LA has added comments, this time with date (LA051107) NEW VERSION OF THIS DOCUMENT, BECAUSE OF A NEW SUGGESTION TO SEM981! Here are som sentences with a verbal as SUBJ or as an attribute (@N< or @A<). SOURCE: text SME970 Dál lea Bireha vuorru geahččat sávzzaid. A1 A:adv('dál') Dál P:v('leat',IV,ind,pr,3sg) lea Cs:g <===== =H:g =D:prop('Biret',Fem,sg,gen) Bireha <===== ==D:n =H:n('vuorru',sg,nom) vuorru <===== ==H:n <===== =D:cl :v('geahččat',TV,inf,n<) geahččat <===== ==P:v(nfin) @-FOBJ:n('sávza',pl,acc) sávzzaid <===== ==O:n . ::: Fixed. SOURCE: text SME971 Áddjá lea čeahppi muitalit ságaid. A1 S:n('áddjá',sg,nom) Áddjá P:v('leat',IV,ind,pr,3sg) lea <===== Cs:g Cs:adj('čeahppi',sg,nom) čeahppi <===== =H:adj <===== =D:icl :v('muitalit',TV,inf,adj<) muitalit <===== ==P:v(nfin) @-FOBJ:n('sáhka',pl,acc) ságaid <===== ==Od:n . ::: Fixed. SOURCE: text SME972 Mus livččii miella mannat teáhteris. A1 A:pron('mun',,1sg,loc) Mus P:v('leat',IV,cond,pr,3sg) livččii <===== S:g S:n('miella',sg,nom) miella <===== =H:n <===== =D:cl :v('mannat',IV,inf,n<) mannat <===== ==P:v(nfin) @-FADVL:n('teáhter',sg,loc) teáhteris <===== ==A:n . ::: Fixed. SOURCE: text SME974 Gurssa ulbmil lea oahppat goarrut Ivgu gávtti. A1 S:g =D:n('gursa',sg,gen) Gurssa =H:n('ulbmil',sg,nom) ulbmil P:v('leat',IV,ind,pr,3sg) lea <===== Cs:cl Cs:v('oahppat',TV,inf) oahppat <===== =P:v(nfin) <===== =Od:cl @-FOBJ:v('goarrut',TV,inf) goarrut <===== ==P:v(vfin) @-FOBJ:g <===== ==Od:g =D:prop('Ivgu',plc,sg,gen) Ivgu <===== ===D:n =H:n('gákti',sg,acc) gávtti <===== ===H:n . ::: Fixed the Od:g. The analysis of goarrut is nowadays "goarrut" V TV Inf @ADVL, thus the group is now A:cl and the result is also otherwise different. LA051107: I have fixed the problem ("goarrut" can also be read as (N Pl Nom)). The analyze is: "" S:11124, 11124, 11124 "gursa" N Sg Gen S:3512 @>N "" S:4488, 4492, 11519, 16466 "ulbmil" N Sg Nom S:4230 @SUBJ "" S:16464 "leat" V IV Ind Prs Sg3 S:4010 @+FMAINV "" S:4488, 4492, 5862, 5862, 9222 "oahppat" V TV Inf S:3891 @SPRED "" S:4488, 4488, 4492, 4492, 4554, 4554, 9639 "goarrut" V TV Inf S:3908 @-FOBJ "" S:4492, 11004, 11281, 11281, 12067 "Ivgu" N Prop Plc Sg Gen S:3512 @>N "" S:10412 "gákti" N Sg Acc S:3916 @-FOBJ "<.>" "." CLB <<< And in VISL: SME1 Gurssa ulbmil lea oahppat goarrut Ivgu gávtti. A1 S:g =D:n('gursa',sg,gen) Gurssa =H:n('ulbmil',sg,nom) ulbmil P:v('leat',IV,ind,pr,3sg) lea Cs:icl =P:v(nfin)('oahppat',TV,inf) oahppat =Od:icl ==P:v(nfin)('goarrut',TV,inf) goarrut ==Od:g ===D:prop('Ivgu',plc,sg,gen) Ivgu ===H:n('gákti',sg,acc) gávtti . This looks good! I wanted to see it as a tree, but visl inlog didn´t work today.. SOURCE: text SME975 Megabyte lea dán áiggi dábálaš vuohki mihtidit sturrodaga. A1 S:n('megabyte',sg,nom) Megabyte P:v('leat',IV,ind,pr,3sg) lea A:g =D:pron('dát',,sg,gen) dán =H:n('áigi',sg,gen) áiggi Cs:g <===== Cs:g <===== =H:g =D:adj('dábálaš',attr) dábálaš <===== ==D:adj =H:n('vuohki',sg,nom) vuohki <===== ==H:n <===== =D:icl :v('mihtidit',TV,inf,n<) mihtidit <===== ==P:v(nfin) @-FOBJ:n('sturrodat',sg,acc) sturrodaga <===== ==Od:n . ::: Here there is a clash between the different modifiers. -FOBJ is now parsed, but otherwise the parse is not the same as you suggest. I have not implemented all the alternative parses so the >N dábálaš is at the moment interpreted as modifying a the whole "vuohki mihtidit sturrodat", even though it should modify only vuohki if I understood correctly.. LA051107: It seems for me that your analyse is correct. "dábálaš" modifies the whole icl. VISL 051107: 740 sets (using 9801 different tags) 517 mappings 29 constraint sections 3047 constraint rules SOURCE: text SME1 Megabyte lea dán áiggi dábálaš vuohki mihtidit sturrodaga. A1 S:n('megabyte',sg,nom) Megabyte P:v('leat',IV,ind,pr,3sg) lea A:g =D:pron('dát',,sg,gen) dán =H:n('áigi',sg,gen) áiggi Cs:g =D:adj('dábálaš',attr) dábálaš =H:g ==H:n('vuohki',sg,nom) vuohki ==D:icl ===P:v(nfin)('mihtidit',TV,inf) mihtidit ===Od:n('sturrodat',sg,acc) sturrodaga . SOURCE: text SME976 E-boasta lea hui álkes ja jođánis vuohki sáddet reivviid birra máilmmi. A1 S:n('e-#boasta',sg,nom) E-boasta P:v('leat',IV,ind,pr,3sg) lea Cs:g =D:g <===== ==D:par <===== ==CJT:g ==D:adv('hui') hui <===== ===D:adv ==H:adj('álki',attr) álkes <===== ===H:adj =SUB:conj-s('ja') ja <===== ==CC:conj =D:adj('jođán',attr) jođánis <===== ==CJT:a <===== =H:g =H:n('vuohki',sg,nom) vuohki <===== ==H:n <===== ==D:icl :v('sáddet',TV,inf,n<) sáddet <===== ===P:v(nfin) @-FOBJ:n('reive',pl,acc) reivviid <===== ===Od:n @-FADVL:g <===== ===A:g =H:prp-pre('birra') birra <===== ====H:p =D:n('máilbmi',sg,gen) máilmmi <===== ====D:n . ::: Still problems with coordination.. other problems as well. SOURCE: text SME978 Lea hávski oažžut reivve. A1 P:v('leat',IV,ind,pr,3sg) Lea Cs:adj('hávski',sg,nom) hávski <===== S:icl S:v('oažžut',TV,inf) oažžut <===== =P:v(nfin) @-FOBJ:n('reive',sg,acc) reivve <===== =Od:n . ::: Fixed SOURCE: text SME979 Mu mielas lea suohtas oahppat sámegiela. A1 A:g =D:pron('mun',,1sg,gen) Mu =H:n('miella',sg,loc) mielas P:v('leat',IV,ind,pr,3sg) lea Cs:adj('suohtas',sg,nom) suohtas <===== Cs:adj <===== S:icl S:v('oahppat',TV,inf) oahppat <===== =P:v(nfin) @-FOBJ:n('sáme#giella',sg,acc) sámegiela <===== =Od:n . ::: This does not have -FOBJ tag anymore. LA051107: It has (at least at the moment), and the VISL works well: "" S:11080 "mun" Pron Pers Sg1 Gen S:3512 @>N "" S:4488, 5929, 7864, 7864 "miella" N Sg Loc S:4375 @ADVL "" S:15613 "leat" V IV Ind Prs Sg3 S:4010 @+FMAINV "" S:4488, 4492, 9169, 11441, 11814, 15885 "suohtas" A Sg Nom S:4230 @SPRED "" S:4488, 4492, 5862, 5862, 9694 "oahppat" V TV Inf S:3873 @SUBJ "" S:12158, 16053 "sáme#giella" N Sg Acc S:4281 @-FOBJ "<.>" "." CLB <<< SOURCE: text SME1 Mu mielas lea suohtas oahppat sámegiela. A1 A:g =D:pron('mun',,1sg,gen) Mu =H:n('miella',sg,loc) mielas P:v('leat',IV,ind,pr,3sg) lea Cs:adj('suohtas',sg,nom) suohtas S:icl =P:v(nfin)('oahppat',TV,inf) oahppat =Od:n('sáme#giella',sg,acc) sámegiela . SOURCE: text SME982 Gean vuorru lea dál bálkestit birccu? A1 Cs:g =D:pron('gii',,sg,gen) Gean =H:n('vuorru',sg,nom) vuorru P:v('leat',IV,ind,pr,3sg) lea A:adv('dál') dál <===== S:icl S:v('bálkestit',TV,inf) bálkestit <===== =P:v(nfin) @-FOBJ:n('bircu',sg,acc) birccu <===== =Od:n ? ::: Fixed. This one is TRICKY because of the "riikkas riikii" (from land to land) which really is a A + A, in other words: Head + Head. It cannot be CJT:cl because there is no verbal? NEW SUGGESTION with acl (verb-elliptic clause) cfr. http://visl.sdu.dk/cafeteria.html SOURCE: text SME981 Mu mielas lea suohtas beassat johtit riikkas riikii ja báikkis báikái. A1 A:g =D:pron('mun',,1sg,gen) Mu =H:n('miella',sg,loc) mielas P:v('leat',IV,ind,pr,3sg) lea Cs:adj('suohtas',sg,nom) suohtas <===== S:icl <===== =P:g D:vaux:v('beassat',IV,inf) beassat <===== ==D:v(nfin) S:v('johtit',IV,inf) johtit <===== ==H:v(nfin) <===== =A:par <===== ==CJT:acl @-FADVL:n('riika',sg,loc) riikkas <===== ===A:n @-FADVL:par <===== remove =CJT:n('riika',sg,ill) riikii <===== ===A:n =SUB:conj-s('ja') ja <===== ==CO:conj <===== ==CJT:acl =CJT:n('báiki',sg,loc) báikkis <===== ===A:n @-FADVL:n('báiki',sg,ill) báikái <===== ===A:n . ::: Not fixed yet. SOURCE: text SME980 Dus ii leat lohpi dan dahkat! A1 A:pron('don',,2sg,loc) Dus P:g =D:vaux:v('ii',IV,neg,ind,3sg) ii =H:v('leat',IV,ind,pr,conneg) leat <===== S:g S:n('lohpi',sg,nom) lohpi <===== =H:n <===== =D:icl @-FOBJ:pron('dat',,3sg,acc) dan <===== ==Od:n :v('dahkat',TV,inf,n<) dahkat <===== ==P:v(nfin) ! ::: Fixed. SME1 Biret-guovttos Juffáin leaba čeahpit duddjot. A1 S:g =H:n('Biret-#guovttos',sg,nom) Biret-guovttos =D:prop('Juffá',Mal,sg,com) Juffáin P:v('leat',IV,ind,pr,3du) leaba <==== Cs:g Cs:adj('čeahppi',pl,nom) čeahpit <==== =H:adj :v('duddjot',TV,inf,adj<) duddjot <==== =D:v(nfin) . ::: Fixed.