is believed to have originated in the Hercynian forest zone of central Europe, diverging from other Palaeolithic populations in Europe approximately ten thousand years ago. The Thurse have no connection with the pygmy populations of the tropics.
| | -fortis: /rd, ld/ |
|
| |
|
| ===Influences===
| | -lenis: /r, l/ |
|
| |
|
| The phonology is inspired by Finnish and Quenya phonaesthetics but twisted towards my sensibilities. I just love lateral obstruents, you know? And I'm not excessively fond of /f/
| | PLOSIVE: |
| Also, Northern Sami has left its mark. The morphology is heavily influenced by Yupik and Inuktitut.
| |
|
| |
|
| ===Goals===
| | -fortis: /pp, tt, kk/ |
|
| |
|
| My intention is to create a polysynthetic elflang that is relatively easy for me to pronounce and which won't have a grammar too conplex for me to use. We'll see how it goes.
| | -lenis: /p, t, k/ |
|
| |
|
| -->
| | FRICATIVE: |
|
| |
|
| <!-- ***Phonology*** -->
| | -fortis: /ss, xx/ |
| <!-- What sounds does your language use? -->
| |
| <!-- Here are some example sub-/other categories:
| |
|
| |
|
| Vowel inventory
| | -lenis: /s, x/ |
| Consonant inventory
| |
| Syllable structure
| |
| Stress
| |
| Intonation
| |
|
| |
|
| -->
| | APPROXIMANT: /v, D, j/ |
|
| |
|
| ==Phonology==
| | 2) VOWELS |
| ===Orthography===
| |
| Thulean is written in the Latin script. The orthography is phonemic but does show the effects of sandhi and consonant gradation. The current alphabet was adopted in 1898 and underwent its last revision in 1965. The orthography is listed in the tables below:
| |
|
| |
|
| | a) MONOPHTHONGS: /i, u, a/ |
|
| |
|
| 1) Consonants
| | b) DIPHTHONGS: /@U, OI, aI, aU/ |
|
| |
|
| {| border="1" cellpadding="1" cellspacing="1" class="bluetable lightbluebg" style="width: 660px; text-align:center;"
| | B) ALLOPHONES |
| ! colspan="3" rowspan="2"| !! rowspan="2"| Labial !! colspan="2"|Alveolar !! rowspan="2"|Palatal !! colspan="2"|Velar !! rowspan="2"|Glottal
| |
| |-
| |
| ! Central !! Lateral !! Plain !! Labialised
| |
| |-
| |
| ! rowspan="4"| Sonorant !! rowspan="2"| Nasal !! Fortis
| |
| | mm || nn || || nnj || gg || ggw || rowspan="5"|
| |
| |-
| |
| ! Lenis
| |
| | m || n || || nj || g || gw
| |
| |-
| |
| ! rowspan="2"| Liquid !! Fortis
| |
| | || rr || ll || llj || colspan="2"|
| |
| |-
| |
| ! Lenis
| |
| | || r || l || lj || colspan="2"|
| |
| |-
| |
| ! rowspan="2" colspan="2"|Stop !! Fortis
| |
| | pp || tt || ttl || ttj || kk || kkw
| |
| |-
| |
| ! Lenis
| |
| | p || t || tl || tj || k || kw || x
| |
| |-
| |
| ! rowspan="2" colspan="2"|Fricative !! Fortis
| |
| | || ss || ssl || ssj || qq || qqw || rowspan="3"|
| |
| |-
| |
| ! Lenis
| |
| | || s || sl || sj || q || qw
| |
| |-
| |
| ! colspan="3"| Approximant
| |
| | v || colspan="2"| || j || h || w
| |
| |}
| |
|
| |
|
| | 1) /N/ is realised as [g] in word-initial position and in onset position following a coda consonant. |
|
| |
|
| 2) Vowels | | 2) All fortis obstruents are realised as lenis plosives in intervocalic position. |
|
| |
|
| {| border="1" cellpadding="1" cellspacing="1" class="bluetable lightbluebg" style="text-align:center; border-spacing: 20px;"
| | 3) All lenis plosives and /s/ are voiced in intervocalic position. |
| |-
| |
| ! colspan="2"|
| |
| ! Front
| |
| ! Back
| |
| |-
| |
| ! rowspan="2" | Monophthong
| |
| ! High
| |
| | i
| |
| | u
| |
| |-
| |
| ! Low
| |
| | e
| |
| | a
| |
| |-
| |
| ! colspan="2" | Diphthong
| |
| | ai
| |
| | au
| |
| |}
| |
|
| |
|
| | 4) /x/ is realised as [Z] in intervocalic position. |
|
| |
|
| Note that the glottal stop is not written word-initially.
| | 5) /D/ is realised as [T] in word-final position. |
|
| |
|
| ===Consonants===
| | 6) /i, u/ are realised as [E, O] when preceding a liquid coda or fortis liquid. |
|
| |
|
| Thulean has 43 consonant phonemes which according to WALS is a large inventory. These are listed in the table below:
| | 7) /i, u/ are reslised as [E, O] when adjacent to /v, j/. |
|
| |
|
| | C) ORTHOGRAPHY |
|
| |
|
| {| border="1" cellpadding="1" cellspacing="1" class="bluetable lightbluebg" style="text-align:center;"
| | "a" = /a/ |
|
| |
|
| ! rowspan="2" colspan="3"|
| | "ai" = /aI/ |
| ! rowspan="2"| Labial
| |
| ! colspan="2"| Alveolar
| |
| ! rowspan="2"| Palatal
| |
| ! colspan="2" |Velar
| |
| ! rowspan="2"|Glottal
| |
| |-
| |
| ! |Central
| |
| ! |Lateral
| |
| ! |Plain
| |
| ! |Labialised
| |
| |-
| |
| ! rowspan="4" | Sonorant
| |
| ! rowspan="2"| Nasal
| |
| ! Fortis
| |
| | /mb/
| |
| | /nd/
| |
| |
| |
| | /ɲʤ/
| |
| | /ŋg/
| |
| | /ŋg<sup>w</sup>/
| |
| |
| |
| |-
| |
| ! Lenis
| |
| | /m/
| |
| | /n/
| |
| |
| |
| | /ɲ/
| |
| | /ŋ/
| |
| | /ŋ<sup>w</sup>/
| |
| |
| |
| |-
| |
| ! rowspan="2"| Liquid
| |
| ! Fortis
| |
| |
| |
| | /rd/
| |
| | /ld/
| |
| | /ʎʤ/
| |
| |
| |
| |
| |
| |
| |
| |-
| |
| ! Lenis
| |
| |
| |
| | /r/
| |
| | /l/
| |
| | /ʎ/
| |
| |
| |
| |
| |
| |
| |
| |-
| |
| ! rowspan="2" colspan="2"| Stop
| |
| ! Fortis
| |
| | /pp/
| |
| | /tt/
| |
| | /ttɬ/
| |
| | /ttʃ/
| |
| | /kk/
| |
| | /kk<sup>w</sup>/
| |
| |
| |
| |-
| |
| ! Lenis
| |
| | /p/
| |
| | /t/
| |
| | /tɬ/
| |
| | /tʃ/
| |
| | /k/
| |
| | /k<sup>w</sup>/
| |
| | /ʔ/
| |
| |-
| |
| ! rowspan="2" colspan="2"|Fricative
| |
| ! Fortis
| |
| |
| |
| | /ss/
| |
| | /ɬɬ/
| |
| | /ʃʃ/
| |
| | /xx/
| |
| | /xx<sup>w</sup>/
| |
| |
| |
| |-
| |
| ! Lenis
| |
| |
| |
| | /s/
| |
| | /ɬ/
| |
| | /ʃ/
| |
| | /x/
| |
| | /x<sup>w</sup>/
| |
| |
| |
| |-
| |
| ! colspan="3" | Approximant
| |
| | /v/
| |
| |
| |
| |
| |
| | /j/
| |
| | /h/
| |
| | /w/
| |
| |
| |
| |}
| |
|
| |
|
| ===Vowels=== | | "au" = /aU/ |
|
| |
|
| Thulean has 4 vowel phonemes which according to WALS is a small inventory. There are also 2 diphthongs. The consonant to vowel ratio is 10.75 which according to WALS is high. The vowels are listed in the table below:
| | "g" = /N/ |
|
| |
|
| | "gg" = /Ng/ |
|
| |
|
| {| border="1" cellpadding="1" cellspacing="1" class="bluetable lightbluebg" style="text-align:center; border-spacing: 20px;"
| | "h" = /x/ |
| |+
| |
| |-
| |
| ! colspan="2"|
| |
| ! Front
| |
| ! Back
| |
| |-
| |
| ! rowspan="2" | Monophthong
| |
| ! High
| |
| | /i/
| |
| | /u/
| |
| |-
| |
| ! Low
| |
| | /ɛ/
| |
| | /ɑ/
| |
| |-
| |
| ! colspan="2" | Diphthong
| |
| | /ai/
| |
| | /au/
| |
| |}
| |
|
| |
|
| ===Allophony=== | | "hh" = /xx/ |
| 1) Fortis obstruents are realised as their corresponding lenis obstruents in intervocalic position.
| |
|
| |
|
| 2) Lenis stops are aspirated in word-initial position.
| | "i" = /i/ |
|
| |
|
| 3) /p, t, ʧ, k, kʷ, s, ʃ, x, xʷ/ are voiced in intervocalic position.
| | "iu" = /@U/ |
|
| |
|
| 4) /tɬ/ is realised as [dð] in intervocalic position.
| | "j" = /j/ |
|
| |
|
| 5) /ɬ/ is realised as [θ] in onset position following a coda consonant of /n, r/.
| | "k" = /k/ |
|
| |
|
| 6) /ɬ/ is realised as [ð] in intervocalic position.
| | "kk" = /kk/ |
|
| |
|
| 7) /x/ is realised as [ʃ] in coda position.
| | "l" = /l/ |
|
| |
|
| 8) The fortis sonorants /mm, nn, ɲɲ, ŋŋ, ŋŋʷ, rr, ll, ʎʎ/ are realised as [mb, nd, ɲʤ, ŋg, ŋgʷ, rd, ld, ʎʤ].
| | "ll" = /ld/ |
|
| |
|
| 9) /ɑ/ is realised as [ɔ] before a liquid coda.
| | "m" = /m/ |
|
| |
|
| ===Prosody=== | | "mm" = /mb/ |
| ====Stress====
| |
| Thulean has stress accent. Primary stress falls within the root. Open syllables without a diphthong are light; open syllables with a diphthong or closed syllables are heavy. Fortis consonants close the syllable preceding them. If the first syllable of a root is heavy then the primary stress falls there otherwise it falls upon the second syllable of the root. In compound nouns, the primary stress falls within the second root.The stress placement within a root is fixed so is not affected by consonant gradation. Anomalous stress is indicated by an acute accent.
| |
|
| |
|
| ====Intonation==== | | "n" = /n/ |
| Secondary stress falls upon every alternate syllable after the primary stress. This gives Thulean a broadly iambic rhythm. Thulean does not have phonemic tone.
| |
|
| |
|
| ===Phonotactics===
| | "nn" = /nd/ |
| <!-- Explain the consonant clusters and vowel clusters that are permissible for use in the language. For example, "st" is an allowed consonant cluster in English while onset "ng" isn't. -->
| |
| 1) The syllable template in Thulean is CV(C).
| |
|
| |
|
| 2) Permitted syllable codas:
| | "p" = /p/ |
|
| |
|
| a) Morpheme-medial only: /ɲ, ŋ, r, l, ʎ, p, s/
| | "pp" = /pp/ |
|
| |
|
| b) Morpheme-medial and morpheme-final: /m, n, t, k, x/
| | "q" = /D/ |
|
| |
|
| 3) Fortis consonants may not occur in word-initial position.
| | "r" = /r/ |
|
| |
|
| 4) Fortis consonants may not occur in the onset of closed syllables.
| | "rr" = /rd/ |
|
| |
|
| 5) Consonant clusters may have no more than two segments.
| | "s" = /s/ |
|
| |
|
| 6) Consonant clusters may only occur across syllable boundaries.
| | "ss" = /ss/ |
|
| |
|
| 7) But suffixes may begin with a consonant cluster.
| | "t" = /t/ |
|
| |
|
| 8) Permitted consonant clusters as per table below:
| | "tt" = /tt/ |
|
| |
|
| {| border="1" cellpadding="1" cellspacing="1" class="bluetable lightbluebg" style="width: 660px; text-align:center;"
| | "u" = /u/ |
| !
| |
| ! m
| |
| ! n
| |
| ! ɲ
| |
| ! ŋ
| |
| ! ŋʷ
| |
| ! r
| |
| ! l
| |
| ! ʎ
| |
| ! p
| |
| ! t
| |
| ! tɬ
| |
| ! ʧ
| |
| ! k
| |
| ! kʷ
| |
| ! ʔ
| |
| ! s
| |
| ! ɬ
| |
| ! ʃ
| |
| ! x
| |
| ! xʷ
| |
| ! v
| |
| ! j
| |
| ! h
| |
| ! w
| |
| |-
| |
| ! m
| |
| | colspan = "8" style = "background: silver"| || mp || colspan = "5" style = "background: silver"| || mʔ || colspan = "5" style = "background: silver"| || mv || colspan = "3" style = "background: silver" |
| |
| |-
| |
| ! n
| |
| | colspan = "9" style = "background: silver"| || nt || ntɬ || colspan = "3" style = "background: silver" | || nʔ || ns || colspan = "8" style = "background: silver"|
| |
| |-
| |
| ! ɲ
| |
| | colspan = "11" style = "background: silver"| || ɲʧ || colspan = "5" style = "background: silver"| || ɲʃ || colspan = "6" style = "background: silver" |
| |
| |-
| |
| ! ŋ
| |
| | colspan = "12" style = "background: silver"| || ŋk || ŋkʷ || ŋʔ || colspan = "3" style = "background: silver"| || ŋx || ŋxʷ || colspan = "2" style = "background: silver" | || ŋh || style = "background: silver" |
| |
| |-
| |
| ! r
| |
| | colspan = "8" style = "background: silver"| || rp || rt || rtɬ || rʧ || rk || rkʷ || rʔ || rs || rɬ || rʃ || rx || rxʷ || rv || rj || rh || rw
| |
| |-
| |
| ! l
| |
| | colspan = "8" style = "background: silver"| || lp || lt || colspan = "2" style = "background: silver"| || lk || lkʷ|| lʔ || ls || colspan = "2" style = "background: silver"| || lx || lxʷ || lv || style = "background: silver"| || lh || lw
| |
| |-
| |
| ! ʎ
| |
| | colspan = "11" style = "background: silver"| || ʎʧ || colspan = "5" style = "background: silver"| || ʎʃ || colspan = "6" style = "background: silver" |
| |
| |-
| |
| ! p
| |
| | colspan = "15" style = "background: silver"| || ps || colspan = "2" style = "background: silver"| || px || colspan = "5" style = "background: silver"|
| |
| |-
| |
| ! t
| |
| | colspan = "15" style = "background: silver"| || ts || colspan = "2" style = "background: silver"| || tx || colspan = "5" style = "background: silver"|
| |
| |-
| |
| ! t
| |
| | colspan = "15" style = "background: silver"| || ks || colspan = "2" style = "background: silver"| || kx || colspan = "5" style = "background: silver"|
| |
| |-
| |
| ! s
| |
| | colspan = "8" style = "background: silver"| || sp|| st|| stɬ || sʧ || sk || skʷ || colspan = "10" style = "background: silver"|
| |
| |-
| |
| ! x
| |
| | colspan = "8" style = "background: silver"| || xp|| xt|| xtɬ || xʧ || xk || xkʷ || colspan = "10" style = "background: silver"|
| |
| |}
| |
|
| |
|
| 9) High vowels may not occur before a liquid coda.
| | "ui" = /OI/ |
|
| |
|
| 10) /i/ may not occur after an onset of /j/.
| | "v" = /v/ |
|
| |
|
| 11) /u/ may not occur after an onset of /w/.
| | Historical note: /D/ was originally written with "þ". However, when printing was introduced to Thule, no type was available for that letter so "q" was substituted instead. |
|
| |
|
| 12) Diphthongs may only occur in the first syllable of a root or in monosyllabic particles.
| | The spelling of Thulean is generally phonemic except that the effects of consonant gradation and sandhi are indicated. |
|
| |
|
| 13) Diphthongs may not occur in a closed syllable.
| | D) PROSODY |
|
| |
|
| 14) Diphthongs may not precede a fortis consonant.
| | Thulean has fixed initial primary stress. Secondary stress falls on every alternate syllable following the primary stress. Rhythm type is trochaic. |
|
| |
|
| 15) Diphthongs may not precede /j, w/.
| | E) PHONOTACTICS |
|
| |
|
| ===Morphophonology===
| | 1) The syllable template is (C)V(C). |
| ===Consonant Gradation===
| |
|
| |
|
| 1) Consonant gradation is word-internal lenition that effects the following:
| | 2) Consonant clusters only occur at syllable boundaries. |
|
| |
|
| a) Fortis sonorants
| | 3) Permitted consonant clusters: |
|
| |
|
| b) Fortis obstruents
| | a) /m/ + /p/ |
|
| |
|
| c) Lenis plosives following a sonorant coda or a vowel
| | b) /n/ + /t, s/ |
|
| |
|
| 2) Consonant gradation is triggered by the closing of a syllable which begins with the above classes of obstruents.
| | c) /N/ + /k, x/ |
|
| |
|
| 3) The sequences of consonant gradation are as per the tables below:
| | d) /r/ + /m, N, p, t, k, s, x, v, j/ |
|
| |
|
| | e) /l/ + /m, N, p, t, k, s, x, v, j/ |
|
| |
|
| a) Fortis obstruents, and lenis stops following a liquid:
| | f) /p, t, k/ + /s/ |
|
| |
|
| {| border="1" cellpadding="1" cellspacing="1" class="bluetable lightbluebg" style="width: 660px; text-align:center;"
| | g) /s/ + /p, t, k/ |
| ! GRADE I !! GRADE II !! GRADE III
| |
| |-
| |
| | pp || p || v
| |
| |-
| |
| | tt || t || r
| |
| |-
| |
| | ttɬ || tɬ || l
| |
| |-
| |
| | tʧ || ʧ || j
| |
| |-
| |
| | kk || k || h
| |
| |-
| |
| | kkʷ || kʷ || w
| |
| |-
| |
| | ss || s || style = "background: silver" |
| |
| |-
| |
| | ɬɬ || ɬ || style = "background: silver" |
| |
| |-
| |
| | ʃʃ || ʃ || style = "background: silver" |
| |
| |-
| |
| | xx || x || style = "background: silver" |
| |
| |-
| |
| | xxʷ || xʷ || style = "background: silver" |
| |
| |}
| |
|
| |
|
| | 4) Permitted word-final codas: /m, n, t, k, s, D/ |
|
| |
|
| b) Lenis stops following a nasal:
| | 5) Vowels in hiatus do not occur. |
| {| border="1" cellpadding="1" cellspacing="1" class="bluetable lightbluebg" style="width: 660px; text-align:center;"
| |
| ! GRADE I !! GRADE II
| |
| |-
| |
| | mp || mb
| |
| |-
| |
| | nt || nd
| |
| |-
| |
| | ntɬ || ld
| |
| |-
| |
| | ɲʧ || ɲʤ
| |
| |-
| |
| | ŋk || ŋg
| |
| |-
| |
| | ŋkʷ || ŋgʷ
| |
| |}
| |
|
| |
|
| | 6) Diphthongs may not occur in a closed syllable, before a fortis consonant or before /P, j/. |
|
| |
|
| | F) SANDHI |
|
| |
|
| 4) Consonant gradation occurs after metathesis from sandhi but before sandhi proper.
| | 1) Sandhi occurs at morpheme boundaries as a result of suffixation. |
|
| |
|
| ===Sandhi===
| | 2) The resulting sandhi transformations are as follows: |
|
| |
|
| 1) Sandhi is the term given to interactions between consonants at morpheme boundaries.
| | a) /m, n, N/ + /m, n, N/ > [mb, nd, Ng] |
|
| |
|
| 2) If a triconsonantal cluster should result from suffixation then an epenthetic /i/ is inserted after the first segment of the triconsonantal cluster.
| | b) /m, n, N/ + /r/ > [rd, rd, rd] |
|
| |
|
| 3) All other sandhi interactions are resolved as per the table below:
| | c) /m, n, N/ + /l/ > [ld, ld, ld] |
|
| |
|
| {| border="1" cellpadding="1" cellspacing="1" class="bluetable lightbluebg" style="width: 660px; text-align:center;"
| | d) /m, n, N/ + /v, D, j/ > [mb, nd, Ng] |
| !
| |
| ! m
| |
| ! n
| |
| ! ɲ
| |
| ! ŋ
| |
| ! ŋʷ
| |
| ! r
| |
| ! l
| |
| ! ʎ
| |
| ! p
| |
| ! t
| |
| ! tɬ
| |
| ! ʧ
| |
| ! k
| |
| ! kʷ
| |
| ! ʔ
| |
| ! s
| |
| ! ɬ
| |
| ! ʃ
| |
| ! x
| |
| ! xʷ
| |
| ! v
| |
| ! j
| |
| ! h
| |
| ! w
| |
| |-
| |
| ! m
| |
| | rowspan ="2"| mb || rowspan ="2"| nd || rowspan ="2"| ɲʤ || rowspan ="2"| ŋg || rowspan ="2"| ŋgʷ || rowspan ="2"| rd || rowspan ="2"| ld || rowspan ="2"| ʎʤ || rowspan ="2"| mp || rowspan ="2"| nt || rowspan ="2"| ntɬ || rowspan ="2"| ɲʧ || rowspan ="2"| ŋk || rowspan ="2"| ŋkʷ || mʔ || rowspan ="2"| ns || rowspan ="2"| nθ || rowspan ="2"| ɲʃ || rowspan ="2"| ŋx || rowspan ="2"| ŋxʷ || rowspan ="2"| mv || rowspan ="2"| ɲ || rowspan ="2"| ŋh || rowspan ="2"| ŋʷ
| |
| |-
| |
| ! n
| |
| | nʔ
| |
| |-
| |
| ! t
| |
| | colspan = "2" | nt || rowspan = "2" | ɲʧ || nt || rowspan = "2" | ŋkʷ || rt || tɬ || rowspan = "2"| ʎʧ || rowspan = "2"| pp || rowspan = "2"| tt || rowspan = "2"| ttɬ || rowspan = "2"| tʧ || rowspan = "2"| kk || rowspan = "2"| kkʷ || tt || ts || rowspan = "2"| tɬ || rowspan = "2"| ʧ|| tx || rowspan = "2"| kkʷ|| rowspan = "2"| kʷ || rowspan = "2"| ʧ || tx || rowspan = "2" | kʷ
| |
| |-
| |
| ! k
| |
| | colspan = "2" | ŋk || ŋk || rk || lk || kk || ks || kx || kx
| |
| |-
| |
| ! x
| |
| | colspan = "2" | ŋx || nʃ || ŋx || ŋxʷ || rx || lx || ʎʃ || ʃp || ʃt || ʃtɬ || ʃʧ || ʃk || ʃkʷ || xʔ || ss || ɬɬ || ʃʃ || xx || colspan = "2" | xxʷ || ʃ || xx || xʷ
| |
| |}
| |
|
| |
|
| | e) /m, N/ + /t, s/ > [nt, ns] |
|
| |
|
| 4) If owing to suffixation a high vowel should precede a liquid coda then it is lowered to its corresponding low vowel.
| | f) /m, n/ + /k, x/ > [Nk, Nx] |
|
| |
|
| ==Morphology==
| | g) /n, N/ + /p/ > [mp] |
| <!-- How do the words in your language look? How do you derive words from others? Do you have cases? Are verbs inflected? Do nouns differ from adjectives? Do adjectives differ from verbs? Etc. -->
| |
|
| |
|
| <!-- Here are some example subcategories:
| | h) /r, l/ + /n/ > [nd] |
|
| |
|
| Nouns
| | i) /r, l/ + /r, l/ > [rd, ld] |
| Adjectives
| |
| Verbs
| |
| Adverbs
| |
| Particles
| |
| Derivational morphology
| |
|
| |
|
| -->
| | j) /r, l/ + /D/ > [rd, ld] |
|
| |
|
| ===General Notes===
| | k) /p, t, k/ + /p, t, k/ > [pp, tt, kk] |
|
| |
|
| Thulean morphology has the following constituents:
| | l) /p, t, k, s/ + /v/ > [ps, ts, ks, ss] |
|
| |
|
| 1) Roots: these are divided into nominal and verbal roots. The latter are divided into transitive and intransitive categories.
| | m) /p, t, k, s/ + /D/ > [ps, ts, ks, ss] |
|
| |
|
| 2) Postbases: these are derivational or adjunctival suffixes which directly follow the root. They are scope-ordered.
| | n) /p, t, k, s/ + /j/ > [ps, ts, ks, ss] |
|
| |
|
| 3) Inflectional affixes: these bear functions such as case, number, possession, agent, patient, tense amongst others. The usual bread and butter, nuts and bolts stuff that inflection does.
| | G) CONSONANT GRADATION |
|
| |
|
| 4) Particles: Conjunctions, interjections and other miscellany that do not fit into the above categories.
| | 1) This process affects fortis consonants and lenis plosives following a nasal, liquid or vowel. |
|
| |
|
| ===Nominal Morphology===
| | 2) It is triggered if the consonants in question form the onset of a closed syllable due to suffixation. |
| ===Nominal Structure===
| |
| 1) Noun template:
| |
|
| |
|
| case prefix + nominal or verbal root + postbase(s) + number suffix + possessive suffix or indefinite suffix + demonstrative suffix
| | 3) The gradation sequences are as follows: |
|
| |
|
| 2) A minimally inflected noun has a case prefix and a number suffix.
| | a) /mp/ > /mb/ > /m/ |
|
| |
|
| 3) Postbases will be covered in the own separate section after verbs.
| | b) /nt/ > /nd/ > /n/ |
|
| |
|
| ===Cases===
| | c) /Nk/ > /Ng/ > /N/ |
| {| class="bluetable lightbluebg"
| |
| |+ '''Nominal Case Declension'''
| |
| |-
| |
| ! Case
| |
| ! Prefix
| |
| ! Function
| |
| ! Examples
| |
| |-
| |
| !Absolutive
| |
| | Ø-
| |
| |
| |
| *Marks the nominal citation form
| |
| *Marks the O argument of a transitive verb
| |
| *Marks the stative S argument of an intransitive verb
| |
| *Marks the recipient of a ditransitive verb
| |
| |
| |
| |-
| |
| ! Ergative
| |
| |a-
| |
| |
| |
| *Marks the A argument of a transitive verb
| |
| *Marks the active S argument of an intransitive verb
| |
| *Marks the possessor of a possessive noun phrase
| |
| |
| |
| |-
| |
| ! Instrumental
| |
| |ki-
| |
| |
| |
| *Indicates the use of an instrument or tool
| |
| *Marks the the focus of an intransitive verb of perception, cognition or affection of stative S arguments
| |
| *Marks direct causal arguments
| |
| *Marks the theme of a ditransitive verb
| |
| |
| |
| |-
| |
| ! Locative
| |
| |jet-
| |
| |
| |
| *Indicates location or place
| |
| *Indicates time, event, or occasion
| |
| *Marks the locative comitative
| |
| *Marks infinite verbal forms in periphrastic constructions
| |
| |
| |
| |-
| |
| ! Allative
| |
| |nu-
| |
| |
| |
| *Indicates direction, goal, or destination
| |
| *Marks the focus of an intransitive verb of perception, cognition or affection of active S arguments
| |
| *Encodes spatial or temporal relations in conjunction with terminative markers (e.g. up to, until, as far as)
| |
| *Marks the beneficiary of an action
| |
| *Marks for the purpose of, for the use of
| |
| |
| |
| |-
| |
| ! Ablative
| |
| | pik-
| |
| |
| |
| *Indicates motion away from
| |
| *Indicates origin
| |
| *Indicates distance from a reference point
| |
| *Indicates time since an event occurred
| |
| *Encodes partitive relations (part of, made from, some of)
| |
| *Marks the focus of intransitive malefactive/adversarial verbs
| |
| *Encodes aversive case relations (for fear of, lest, in case of)
| |
| |
| |
| |-
| |
| ! Perlative
| |
| | me-
| |
| |
| |
| *Indicates motion through, across or along
| |
| *Indicates path, means or mode of transport or transmission
| |
| *Indicates duration
| |
| *Marks the instrumental or collaborative comitative
| |
| *Marks indirect causal arguments
| |
| *Encodes distibutive functions
| |
| |
| |
| |-
| |
| ! Equative
| |
| | sin-
| |
| |
| |
| *Indicates similarity in manner, likeness or composition
| |
| *Encodes comparative marking (as, than)
| |
| *Serves as a topicalizer (e.g. regarding, concerning, about, as for)
| |
| |
| |
| |}
| |
|
| |
|
| <!--
| | d) /rp/ > /rP/ |
| 1) Case prefixes:
| |
|
| |
|
| ABS(olutive): Ø-
| | e) /rt/ > /rd/ > /r/ |
|
| |
|
| NOM(inative): a-
| | f) /rk/ > /rj/ |
|
| |
|
| INST(rumental): ki-
| | g) /lp/ > /lP/ |
|
| |
|
| LOC(ative): jet-
| | h) /lt/ > /ld/ > /ll/ |
|
| |
|
| ALL(ative): nu-
| | i) /lk/ > /lj/ |
|
| |
|
| ABL(ative): pik-
| | j) /pp/ > /p/ > /P/ |
|
| |
|
| PER(lative): me-
| | k) /tt/ > /t/ > /D/ |
|
| |
|
| EQU(ative): sin-
| | l) /kk/ > /k/ > /j/ |
|
| |
|
| | m) /ss/ > /s/ |
|
| |
|
| 2) Case Functions:
| | n) /xx/ > /x/ |
|
| |
|
| a) Absolutive Case:
| |
|
| |
|
| i) Marks the citation form of a noun.
| |
|
| |
|
| ii) Marks the O argument of a transitive verb.
| | [[Category:Sketchlangs]] |
| | |
| iii) Marks the stative S argument of an intransitive verb.
| |
| | |
| iv) Marks the recipient of a ditransitive verb.
| |
| | |
| b) Nominative Case:
| |
| | |
| i) Marks the A argument of a transitive verb.
| |
| | |
| ii) Marks the active S argument of an intransitive verb.
| |
| | |
| iii) Marks the possessor of a possessive noun phrase
| |
| | |
| c) Instrumental Case:
| |
| | |
| i) Marks the use of an instrument or tool.
| |
| | |
| ii) Marks the the focus of an intransitive verb of perception, cognition or affection if the S argument is stative.
| |
| | |
| iii) Marks direct causation.
| |
| | |
| iv) Marks the theme of a ditransitive verb.
| |
| | |
| d) Locative Case:
| |
| | |
| i) Marks location or place.
| |
| | |
| ii) Marks occasion, time or event.
| |
| | |
| iii) Marks the locative comitative.
| |
| | |
| iv) Marks the infinite form of the verb in periphrastic constructions.
| |
| | |
| e) Allative Case:
| |
| | |
| i)Marks destination or motion towards.
| |
| | |
| ii) Marks the focus of an intransitive verb of an intransitive verb of perception, cognition or affection if the S argument is active.
| |
| | |
| iii) Marks terminative relations (up to, until, as far as) spatially and temporarally.
| |
| | |
| iv) Marks benefactive relations (for, on behalf of).
| |
| | |
| f) Ablative Case:
| |
| | |
| i) Marks origin or motion away from.
| |
| | |
| ii) Marks distance from, time since.
| |
| | |
| iii) Marks the focus of intransitive verbs of opposition or repulsion.
| |
| | |
| iv) Marks aversive functions (for fear of, lest, in case of).
| |
| | |
| g) Perlative Case:
| |
| | |
| i) Marks motion through, across or along.
| |
| | |
| ii) Marks duration.
| |
| | |
| iii) Marks path, means or mode of transport or transmission.
| |
| | |
| iv) Marks the instrumental or collaborative comitative.
| |
| | |
| v) Marks indirection causation.
| |
| | |
| h) Equative Case:
| |
| | |
| i) Marks similarity in manner, likeness or composition.
| |
| | |
| ii) Marks comparative functions (as, than).
| |
| | |
| iii) Marks topical functions (re, regarding, concerning, about, as for).
| |
| | |
| ===Postbases===
| |
| | |
| Postbases are covered on their own section.
| |
| -->
| |
| | |
| ===Number===
| |
| 1) For the purposes of number, Thulean nouns are divided into count nouns and mass nouns.
| |
| | |
| 2) For count nouns there are two systems of number:
| |
| | |
| a) singular-plural
| |
| | |
| b) collective-singulative
| |
| | |
| 3) The singular-plural system works as it does in most European languages, the singular form of the noun is default and marks a single instance of that noun. The plural form marks multiple instances of that noun. EG:
| |
| | |
| SG: kattu = cat
| |
| | |
| PL: katut = cats
| |
| | |
| 4) With the collective-singulative system the collective form of the noun is default and marks multiple instances of that noun. The singulative form marks a single instance of that noun. Collective nouns usually indicate entities that are found in groups. EG:
| |
| | |
| COL: makku = pigs
| |
| | |
| SGV: makuq = pig
| |
| | |
| 5) Mass nouns are considered to be pluralia tanta and thus use the collective-singulative system. The collective form indicates a lump or mass and the singulative form indicates a part of that lump or mass. EG:
| |
| | |
| COL: ninta = water
| |
| | |
| SGV: ninnuq = a drop or sip of water
| |
| | |
| 6) Number suffixes:
| |
| | |
| a) After vowel:
| |
| | |
| PL: -t
| |
| | |
| SGV: -q
| |
| | |
| b) After consonant or before suffix:
| |
| | |
| PL: -ta
| |
| | |
| SGV: -qe
| |
| | |
| ===Possession===
| |
| 1) Pronominal Possession:
| |
| | |
| This is indicated by the following set of suffixes:
| |
| | |
| 1SG: -nne
| |
| | |
| 2SG: -gke
| |
| | |
| 3SG ANIM: -nte
| |
| | |
| 3SG INAN: -mme
| |
| | |
| 4SG ANIM: -lle
| |
| | |
| 4SG INAN: -nse
| |
| | |
| 1PL EXCL: -gka
| |
| | |
| 1PL INCL: -ppa
| |
| | |
| 2PL: -kka
| |
| | |
| 3PL ANIM: -tta
| |
| | |
| 3PL INAN: -mma
| |
| | |
| 4PL ANIM: -lka
| |
| | |
| 4PL INAN: -ksa
| |
| | |
| EG:
| |
| | |
| a)
| |
| | |
| tlantante
| |
| | |
| = her children
| |
| | |
| b)
| |
| | |
| tlannaqeppa
| |
| | |
| = our child
| |
| | |
| 2) Nominal Possession:
| |
| | |
| This is indicated by the following construction:
| |
| | |
| NOM + possessor possessum + pronominal possessive suffix.
| |
| | |
| EG:
| |
| | |
| akimet katunte = the woman's cat
| |
| | |
| a-kimet kattu-nte
| |
| | |
| ERG-woman cat-3SG.ANIM.POS
| |
| | |
| ===The Indefinite===
| |
| 1) The indefinite suffix -li imparts the meaning of a, a certain, some, any.
| |
| | |
| EG:
| |
| | |
| jelelli = a valley
| |
| | |
| 2) Used with the ablative case has the meaning any of, any one of.
| |
| | |
| EG:
| |
| | |
| a)
| |
| | |
| pikselhat telli
| |
| | |
| = any one of the people
| |
| | |
| pik-selka-t tin-Li
| |
| | |
| ABL-person-PL 3SG.ANIM.PRON-INDEF
| |
| | |
| b)
| |
| | |
| pikselhat talki
| |
| | |
| = any of the people
| |
| | |
| pik-selka-t tak-Li
| |
| | |
| ABL-person-PL 3PL.ANIM.PL.INDEF
| |
| | |
| ===Demonstratives ===
| |
| 1) Thulean has four demonstrative suffixes which encode the following distances:
| |
| | |
| a) The proximal citerior which marks a person or object near the speaker.
| |
| | |
| b) The distal citerior which marks a person or object near the addressee.
| |
| | |
| c) The proximal ulterior which marks a person or object away from both speaker and addressee but within line of sight.
| |
| | |
| d) The distal ulterior which marks a person or object away from both speaker and addressee but outside visual range.
| |
| | |
| 2)The demonstrative suffixes are as follows:
| |
| | |
| PROX CIT: -ksi
| |
| | |
| DIST CIT: -psu
| |
| | |
| PROX ULT: -tje
| |
| | |
| DIST ULT: -nja
| |
| | |
| 3) Examples:
| |
| | |
| a)
| |
| | |
| janaqiksi
| |
| = this man
| |
| | |
| b)
| |
| | |
| avallatitje
| |
| = yon apples that are in sight
| |
| | |
| ===Gender===
| |
| | |
| 1) Thulean has two genders, animate and inanimate.
| |
| | |
| 2) The animate gender contains nouns referring to people, animals and dynamic physical phenomena such as fire or wind.
| |
| | |
| 3) The inanimate gender contains the residuum.
| |
| | |
| 4) Nouns are not overtly marked for gender but they govern the appropriate pronouns, possessive suffixes and verbal pronominal markers.
| |
| | |
| EG:
| |
| | |
| a)
| |
| | |
| paslanti kattu
| |
| | |
| = the pretty cat
| |
| | |
| passla-nti Ø-kattu
| |
| | |
| be.pretty-3SG.ANIM.ABS ABS-cat
| |
| | |
| b)
| |
| | |
| helhammu ninta
| |
| | |
| = the hot water
| |
| | |
| helka-mmu Ø-ninta
| |
| | |
| be.hot-3PL.INAN.ABS ABS-water
| |
| | |
| ===Pronouns===
| |
| 1) Pronouns are treated like any other nominals.
| |
| | |
| 2) Personal Pronouns:
| |
| | |
| a) These are listed as follows:
| |
| | |
| 1SG: nin
| |
| | |
| 2SG: kin
| |
| | |
| 3SG ANIM: tin
| |
| | |
| 3SG INAN: min
| |
| | |
| 4SG ANIM: lin
| |
| | |
| 4SG INAN: sin
| |
| | |
| 1PL EXCL: nuk
| |
| | |
| 1PL INCL: puk
| |
| | |
| 2PL: kuk
| |
| | |
| 3PL ANIM: tuk
| |
| | |
| 3PL INAN: muk
| |
| | |
| 4PL ANIM: luk
| |
| | |
| 4PL INAN: suk
| |
| | |
| b) Animate pronouns can also encode the meaning of person in the generic sense.
| |
| | |
| EG:
| |
| | |
| paslanti tin
| |
| | |
| = the pretty one
| |
| | |
| be.pretty.3SG.ANIM.ABS 3SG.ANIM.PRON
| |
| | |
| c) Inanimate pronouns can also encode the meaning of object in the generic sense.
| |
| | |
| EG:
| |
| | |
| mukqalla
| |
| | |
| = big things
| |
| | |
| muk-halla
| |
| | |
| 3PL.INAN.PRON-big
| |
| | |
| d) 4th person pronouns come into play when there are two referents of the same gender and number that need to be distinguished. Thus, 4th person pronouns can be translated into English as "the latter".
| |
| | |
| e) Exclusive 1st person plural pronouns indicate that the addressee is not included within the referent.
| |
| | |
| IE: we but not you.
| |
| | |
| f) Inclusive 1st person plural pronouns indicate that the addresses is included with I the referent.
| |
| | |
| IE: we and you.
| |
| | |
| 3) Indefinite Pronouns:
| |
| | |
| a) These encode the concepts of somebody, something, anybody, anything.
| |
| | |
| b) They are formed by affixing the indefinite suffix to the appropriate pronoun.
| |
| | |
| EG:
| |
| | |
| melli = anything
| |
| | |
| c) In conjuction with the ablative case express the concepts of anyone of, any of.
| |
| | |
| EG:
| |
| | |
| i)
| |
| | |
| pitjanaqta
| |
| | |
| = anyone of the men
| |
| | |
| pik-yanaq-ta tin-li
| |
| | |
| ABL-man-PL 3SG.ANIM.PRON-INDEF
| |
| | |
| ii)
| |
| | |
| pikavalla malki
| |
| | |
| = any of the apples
| |
| | |
| pik-avalla muk-Li
| |
| | |
| ABL-apples 3PL.INAN.PRON-INDEF
| |
| | |
| 4) Reflexive Pronouns:
| |
| | |
| a) There are no dedicated reflexive pronouns.
| |
| | |
| b) Reflexivity is indicated by employing the absolutive, oblique or possessive form which refers to the ergative or sole argument of the verb.
| |
| | |
| EG:
| |
| | |
| i)
| |
| | |
| temmaiserittu
| |
| | |
| = he misleads himself
| |
| | |
| ten-maiset-ttu
| |
| | |
| 3SG.ANIM.ERG-mislead-3SG.ANIM.ABS
| |
| | |
| VS
| |
| | |
| temmaiserelku
| |
| | |
| = he misleads him
| |
| | |
| ten-maiset-lku
| |
| | |
| 3SG.ANIM.ERG-mislead-4SG.ANIM.ABS
| |
| | |
| ii)
| |
| | |
| kitin qinnunti
| |
| | |
| = he sees himself
| |
| | |
| ki-tin qintu-nti
| |
| | |
| INST-3SG.ANIM.PRON see-3SG.ANIM.ABS
| |
| | |
| VS
| |
| | |
| kilin qinnunti
| |
| | |
| = he sees him
| |
| | |
| ki-lin qintu-nti
| |
| | |
| INST-4SG.ANIM.PRON see-3SG.ANIM.ABS
| |
| | |
| iii)
| |
| | |
| avallaqente tenarqansi
| |
| | |
| = she eats her own apple
| |
| | |
| avalla-qe-nte ten-arqa-nsi
| |
| | |
| apples-SGV-3SG.ANIM.POS 3SG.ANIM.ERG-eat-4SG.INAN.ABS
| |
| | |
| VS
| |
| | |
| avallaqelle tenarqansi
| |
| | |
| = she eats her apple
| |
| | |
| avalla-qe-lle ten-arqa-nsi
| |
| | |
| apples-SGV-4SG.ANIM.POS 3SG.ANIM.ERG-eat-4SG.INAN.ABS
| |
| | |
| 5) Reflexive Pronouns:
| |
| | |
| There are no dedicated reflexive pronouns. Their functions are performed by a verbal suffix.
| |
| | |
| 6) Negative Pronouns:
| |
| | |
| a)These encode the concepts of nobody and nothing.
| |
| | |
| b) The negative pronouns are listed as follows:
| |
| | |
| 3SG ANIM: vannin
| |
| | |
| 3SG INAN: vámin
| |
| | |
| 4SG ANIM: válin
| |
| | |
| 4SG INAN: vansin
| |
| | |
| 3PL ANIM: vátuk
| |
| | |
| 3PL INAN: vámuk
| |
| | |
| 4PL ANIM: váluk
| |
| | |
| 4PL INAN: vaksuk
| |
| | |
| c) Diachronically, the negative pronouns are derived from relative clauses of negative auxiliary plus pronoun.
| |
| | |
| EG:
| |
| | |
| vannin
| |
| | |
| < *avanti tin
| |
| | |
| = the one who is not
| |
| | |
| apa-nti tin
| |
| | |
| NEG-3SG.ANIM.ABS 3SG.ANIM.PRON
| |
| | |
| 7) Demonstrative Pronouns:
| |
| | |
| a) These are derived from affixing the appropriate demonstrative suffix.
| |
| | |
| EG:
| |
| | |
| kiminiksi teghannalli
| |
| | |
| = she gives this to him
| |
| | |
| ki-min-ksi ten-hanta-lli
| |
| | |
| INST-3SG.INAN.PRON-PROX.CIT 3SG.ANIM.ERG-give-4SG.ANIM.ABS
| |
| | |
| b) Inanimate demonstrative pronouns can also encode the concepts of here and there.
| |
| | |
| EG:
| |
| | |
| numinipsu kimpalka
| |
| | |
| = you go there
| |
| | |
| nu-min-psu kin-palka
| |
| | |
| ALL-3SG.INAN.PRON-DIST.CIT 2SG.ERG.go
| |
| | |
| 8) Reflexive Pronouns:
| |
| | |
| There are no reflexive pronouns. Thulean uses gap relativisation.
| |
| | |
| EG:
| |
| | |
| kimet avalla lenarqammu kijanaq qinnunti
| |
| | |
| = the woman sees the man who eats the apples
| |
| | |
| Ø-kimet Ø-avalla len-arqa-mmu ki-janaq qinnu-nti
| |
| | |
| ABS-woman ABS-apples 4SG.ANIM.ERG-eat-3PL.INAN INST-man see-3SG.ANIM.ABS
| |
| | |
| 9) Interrogative Pronouns:
| |
| | |
| ===Verbal Morphology===
| |
| | |
| ===Verbal Structure===
| |
| | |
| ==Syntax==
| |
| ===Constituent order===
| |
| ===Noun phrase===
| |
| ===Verb phrase===
| |
| ===Sentence phrase===
| |
| ===Dependent clauses===
| |
| <!-- etc. etc. -->
| |
| | |
| ==Example texts==
| |
| ==Other resources==
| |
| <!-- Example: Word order, qualifiers, determinatives, branching, etc. -->
| |
| | |
| <!-- Template area -->
| |
| | |
| | |
| [[Category:{{PAGENAME}}]]
| |
| [[Category:Languages]] | |
| ===Thulean Lexicon===
| |
| ahaq = fire
| |
| | |
| aira = tree
| |
| | |
| alwa = lark
| |
| | |
| ana = fruits
| |
| | |
| anja = copper, small change
| |
| | |
| apa = to be not, negative auxiliary
| |
| | |
| aqqa = to hate (stative), to dislike (active)
| |
| | |
| arhat = silver, money, cash
| |
| | |
| arqa = to eat
| |
| | |
| aru = kernel, core
| |
| | |
| avalla = apple
| |
| | |
| ekwa = horse
| |
| | |
| elet = bronze
| |
| | |
| eqpe = to taste (stative), to savour (active)
| |
| | |
| esak = salmon
| |
| | |
| esel = leaves
| |
| | |
| gaxala = to heal, to doctor
| |
| | |
| hakin = to know (stative), to learn (active)
| |
| | |
| -halla = augmentative
| |
| | |
| -hansle = too much, excessively
| |
| | |
| hanta = to give
| |
| | |
| helka = to be hot (stative), to warm up (active)
| |
| | |
| isan = iron
| |
| | |
| iskalva = eagle, falcon, hawk
| |
| | |
| issit = hair
| |
| | |
| istilu = pen
| |
| | |
| jaipiq = fishes
| |
| | |
| jakken = boat
| |
| | |
| janaq = man
| |
| | |
| jelen = valley
| |
| | |
| jensa = pine tree
| |
| | |
| jiggat = to drink
| |
| | |
| kai = but
| |
| | |
| kallun = witch, sorcerer
| |
| | |
| kamma = be sick (stative), become ill (active)
| |
| | |
| kaqqwe = coffee
| |
| | |
| karra = stone
| |
| | |
| karvat = wagon, cart
| |
| | |
| katen = chains
| |
| | |
| kategkarvat= train
| |
| | |
| kategwala = laser
| |
| | |
| katta = to be tired (stative), to exhaust oneself (active)
| |
| | |
| kattu = cat
| |
| | |
| kerta = skeleton, frame
| |
| | |
| kespe = moon, month
| |
| | |
| kimet = woman
| |
| | |
| kinje = to love (stative), to like (active)
| |
| | |
| kwena = a smile
| |
| | |
| -kwepsu = never
| |
| | |
| kwersi = piece, portion
| |
| | |
| kwetsala = feathers
| |
| | |
| lakka = duck
| |
| | |
| latju = thug, bravo
| |
| | |
| -limut = should, must, ought
| |
| | |
| lira = sea ocean
| |
| | |
| -lkwa = white, blonde
| |
| | |
| lukat = mice
| |
| | |
| maiset = to mislead
| |
| | |
| -mak = can, able to
| |
| | |
| makku = pigs
| |
| | |
| makkusluse = pork
| |
| | |
| -mani = much, a lot
| |
| | |
| maslen = bread
| |
| | |
| mesalka = blackbird
| |
| | |
| -mmelet = black, brunette
| |
| | |
| nara = be located
| |
| | |
| nasura = bones
| |
| | |
| nina = ash tree
| |
| | |
| ninta = water
| |
| | |
| njara = to purr
| |
| | |
| numme = to be red
| |
| | |
| nutje = eels
| |
| | |
| palka = to go
| |
| | |
| pantju = to buy
| |
| | |
| passla = be beautiful
| |
| | |
| patata= potatoes
| |
| | |
| pattu = hare
| |
| | |
| pattuhalla = donkey
| |
| | |
| penelku = pencil
| |
| | |
| penta = to fill
| |
| | |
| pilun = lead
| |
| | |
| qaneq = willows
| |
| | |
| qaste = world
| |
| | |
| qintu = to see (stative), to watch (active)
| |
| | |
| -qpak = dimunitive
| |
| | |
| qwakka = fuck
| |
| | |
| qwenestla = window
| |
| | |
| qwestu = winds
| |
| | |
| rarka = be poor in quality (stative), be wicked (active)
| |
| | |
| rauna = to be silent (stative), to keep a secret (active)
| |
| | |
| runna = fist
| |
| | |
| russla = spirit, soul
| |
| | |
| -rusta = red, ginger
| |
| | |
| sahu = sun, day
| |
| | |
| selka = person
| |
| | |
| sinarva = gold
| |
| | |
| sinnak = fox
| |
| | |
| slaka = maggots
| |
| | |
| slasu = to hear (stative), to listen (active)
| |
| | |
| slatan = wing
| |
| | |
| -slempi = certain, sure
| |
| | |
| sluse = meat, flesh
| |
| | |
| sukkaru = sugar
| |
| | |
| sunik = snow
| |
| | |
| suvi = strawberries
| |
| | |
| tagka = tin
| |
| | |
| taha = sword
| |
| | |
| tamate = tomato
| |
| | |
| teninta = tea
| |
| | |
| tjakalat = chocolate
| |
| | |
| tjave = to feel (stative), to touch (active)
| |
| | |
| tlanta = children
| |
| | |
| unnu = pillow
| |
| | |
| unnujakken = hovercraft
| |
| | |
| uruk = foreigners
| |
| | |
| vagwa = to find (stative), to seek (active)
| |
| | |
| vakkan = rain
| |
| | |
| vanna = raven
| |
| | |
| varka = badger
| |
| | |
| varra = wolves
| |
| | |
| vassli = wasp
| |
| | |
| vasslislatan = helicopter
| |
| | |
| -verri = new, young, fresh
| |
| | |
| verru = be short (stative), to shrink (active)
| |
| | |
| vika = bee
| |
| | |
| -vinja = old, mature, wise
| |
| | |
| vussu = mouth
| |
| | |
| vuta = house
| |
| | |
| wala = light
| |
| | |
| wervu = amber
| |
Introduction
Thulean is the sole surviving member of the Insular branch of the Hercynian language family. It has an estimated 33 million speakers. The language is agglutinative and polysynthetic with an active-stative morphosyntax of the fluid-S subtype. It is written using the Latin alphabet as is universal amongst Western European languages.
Phonology
A) PHONEME INVENTORY
1) CONSONANTS
NASAL:
-fortis: /mb, nd, Ng/
-lenis: /m, n, N/
LIQUID:
-fortis: /rd, ld/
-lenis: /r, l/
PLOSIVE:
-fortis: /pp, tt, kk/
-lenis: /p, t, k/
FRICATIVE:
-fortis: /ss, xx/
-lenis: /s, x/
APPROXIMANT: /v, D, j/
2) VOWELS
a) MONOPHTHONGS: /i, u, a/
b) DIPHTHONGS: /@U, OI, aI, aU/
B) ALLOPHONES
1) /N/ is realised as [g] in word-initial position and in onset position following a coda consonant.
2) All fortis obstruents are realised as lenis plosives in intervocalic position.
3) All lenis plosives and /s/ are voiced in intervocalic position.
4) /x/ is realised as [Z] in intervocalic position.
5) /D/ is realised as [T] in word-final position.
6) /i, u/ are realised as [E, O] when preceding a liquid coda or fortis liquid.
7) /i, u/ are reslised as [E, O] when adjacent to /v, j/.
C) ORTHOGRAPHY
"a" = /a/
"ai" = /aI/
"au" = /aU/
"g" = /N/
"gg" = /Ng/
"h" = /x/
"hh" = /xx/
"i" = /i/
"iu" = /@U/
"j" = /j/
"k" = /k/
"kk" = /kk/
"l" = /l/
"ll" = /ld/
"m" = /m/
"mm" = /mb/
"n" = /n/
"nn" = /nd/
"p" = /p/
"pp" = /pp/
"q" = /D/
"r" = /r/
"rr" = /rd/
"s" = /s/
"ss" = /ss/
"t" = /t/
"tt" = /tt/
"u" = /u/
"ui" = /OI/
"v" = /v/
Historical note: /D/ was originally written with "þ". However, when printing was introduced to Thule, no type was available for that letter so "q" was substituted instead.
The spelling of Thulean is generally phonemic except that the effects of consonant gradation and sandhi are indicated.
D) PROSODY
Thulean has fixed initial primary stress. Secondary stress falls on every alternate syllable following the primary stress. Rhythm type is trochaic.
E) PHONOTACTICS
1) The syllable template is (C)V(C).
2) Consonant clusters only occur at syllable boundaries.
3) Permitted consonant clusters:
a) /m/ + /p/
b) /n/ + /t, s/
c) /N/ + /k, x/
d) /r/ + /m, N, p, t, k, s, x, v, j/
e) /l/ + /m, N, p, t, k, s, x, v, j/
f) /p, t, k/ + /s/
g) /s/ + /p, t, k/
4) Permitted word-final codas: /m, n, t, k, s, D/
5) Vowels in hiatus do not occur.
6) Diphthongs may not occur in a closed syllable, before a fortis consonant or before /P, j/.
F) SANDHI
1) Sandhi occurs at morpheme boundaries as a result of suffixation.
2) The resulting sandhi transformations are as follows:
a) /m, n, N/ + /m, n, N/ > [mb, nd, Ng]
b) /m, n, N/ + /r/ > [rd, rd, rd]
c) /m, n, N/ + /l/ > [ld, ld, ld]
d) /m, n, N/ + /v, D, j/ > [mb, nd, Ng]
e) /m, N/ + /t, s/ > [nt, ns]
f) /m, n/ + /k, x/ > [Nk, Nx]
g) /n, N/ + /p/ > [mp]
h) /r, l/ + /n/ > [nd]
i) /r, l/ + /r, l/ > [rd, ld]
j) /r, l/ + /D/ > [rd, ld]
k) /p, t, k/ + /p, t, k/ > [pp, tt, kk]
l) /p, t, k, s/ + /v/ > [ps, ts, ks, ss]
m) /p, t, k, s/ + /D/ > [ps, ts, ks, ss]
n) /p, t, k, s/ + /j/ > [ps, ts, ks, ss]
G) CONSONANT GRADATION
1) This process affects fortis consonants and lenis plosives following a nasal, liquid or vowel.
2) It is triggered if the consonants in question form the onset of a closed syllable due to suffixation.
3) The gradation sequences are as follows:
a) /mp/ > /mb/ > /m/
b) /nt/ > /nd/ > /n/
c) /Nk/ > /Ng/ > /N/
d) /rp/ > /rP/
e) /rt/ > /rd/ > /r/
f) /rk/ > /rj/
g) /lp/ > /lP/
h) /lt/ > /ld/ > /ll/
i) /lk/ > /lj/
j) /pp/ > /p/ > /P/
k) /tt/ > /t/ > /D/
l) /kk/ > /k/ > /j/
m) /ss/ > /s/
n) /xx/ > /x/