Meänkieli twolc file

This file documents the Meänkieli twolc file (the file governing gradation, gemination, vowel harmony and other morphophonological processes).

The first part of the file contains definitions, the second part contains rules.

Declaring the alphabet, sets and definitions

Alphabet

This defines all symbols (letters, archiphonemes, triggers) to be used.

• a b c d e f g h i j k l m n o p q r s t u v w x y z å ä ö æ ø = the letters
• %^WG:0 = weak grade
• %^E2I:0 = kiele- > kieli
• %^HMETA:0 = vow metathesis for ill (sauna+haan > saunhaan)
• %^HMET2:0 = vow metathesis for sauhnaan etc (sauna+haan > sauhnaan)
• %^AE:0 = a to e in otta- > otethaan
• %^IDEL:0 = trigger to delete i
• %> = suffix boundary
• h2:h = clitic hAAn
• i2:i = plural of nouns
• i3:i = past tense of verbs
• i4:i = i in conditional of isi
• i5:i = superlative i of adjectives
• i9:i = plural-i of adjectives (kamalissa)
• i8:i = past tense of verbs that drops in past antaa : anto
• p2:p t2:t k2:k = always p t k !mettä > mettän etc
• t3:t = t participating in gradation, but not in t:s !t3 not in use?
• k4:k = k to j in certain words
• t4:t goes to 0 (ie rt goes to r) in imartelee : imarella
• n9:n nouns that keep nn; juvanne
• %^A2O:0 = päärynä > päärynöitä
• %^A:ä %^O:ö %^U:y vowel harmony archiphonemes
• %^V:a %^V:e %^V:o %^V:u %^V:y %^V:i %^V:ä %^V:ö = for vowel lenghtening
• %^N:n for participle -nut
• ’ 0027 aphstrophe for saa’a
• ’ 2019 right single quot mark for saa’a
• ˊ 02ca letter prime for saaˊa

Sets

Here we group the symbols in convenient sets.

• Dummy = %+ %^WG %^E2I %^HMETA %^HMET2 %^VDEL %^EDEL %^AE %^AO %^¤ %^IDEL %^A2O ;
• DummyBorder = Dummy %> ;
• ArchiVowel = %^A %^O %^U ;
• SomeVowel = %^V ;
• NeutralVowel = e i i2 i3 i4 i5 i9 i8 E I ;
• FrontVowel = e i y ä ö ü æ ø E I Y Ä Ö Ü Æ Ø ;
• BackVowel = a o u å A O U Å ;
• RndVow = y ö o u å Y Ö O U Å ;
• UnRndVow = e i ä a E I Ä A ;
• VowelNotUY = e i ä ö a o å ü æ ø E I Ä Ö A O Å Ü Æ Ø ;
• VowNotI = e y ä ö a o u å ü æ ø E Y Ä Ö A O U Å Ü Æ Ø ;
• VowNotIU = e y ä ö a o å ü æ ø E Y Ä Ö A O Å Ü Æ Ø ;
• VowelNotEI = y ä ö a o u å ü æ ø Y Ä Ö A O U Ü Ä Ö ;
• Vow = FrontVowel BackVowel ArchiVowel SomeVowel ;
• ArchiCns = %^N ;
• LNRM = l n r m ;
• SurfaceCns = b c d đ f g h j k l m n p q r s š t v w x z ;
• Cns = SurfaceCns ArchiCns p2 t2 t3 t4 k2 k4 ’ ; testing with ’ 2019 for saa’a.
• Segment = Vowel Cns ;
• NonFront = BackVowel ArchiVowel ArchiCns SomeVowel Cns NeutralVowel Dummy ;
• NonLNR = b c d đ f g h j k m p q s š t v w x z ;

Definitions

This defines strings used often in rules.

WeakGrade = ([l|n|r]) (%^AE:) %^WG:

Rules

This chapter gives the rules themselves.

Consonant rules

For the gradation rules, each consonant deletion or change is given its own rule. Thus, both kk:k and k:0 are handled in the same *k:0 rule. This to avoid rule conflicts. The change rules (k:g, k:j etc.) are restricted by context (k:g only after n, etc.).

f rules

RULE: f:0

j rules

RULE: j:0

k rules

RULE: k:g

Tests:

RULE: k:0

Tests:

RULE: k:j

RULE: k4:j

Tests:

RULE: k:v

Tests:

l rules

RULE: k:v

m rules

RULE: m:0

n rules

RULE: n:0

p rules

RULE: p:0

Tests:

RULE: p:v

Tests:

RULE: p:m

r rules

RULE: p:m

s rules

RULE: r:0

t rules

RULE: t:j

RULE: t4:0 where t4 is t in rt that shall not become rr

RULE: t4:s where t4 becomes s in rt that shall not become rr

Tests:

RULE: t:0

Tests:

RULE: t:s

Tests:

**RULE: t:l ** for lt:ll

Tests:

**RULE: t:n ** for nt:nn

Tests:

**RULE: t:r ** for rt:rr

Tests:

Tests:

v rules

RULE: v:0

Gemination rules

The gemination rules insert the geminated consonant (thus 0:h if h to the left). There is one subrule for each vowel context, in order to avoid confilcts.

RULE: Gemination 0:h

RULE: Gemination 0:j

RULE: Gemination 0:k

Tests:

RULE: Gemination 0:l

Tests:

RULE: Gemination 0:m

RULE: Gemination 0:n

RULE: Gemination 0:p

RULE: Gemination 0:s

Tests:

RULE: h:0

RULE: h:0

RULE: h:0

kasva>hm^A^An kasva>mhaan

saarna>^A>hm^A^An saarna>a>hmaan

tule>hm^A^An tule>mhaan

RULE: Gemination 0:t

Tests:

RULE: Gemination 0:v Tests:

Assimilation rules

These are assimilation rules for n on suffix borders of LNRS consonant stems. There is also a rule j:0 avoiding a lji sequence.

RULE: Alveolar assimilation for consonant stem l

Tests:

RULE: Alveolar assimilation for consonant stem r

RULE: Alveolar assimilation for consonant stem s in infinitives Tests:

RULE: Alveolar assimilation for consonant stem s in participles

Vowel change rules: a - ä - e - i - o - ö - u - y

Here come the rules for stem vowel changes in front of suffix -i- (be it plural, present, comparative or conditional). Vowels are deleted or changed according to context. There are also some other vowel change rules.

a rules

RULE: a:e before the ^AE trigger

RULE: a:0 before metathesis h

Tests:

RULE: a:o when nonrounded root vowel and before i

Tests:

ä rules

RULE: ä:0

Tests:

RULE: ä:e

e rules

RULE: e:0 deletes -e- in LNR stems as well as before -i-

Tests:

RULE: e:i

Tests:

RULE: ä:ö

Tests:

i rules

RULE: i:0

Tests:

RULE: i:j

RULE: i2:j

RULE: i8:0

Tests:

RULE: i:e

o rules

RULE: o:0

Tests:

ö rules

RULE: ö:0

Tests:

u rules

RULE: u:0

Tests:

y rules

RULE: y:0

Tests:

Vowel copying rules

These are the rules connected to the Meänkieli -h- suffixes. The vowel must be copied from the stem to the right of the h and also deleted in the stem (cf. talo : talhoon)

RULE: a copying for h metathesis

Tests:

RULE: o copying for h metathesis

Tests:

RULE: i copying for h metathesis

Tests:

RULE: ä copying for h metathesis

RULE: e copying for h metathesis

RULE: ö copying for h metathesis

RULE: y copying for h metathesis

RULE: u copying for h metathesis

Vowel harmony rule

All vowel harmony is taken care of with one rule.

RULE: Back harmony

Tests:

• ★katto^WG (is not standard language)
• ★katto0 (is not standard language)

---

This (part of) documentation was generated from src/fst/morphology/phonology.twolc