! ========================================================================== ! ! North Sámi morphophonological rule set ! ! ========================================================================== ! ! twol-sme.txt ! The file contains the rule set for the non-segmental ! North Sámi morphphonological rules Alphabet a b c d e f g h i j k l m n o p q ! small r s t u v w x y z å ä ö ø æ %- á é ó ú í à è ò ù ì ë ü ï â ê ô û î ã ý þ ñ ð ß ª ß ç A B C D E F G H I J K L M N O P Q ! capital R S T U V W X Y Z Å Ä Ö Ø Æ Á É Ó Ú Í À È Ò Ù Ì Ë Ü Ï Â Ê Ô Û Î Ã Ý þ Ñ Ð e1:e i1:i o1:o o3:o u1:u ! Morphophonemes č đ ŋ š ŧ ž ! Sámi letters. In an UTF-8 version these will Č Đ Ŋ Š Ŧ Ž ! be replaced by the real Sámi letters. ':0 %/ ¤:0 '1:' #:0 ! ' for CnsGrad of the lg:lgg and l'l:ll type ! ¤:0 prevents ConsGrad in certain words ! '1 is the real apostroph ! # is used to mark both lexicalised and ! derived compounds ! ':0 %/:0 %^%/:%/ ¤:0 '1:' ! This line will be deleted after the dust has ! been settled. h1:h h2:h g2:g m2:m n2:n ! the x2 ones are consonants that alternate in H1:H H2:H G2:G M2:M N2:N ! stem-final positions. j3:j b3:b d3:d g3:g h3:h m3:m n3:n p3:p s3:s t3:t z3:z J3:J B3:B D3:D G3:G H3:H M3:M N3:N P3:P S3:S T3:T Z3:Z ! The x3 ones are consonants that never alternate. ! The capital J3 etc. do not work. X1:0 X2:0 X3:0 X4:0 X5:0 X6:0 X7:0 X8:0 X9:0 ! diacritics Q1:0 Q2:0 Q3:0 Q4:0 Q5:0 Q6:0 Q7:0 Q8:0 Q9:0 ! They trigger morphophono- Y1:0 Y2:0 Y3:0 Y4:0 Y5:0 Y6:0 Y7:0 Y8:0 Y9:0 ! logical rules W1:0 W2:0 ; Sets Vow = a á e e1 i i1 o o1 o3 u u1 y æ ø å ä ö A Á E I O U Y Æ Ø Å Ä Ö Å é ó ú í à è ò ù ì ë ü ï â ê ô û î ã ý É Ó Ú Í À È Ò Ù Ì Ë Ü Ï Â Ê Ô Û Î Ã Ý ; Cns = b b3 c č d đ d3 f g g2 g3 h h2 h3 j j3 k l m m2 m3 n ŋ n2 n3 B B3 C Č D Đ D3 F G G2 G3 H H2 H3 J J3 K L M M2 M3 N Ŋ N2 N3 p p3 r s š t ŧ v z ž z3 ' '1 P P3 R S Š T Ŧ V Z Ž Z3 ; StemCns = b b3 d d3 g g2 g3 h h2 h3 j j3 l m m2 m3 n n2 n3 r s š B B3 D D3 G G2 G3 H H2 H3 J J3 L M M2 M3 N N2 N3 R S Š ; WeG = X4 X5 X6 X8 X9 Q4 Q5 Q6 Q7 Q8 W1 ; Len = Y1 Y2 Y3 Y4 Y5 Y6 Y7 Y9 ¤ ; Dummy = X1 X2 X3 X4 X5 X6 X7 X8 X9 Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Y1 Y2 Y3 Y4 Y5 Y6 Y7 Y8 W1 W2 ¤ ; FinalDummy = X2 X4 X7 X8 ; NonMPDummy = X1 X4 X7 X8 ¤ ; NonMPDummy2 = X1 X4 ¤ ; HighVow = i u I U ; NonHighVow = a e o A E O ; Stop = k p t c č K P T C Č ; DelCns = g2 m2 n2 h2 G2 M2 N2 H2 ; CntrCns = s j h ž S J H Ž ; Hyph = %- ; Rules !! Remainder! !! Change all #: to (Hyph) #: in order to account for ealáhus- ja ... ! j rule ! ------ "Postvocalic j Surfacing" j:i <=> Vow: ( :0 ) ( Dummy: ) _ [ Cns | ': | ( :0 - Y5: ) (Hyph) #: ] ; ! áj'gi : áigi, mánnáX5jd : mánáid ! Added ( :0 ) because of olmmoš ! new version for Troandin-bisma ! All j:i alternation is conflated into one rule. LC is vowel and optional ! (suffixinitial) dummy symbol, RC is consonant, dummy or word boundary, ! The intention is to have the strict operator (<=>), but catch the ! variation with a lot of (=all) contextual alternatives. ! exchanging DummyNotY5 with :0 - Y5: , better luck this time. ! j3:j is the j that does not turn to i in this context. It was added for ! the sake of Budejju and Stálobajjávri, to avoid Budeiju and Stálobaijávri ! Word final consonant alternations ! --------------------------------- "Word Final Consonant Neutralization" !smirezit : smires, Troandin-bisma Cx:Cy <=> Vow: (( CntrCns:) Dummy:+) _ ( :0 - Y5: ) [ Hyph | #: ] ; where Cx in (z ž m h p d g b) Cy in (s š n t t t t t) matched ; "Deletion: Final h s j ž in Contracted Stems" ! fijdnisY5t:fiidnát Cx:0 <=> Vow _ [ Y5: | Y6: j: ] ; where Cx in (h s j ž) ; "Word Final Cluster Simplification 1" ! vuordildit : vuordil Cy:0 <=> Vow Cx _ (Hyph) #: ; where Cx in (l r) Cy in (d g) ; "Word Final Cluster Simplification 2" ! attestit : attes Cx:0 <=> Vow s _ (Hyph) #: ; where Cx in (t k m) ; "Word Final Cluster Simplification 3" ! beroštit : beroš t:0 <=> Vow š _ (Hyph) #: ; "Word Final Cluster Simplification 4" ! bearjadahka : bearjadat Cx:0 <=> Vow h: ':0 _ (Hyph) #: ; where Cx in (t k) ; ! Changing bearjadahk to bearjadah'k, adding ':0 "Word Final Cluster Simplification 5" ! bearjadahka : bearjadat h:t <=> Vow _ ':0 [t:0 | k:0] (Hyph) #: ; ! Splitting WFCF 4 into 2 to handle the inserted ':0 symbol. "Word Final Double Consonant Simplification" ! muitaluss : muitalus Cy:0 <=> Vow Cx: _ (Hyph) #: ; where Cx in (s š d ž) Cy in (s š d ž) matched ; ! Consonant gradation rules ! ------------------------- !Now that long double cons. have been analyzed having lexical triple consonants, !many of the grad. rules of the form ':Cx may be collapsed to a single rule !without left context: _ Cx Vow: WeG: ; where XXXX matched ; . This has not !been done yet (PS 29.11.1993). ! But the long double cons have not been analyzed as triple consonants! "Gradation: h Loss" ! johka : joga (h:0) h:0 <=> _ ': Cy: Vow ( StemCns:) (:StemCns) WeG: ; where Cy in (p t k c č) ; "Gradation: Prenasal Stops" ! sápmi : sámi (p:0) Cx:0 <=> _ Cy Vow ( StemCns:) (:StemCns) WeG: ; where Cx in (p t k) Cy in (m n ŋ) matched ; "Gradation: tnj" ! latnja : lanja (t:0) t:0 <=> _ n j Vow ( StemCns:) (:StemCns) WeG: ; "Gradation: dj" ! vuodjit:vuoján d:0 <=> Vow: _ ': j Vow ( StemCns:) (:StemCns) WeG: ; "Gradation: Double Consonant" ! káffe:káfes Cx:0 <=> Vow: _ Cy Vow ( StemCns:) (:StemCns) WeG: ; where Cx in (đ f l m n ŋ r s š ŧ v) Cy in (đ f l m n ŋ r s š ŧ v) matched ; "Gradation: šž" Cx:0 <=> Vow: _ Cy X2: Vow ( StemCns:) (:StemCns) ; ! njunnoša:njunnožis where Cx in (š) Cy in (ž) matched ; !"Gradation: Long Double Consonant" ! Cx:0 <=> Vow: Cy _ Cy Vow ( StemCns:) (:StemCns) WeG: ; ! where Cx in (l m n ŋ r s š v) ! Cy in (l m n ŋ r s š v) ! matched ; ! This rule was made for a lexicon with tálllaX4 : tálla. We have a ! tál'laX4 : tálla lexicon, hence the rule is removed. "Gradation: Preaspirated Geminates" ! áhkku:áhkus Cx:0 <=> h _ Cy Vow ( StemCns:) (:StemCns) WeG: ; where Cx in (k p t c č l r) Cy in (k p t c č l r) matched ; "Gradation: Jodded Double Consonants" ! viellja:vieljas Cx:0 <=> Vow: _ Cy j Vow ( StemCns:) (:StemCns) WeG: ; ! mannji:manjis where Cx in (l d n) ! áddjá:ádjás Cy in (l d n) matched ; "Gradation: Cluster Non-nasal Sonorant + Non-sonorant" ! láj'bi:láibbi ':Cx <=> Vow: [ l | r | đ | v | j: ] _ Cz Vow ( StemCns:) (:StemCns) WeG: ; where Cx in (f g d b k p t z ž c č s š) Cz in (f g d b k p t z ž c č s š) matched ; "Gradation: Cluster ŋ + Non-sonorant" ! seaŋ'ga:seaŋgga ':Cx <=> Vow: ŋ _ Cz Vow ( StemCns:) (:StemCns) WeG: ; ! boŋ'kit:boŋkkii where Cx in ( g k ) Cz in ( g k ) matched ; "Gradation: Cluster m + Non-sonorant" ! lám'pa:lámppas ':Cx <=> Vow: m _ Cz Vow ( StemCns:) (:StemCns) WeG: ; ! rán'ska:ránskkas where Cx in (f b p s š t) Cz in (f b p s š t) matched ; ! mt : mtt added bec. of ámta:ámttas, although not in grammar. Check norm! "Gradation: Cluster n + Non-sonorant" ! gán'da:gánddat ':Cx <=> Vow: n _ Cz Vow ( StemCns:) (:StemCns) WeG: ; where Cx in (d t z ž c č s š) Cz in (d t z ž c č s š) matched ; "Gradation: Cluster j/v + Liquid" ! áj'ru:áirrut ':Cx <=> Vow: [ j: | v ] _ Cz Vow ( StemCns:) (:StemCns) WeG: ; where Cx in (l r) Cz in (l r) matched ; "Gradation: Clusters lv, rj etc." ! ol'ju:oljjus ':Cx <=> Vow: [ đ | l | r ] _ Cz Vow ( StemCns:) (:StemCns) WeG: ; where Cx in (v j) Cz in (v j) matched ; "Gradation: Clusters tk, tm" ! mát'ki:mátkkis ':Cx <=> Vow: [ t | ŧ ] _ Cy Vow ( StemCns:) (:StemCns) WeG: ; where Cx in (k m) Cy in (k m) matched ; ! ŧ is missing in the grammar. example? "Gradation: Clusters ts, ks" ! 0:s <=> Vow: [ t | k ] _ s Vow ( StemCns:) (:StemCns) WeG: ; ! I cannot find any such clusters. "Gradation: Clusters sk, sm etc." ! beas'ka:beaskka ':Cx <=> Vow: [ s | š ] _ Cz: Vow ( StemCns:) (:StemCns) WeG: ; where Cx in (k p t l m v) Cz in (k p t l m v) matched ; "Gradation: Cluster vj" ! čoav'ji:čoavjji ':j <=> Vow: v _ j Vow ( StemCns:) (:StemCns) WeG: ; "Gradation: Cluster jv" ! beaj'vi:beaivvit ':v <=> Vow: j: _ v Vow ( StemCns:) (:StemCns) WeG: ; "Gradation: ihm, vhl etc." ! dujh'mi:duihmmit ':Cx <=> Vow: [ j: | v ] h _ Cy Vow ( StemCns:) (:StemCns) WeG: ; where Cx in (l m n) Cy in (l m n) matched ; "Gradation: Clusters lbm, jdn, vdn etc." ! čuolbma:čuolmma Cx:Cy <=> [ l | j: | v ] _ Cz Vow (Cns) ( StemCns:) (:StemCns) WeG: ; where Cx in (b d g ) Cy in (m n ŋ) Cz in (m n ŋ) matched ; ! "(Cns)" is added. I do not like it. But the rule as it was blocked the ! "olmmožin" essive of "olmmoš", since the "ž" was neither an underlying ! nor a surface stem consonant, I had to do something. The rule as it stands ! is meaningless, "(Cns)" contains both (StemCns:) and (:StemCns). ! Think through this. "Gradation: Clusters ldnj, vdnj etc." ! sávdnji:sávnnji d:n <=> [ l | v ] _ n j Vow ( StemCns:) (:StemCns) WeG: ; "Gradation: Clusters rbm, rdn rgŋ" ! čorbma:čorpma Cx:Cy <=> r _ Cz Vow ( StemCns:) (:StemCns) WeG: ; where Cx in (b d g) Cy in (p t k) Cz in (m n ŋ) matched ; "Gradation: Clusters rdnj" ! skurdnji:skurtnji Cx:Cy <=> r _ Cz j Vow ( StemCns:) (:StemCns) WeG: ; where Cx in (b d g) Cy in (p t k) Cz in (m n ŋ) matched ; "Gradation: ist, vsk etc." ! návs'tu:návsttu ':Cx <=> Vow: [ j: | v | l | r | m | n ] s _ Cy Vow ( StemCns:) (:StemCns) WeG: ; where Cx in (k t) Cy in (k t) matched ; "Gradation: mšk, ršk" ! limš'ki:limškki ':k <=> Vow: [ m | r ] š _ k Vow ( StemCns:) (:StemCns) WeG: ; "Gradation: Clusters kc ks etc. 1" ! uk'sa:uvssa k:v <=> Vow: _ ': Cx Vow ( StemCns:) (:StemCns) WeG: ; where Cx in (t c č s š) ; "Gradation: Clusters kc ks etc. 2" ! uk'sa:uvssa ':Cx <=> Vow: k: _ Cy Vow ( StemCns:) (:StemCns) WeG: ; where Cx in (t c č s š) Cy in (t c č s š) matched ; "Gradation: Cluster kst 1" ! teaks'ta:teavstta k:v <=> Vow: _ s ': t Vow ( StemCns:) (:StemCns) WeG: ; "Gradation: Cluster kst 2" ! teaks'ta:teavstta ':t <=> Vow k: s _ t Vow: ( StemCns:) (:StemCns) WeG: ; "Gradation: bb dd etc.1" ! spábba:spáppat Cx:Cy <=> Vow: _ Cz: Vow ( StemCns:) (:StemCns) WeG: ; where Cx in (b d g z ž) Cy in (p t k c č) Cz in (b d g z ž) matched ; "Gradation: bb dd etc.2" ! spábba:spáppat Cx:Cy <=> :Cy _ Vow ; where Cx in (b d g z ž) Cy in (p t k c č) matched ; "Gradation: dn to tn, etc." ! eadni:eatni Cx:Cy <=> Vow: _ Cz: Vow ( StemCns:) (:StemCns) WeG: ; where Cx in (b d g ) Cy in (p t k ) Cz in (m n ŋ) matched ; "Gradation: dnj to tnj" ! boadnji:boatnji d:t <=> Vow: _ n j Vow ( StemCns:) (:StemCns) WeG: ; ! Lengthening of consonants ! ------------------------- ! This takes place in certain verb forms, e.g. imperative and certain ! nominalisation forms. "Lengthening of Central Consonants hk etc." ! boah'ti:boahtti ':Cy <=> h: _ Cz Vow ( CntrCns:) Len: ; where Cy in (p t k c č) Cz in (p t k c č) matched ; "Lengthening of Central Consonants pm etc." ! dapmaY1:dabmi Cx:Cy <=> Vow: _ Cz Vow ( CntrCns:) Len: ; where Cx in (p t k ) Cy in (b d g ) Cz in (m n ŋ) matched ; "Lengthening of Central Consonants tnj" ! bitnju:bidnju t:d <=> Vow: _ n j Vow ( CntrCns:) Len: ; "Lengthening of Central Consonants dj" ! dad'ja:daddji ':d <=> Vow: d _ j Vow ( CntrCns:) Len: ; ! Lenition ! -------- "Lenition: Intervocalic Stops and Affricates" ! joh'kaX4:jo00ga Cx:Cy <=> h:0 ':0 _ Vow ; where Cx in (k p c č t) Cy in (g b z ž đ) matched ; ! Here, the k is reduced to g as an automatic result of the h deletion in the ! "Gradation: h loss" rule. ! Vowel Shortening ! ---------------- "Vowel Shortening in Bisyllabic Sg2 Imperatives and Connegatives" Vx:Vy <=> Cns _ Q6: ; where Vx in (u i) ! čuor'vu:čuorvo Vy in (o e) ! boah'ti:boađe matched ; "Vowel Shortening in Vowel-Final Compounds after Long 1st Syllable 1" Vx:Vy <=> [ Vow Vow | á ] Cns:+ Cns: _ X7: ; where Vx in (u i) ! várregirku Vy in (o e) ! < várri + girku matched ; ! This rule was difficult. Look at *)footnote 1 below for comments. "Optional Vowel Shortening in Cns-Final Compounds after Long 1st Syllable 1" Vx:Vy => [ Vow Vow | á ] Cns:+ Cns: _ StemCns:+ X7: ; where Vx in (u i) ! example? Vy in (o e) matched ; ! This is for the vuovdin# and vuovden#, if vuovden# exists. ! If vuovden# etc. is not possible, this rule should be removed. "Vowel Shortening in Compounds of Contract Stems" Vx:Vy <=> [ Vow Vow | á ] Cns:+ Cns: _ StemCns:* X4: X7: ; where Vx in (u i) Vy in (o e) matched ; ! To account for boazoviessu, the u:o alternation does not work above ! The reason is that we have boa'hcuX4viessu, not X4X7, as required here. "Allegro Forms in Compounds" ! gáv'piX7:gáv0pe Vx:Vy <=> Cns:+ _ ¤: X7: ; ! bassi¤X7:basse where Vx in (i u á ) Vy in (e o a ) matched ; ! Overlapping => constraint wrt. 'i:e', Optional Vow Shortening ! After short 1st Syllable ! The ¤: condition was added in order to account for VIVVA nouns. ! First, it was added as optional, with parentheses, but that did not ! work, for some strange reason. Then the parentheses were removed, and to our ! surprise both forms with and without ¤ worked. "Optional Vowel Shortening after Short 1st Syllable" ! Vx:Vy => [ (Hyph) #: | Cns ] [a|i|o|u] [ Cns:+ Cns | h: ': Cns: ] Vx:Vy => [ (Hyph) #: | Cns ] [a|i|o|u] [ Cns Cns:+ | h: ': Cns: ] _ (StemCns:*) [ X7: | X8: ] ; where Vx in (u i) Vy in (o e) matched ; ! Added + to Cns:+ and made StmCns: optional in LC, for biv'du:bivdo "Optional Shortening of á in Compounds after Long 1st Syllable 1" á:a => [ Vow Vow | á ] Cns:+ Cns _ StemCns:* [ X7: | X8: ] ; "Optional Shortening of á in Compounds after Long 1st Syllable 2" á:a => [ (Hyph) #: | Cns ] Vz Cns+ Cns: Cns+ _ StemCns:* [ X7: | X8: ] ; where Vz in (a i o u) ; "Vowel lowering for buorre" Vx:Vy <=> [ Vow Vow | á ] Cns:+ Cns _ X8: ; where Vx in (u i) Vy in (o e) matched ; ! This rule was actually made for one word only, buorre:buoret ! It replaces the missing "| X8:" sequence that is commented out in the ! "Vowel shortening ... 1" rule above. Why X8: does not work there is unclear. ! This rule should in any case "Compulsory Vowel Shortening after ij" ! liigi : liigeruđaid Vx:Vy <=> [ (Hyph) #: | Cns ] i j:i ': Cns _ [ X7: | X8: ] ; where Vx in (u i) Vy in (o e) matched ; ! The rule is too special, but we do not have illu : illosáhka, ! so not every cns after i will do. ! j: changed to j:i in order to allow biv'du: biv'do# ! but biv'do# is still blocked. WHY??!! "Optional Vowel Shortening in Caritives" Vx:Vy => [ (Hyph) #: | Cns ] [a|i|o|u] Cns:+ _ ( DelCns:) X9: ; where Vx in (u i) Vy in (o e) matched ; ! Hmm. X9 is used only in the DenominalAdjsV lexicon., in front of ! heapmi, huvva, huhtti, hudda. "Vowel Shortening in Caritives" Vx:Vy <=> [ Vow Vow | á ] Cns:+ _ ( DelCns:) X9: ; where Vx in (u i) Vy in (o e) matched ; "Vowel Shortening in Attributive Forms, V-final Adjectives" ! láj'kiW1s:láikkes Vx:Vy <=> Cns:+ _ [ W1: | W2: ] s (Hyph) #: ; ! ál'kiW2s:álkes where Vx in (u i) Vy in (o e) matched ; ! W1 for WeG Attr, W2 for non-CG Attr form. "Vowel Shortening in Attributive Forms, s-final Adjectives" ! váttisW1:váttes Vx:Vy <=> Cns:+ _ s W1: (Hyph) #: ; where Vx in (u i) Vy in (o e) matched ; ! Tentative rule, but works fine. ! Diphtong simplification ! ----------------------- "Diphthong Simplification in i-Stems before Suffixes Beginning with j:" Vx:0 <=> Vow _ Cns:+ [ i: | e1: ] X5: ; where Vx in (e o a) ; ! goah'tiX5jd:go0điid "Diphthong Simplification in Actor Nouns from i- and a-Stems (Optional)" Vx:0 => Vow _ Cns:+ [ i: | a: ] ¤: X5: ; where Vx in (e o a) ; "Diphthong Simplification in Certain Forms from u-Stems" Vx:0 <=> Vow _ Cns:+ [ u: | o1: ] [ X2: | X3: | (DelCns:) X6: | Q8: | Y5: | Y6: ] ; where Vx in (e o a) ; "Diphthong Simplification in Actor Nouns from u-Verbs (Optional)" Vx:0 => Vow _ Cns:+ u: ¤: [ X2: | X3: | X6: ] ; where Vx in (e o a) ; "Diphthong Simplification in Diminutives etc. from Trisyllabics with 2nd Syll u/o" Vx:0 <=> Vow _ Cns:+ [ u: | o1: ] ( NonMPDummy:) Cns:+ X2: ; where Vx in (e o a) ; "Diphthong Simplification: Conditional in -alla- etc. Derivatives from Trisyllabics" Vx:0 <=> Vow _ Cns:+ u:o ( NonMPDummy:) Cns:+ X1: ; where Vx in (e o a) ; "Diphthong Simplification in IllSg and Certain Conjug. Forms of a-Stems" Vx:0 <=> Vow _ Cns:+ a: [ X3: | Q2: | Q3: | Y7: ] ; where Vx in (e o a) ; "Diphthong Simplification in Certain Conjugational Forms of i- and u-Stems" Vx:0 <=> Vow _ Cns:+ HighVow: [ Q2: | Q3: | Q7: | Y7: ] ; where Vx in (e o a) ; "Diphthong Simplification: Optional before Deverbal -alla-, -adda-, -us" Vx:0 => Vow _ Cns:+ o1: ( NonMPDummy:) Cns:+ X1: ; where Vx in (e o a) ; "Diphthong Simplification in u-Stems before Conditional Marker" Vx:0 => Vow _ Cns:+ u: Q5: ; where Vx in (a e o) ; "Dipthtong simplification in the attribute forms of AD-adjectives" Vx:0 <=> Vow _ Cns:+ Y9: ; where Vx in (a e o) ; ! Vowel Rising ! ------------ "Vowel Rising" Vx:Vy <=> Vow: Cns:+ _ ( CntrCns:0 ) ( Dummy:+) j:i ; where Vx in (e o) Vy in (i u) matched ; ! Metaphony ! --------- "General Metaphony" Vx:Vy <=> Cns _ Cns:* [ X2: | X6: ] ; where Vx in (i e1 u) Vy in (á á o) matched ; "Metaphony in Derivatives -us, -alla, -adda 1" u:o <=> Cns _ ( NonMPDummy2:) Cns:+ X1: ; ! Here I have a new set, NonMPDummy2, without X7 and X8, to avoid clash with ! the allegro shortening rule. Check for alternative solutions !!! "Metaphony in Derivatives -us, -alla, -adda 2" Vx:á <=> Cns _ ( NonMPDummy2:) Cns:+ X1: ; where Vx in (i e1) ; "Illative Singular Metaphony" Vx:Vy <=> Cns _ X3: ; where Vx in (i a e1) Vy in (á i á ) matched ; "Metaphony in Actor Nouns of i-, a-, and u-Verbs" Vx:Vy <=> Cns _ ¤: [ X2: | X3: | X6: ] ; where Vx in (i a u) Vy in (á á o) matched ; "Metaphony in MUSH-Derivatives" i:á => Cns _ m: u: š: š: ; "Potential Marker Metaphony" Vx:Vy <=> Cns _ Q7: ; where Vx in (i u) Vy in (e o) matched ; ! Stem vowel alternations ! ----------------------- "Stem Vowel in Contracted Stems" ! fijdnisY5t:fiidnát Vx:Vy <=> Cns _ ( CntrCns:) Y5: ; where Vx in (i a u) Vy in (á á o) matched ; "Stem Vowel before Frequentative d" ! oažžut:oččodit Vx:Vy <=> Cns _ Q8: ; ! coahkut:coagostit where Vx in (i u) Vy in (a o) matched ; ! Note that this rule also triggers other contexts than frequentative -d- !"Stem Vowel before Frequentative l" ! i:a <=> Cns _ Q9: ; ! This rule is not implemented, there is no Q9 in the sme-lex.txt file (yet). !"Stem Vowel before Reflexive s" ! Vx:Vy <=> Cns _ Y8: ; ! where Vx in (i u) ! Vy in (á a) ! matched ; ! This rule is not implemented, there is no Y8 in the sme-lex.txt file (yet). "Indicative Present Singular 3rd Final Vowel" ! jearraQ1:jearrá Vx:á <=> Cns _ Q1: ; ! boahtiQ1:boahtá where Vx in (a i) ; "Imperative Present 1st/3rd Person and Connegative Final Vowel" Vx:o <=> Cns _ Q3: ; ! jearraQ3n:jerron where Vx in (a i u) ; ! boahtiQ3n:bohton ! čuorvuQ3n:čurvon "Stem Vowel Shift a:e, i:e, u:o" ! jearra:jerre Vx:Vy <=> Cns _ Q2: ; ! boahtit:bohtet where Vx in (a i u) ! čuorvu:čurvo Vy in (e e o) matched ; ! This is for Prs Du1, Pl3, and 2nd Person Imperative "Stem Vowel in 1st and 2nd Person Singular Present" ! boah'ti:boađán i:á <=> Cns _ Q4: ; "Stem Vowel before Conditional Marker" ! boahtit:boađálin i:á <=> Cns _ Q5: ; ! Vx:Vy <=> Cns _ Q5: ; ! The u:o alternation ! where Vx in (u i) ! removed because it gave ! Vy in (o á) ! *čuorvvolin, but Nickel ! matched ; ! gives čuorvvulin. "Stem Vowel in Dual 2nd Person Imperative" ! jearra:jearri a:i <=> Cns _ Y1: ; "Stem Vowel in Present Participials from i-Verbs" a:i <=> Cns _ ¤: [ X1: | X4: | X5: | X7: | X8: | (Hyph) #: | s | n ] ; ! a:i <=> Cns _ ¤: [ X1: | X4: | X5: | X7: | X8: | (Hyph) #: ] ; ! The problem with the old one was that it didn't trigger for Essive and for ! the Px forms of the nominative. ! The new one isn't perfect, the s and n are rather quasi for Ess and Px. "Stem Vowel in Dual 1st Person Imperative" ! jearra:jearru Vx:u <=> Cns _ Y2: ; ! boah'ti:boahttu where Vx in (a i) ; "Stem Vowel in Plural 2nd Person Imperative" ! jearra:jerret ! Vx:Vy <=> Cns _ Q2: ; ! boah'ti:bohtet Vx:Vy <=> Cns _ Y3: ; ! čuor'vu:čurvot where Vx in (a i u) Vy in (e e o) matched ; ! Q2 context eplaced by other rule for both imperative and other categories "Stem Vowel in Imperative Plural 1st Person and Passive Stem" Vx:o <=> Cns _ [ Y4: | Y7: ] ; ! jearra:jerrot where Vx in (a i u) matched ; ! boah'ti:bohtot ! čuor'vu:čurvot ! Special word final consonants ! ----------------------------- "Word Final Deletion of n2 m2 g2 h2" ! beatnag2:beana Cx:0 <=> Vow _ [ X4: | X8: | X6: Cns: | X9: ] ; ! luopmán2:luomi where Cx in (DelCns) matched ; ! giellum2:gielu "Word Final Neutralization of g2, h2, m2" ! giellum2:gielun Cx:Cy <=> Vow _ [ X2: | X7: ] (Hyph) #: ; where Cx in (g2 h2 m2) Cy in (t t n ) matched ; ! Orthographical rules ! -------------------- !"Downcasing of adjectives made of proper nouns" ! Xx:Xy <=> #: _ :* *: l a ž: ž: ; ! where Xx in ( 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 Ø Æ ! Á É Ó Ú Í À È Ò Ù Ì Ë Ü Ï Â Ê Ô Û Î Ã Ñ ) ! Xy in ( 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 ø æ ! á é ó ú í à è ò ù ì ë ü ï â ê ô û î ã ñ ) ! matched ; ! Oslo -> oslolaš ! Kárášjohka -> kárášjohkalaš ! Appendix: Deleted rules that still need attention ! ------------------------------------------------- ! *) footnote 1 ! Older, removed versions: ! Vx:Vy <=> [ Vow Vow | á ] Cns:+ Cns: _ StemCns:* X7: ; ! This one gave vuovden# as only possible form of vuovdin in compounds. ! Vx:Vy <=> [ Vow Vow | á ] Cns:+ Cns _ StemCns:* [ X7: | X8: ] ; ! the problem with this one was that it was not restricted to compounds !! ! The X7: | X8: line was commented out since no lexicon in sme-lex.txt ! pointed from X8 to R. This move should be checked for compounds later. !"Vowel Shortening in Compounds after Long 1st Syllable 2" ! Vx:Vy <=> [ (Hyph) #: | Cns ] Vz Cns+ Cns: Cns+ _ StemCns:* [ X7: | X8: ] ; ! where Vz in (a i o u) ; ! where Vx in (u i) ! Vy in (o e) ! matched ; ! XXX Remainder! This rule is commented out since including it ! spoils the i:á alternation of verbs (boahtit:boađán, for some reason. ! This should be looked into.