%***************************************************************************** % % % A Link Grammar Parser Dictionary for Persian % % (C) Jon Dehdari, 2004-2005 % % This file is distributed under the terms of the General Public License v.2 % % %***************************************************************************** % Dictionary version number is 5.0.1 (formatted as V5v0v1+) : V5v0v1+; % PERSIAN SYNTAX NOTES: % -Normal Order: (S) (PP) (O) V % -If Object if Definite: (S) (O & rA) (PP) V % -Marked categories in a phrase typically move to % the front, allowing pretty free word order % afterwards. Just make sure the grammatical % markers (rA, be, dar, az, ...) are present in % this situation. % COMPOUND VERB NOTES: % jAru: broom.n % zadan: hit.v % jAru zadan: sweep.v % jAru xub mi zad: "he/she used to sweep well" % jAru rA zad: "he/she hit the broom" % (see link 'K' for similar concept) % % EZAFE NOTES: -nouns can take ez. link afterwards and beforehand % -many preps (Geir, qAbel, sar, zir) must take ez. link afterwards, % -arabic preps (Geir, qAbel) can take ez. link beforehand, % -persian preps cannot take ez. link beforehand, % -pronouns cannot take ez. link afterwards, % -adj. and pronouns after nouns and prep. must form an ez. link % -adj. must take ez. link beforehand. % LGPARSER NOTE: The more to the left something is (ceteris % paribus) the closer it will link. This is important % for the SOV nature of Persian. Thus links from a verb: % {O-} & {S-}; % This makes it SOV % {S-} & {O-}; % And this makes it OSV % TODO: % q's (non y/n), % past participle verbs to form adj's % verbless things like 'fqirm' and ?'bzrg |m'? % complementizers for NP's 'the book that is green' % (situation: 'kh |gr dr |irAn hst m, verb', and other "if" situations at both the beginning and end of a sentence % more time and number situations % 0. Definitions % I. Nouns % II. Pronouns % III. Determiners % IV. Verbs % A. Normal Verbs % B. Intransitive Verbs % C. Compound Verbs % D. Auxiliary Verbs % E. Subordinating Verbs % F. Passive Verbs % G. Future Verbs % H. Verb Prefixes % I. Verb Suffixes % V. Prepositions and Case Particles % VI. Adjectives % VII. Adverbs % VIII. Other Grammatical Words % 0. Definitions % AV Adverb [CO|E|MV] % AJM Adjective Morphological Suffix (comparative|superlative) [?|L] % C Complementizer (aka Subordinate Clauses) [C] % CC Conjunction [CC] % CCcc Conjunction for conjunctions % CCAM Conjunction for attributive adjectives % CCAJM Conjunction for attributive adjectives, top level % CCAMr Conjunction for attributive adjectives on the Right side % CCAML Conjunction for attributive adjectives on the Left side % CCAP Conjunction for predicate adjectives % CCAJP Conjunction for predicate adjectives, top level % CCAPR Conjunction for predicate adjectives on the Right side % CCAPl Conjunction for predicate adjectives on the Left side % CCF Conjunction for CP's on the Left side % CCO Conjunction for Object (Accusative) phrases (both nouns and pronouns) % CCOR Conjunction for Acc. words on the Right side % CCOl Conjunction for Acc. words on the left side % CCOB Conjunction for Acc. phrases, top level % CCN Conjunction for Noun phrases, nominative % CCNR Conjunction for Nouns on the Right side % CCNl Conjunction for Nouns on the Left side % CCP Conjunction for Prepositional phrases % CCPP Conjunction for Prep phrases, top level % CCPr Conj. for objects in a PP on the Right side % CCPL Conj. for objects in a PP on the Left side % CCS Conjunction for Subject (nom), top level % COMMA Connects a final comma with the following conjunction (eg. Bob, Sue, and Bill) % D Determiner (eg. yek-ketAb) [D] % EZ Ezafe (links Noun and M (noun|adj|pronoun)) % I Verb Future [I] % IK Light Verbs separated by linearity violations from non-verbal element (in future tense - eg. man dast xuAham zad) % K Kompound verbs [K|ID?] % M Ezafe Construct or Genitive Pronominal Clitic (noun|adj|pronoun) [M|A]? (eg. ketAb e mA or ketAb-emAn) % NMS Noun Morphological Suffix - plural or indefinite (eg. ketAb-hA) % O Object (acc) [O] % On Object noun % Opn Object pronouns % ORDM Ordinal number suffix morpheme % P Predicate of stative verbs, for predicate adjectives, nouns [M] (eg. in ketAb ast; in bozorg ast) % PA Postposition Accusative (rA) % PO Preposition Object [J] % PP Preposition [MVp] % REL Relative clause [R] (eg. the book that I wrote is here) (note. using 'R' messes up stuff) % S Subject (nom) [S] % Sn Subject noun % Spn Subject pronouns (1st person,2nd... Plural, Singular) % SUB Subordinating word (eg. keh(that), AyA(y/n)) [TH] % SUBP Subordinating Phrase, followed by 'keh' (eg. vaqti keh(when), barAye in keh(so that)) % VM Verb Morpheme (prefixes: mi(DUR),be(IMPER or SUBJ),na(NEG)) [Pg| ] % VMO Verb Morpheme Object suffix (eg. didam-eC) [O] % VMP Verb Morpheme Person suffix (eg. did-am) % VMPK Verb Morpheme Person suffix for split complex predicates % VMPP Verb Morpheme Past Participle (eg. kard-eh) % VMT Verb Morpheme Tense suffix (d|t) % VPAS Verb Passive (Become + Past Participle) (eg. kardeh Codan) [Pv] % VPPP Verb Past Participle Person (eg. kardeh-am) % Wcc left Wall for sentences with coordinating conjunction % Wd left Wall for Declarative sentences with explicit subject % Wi left Wall for Imperative sentences, or any other sentence without explicit subject % Xp Sentence % I. Nouns : Sns+; : Sn3p+; : [On+]; : [Pn+]; : NMSp+; : NMSi+; : ( NMSp+ or NMSi+ ); : PA+; : PO-; : REL+; : D-; : DI+; : (EZ+ or M+); : M-; : (CCNR- or CCNl+); : (CCOR- or CCOl+); : (CCPr- or CCPL+); : CCAJM+; : {Wd-}; /fa/words/words.n /fa/words/words.proper UNKNOWN-WORD.n : % Nouns, normal ( % Nouns in prepositional phrases {} & {} & {( or )} & {} & {} & ( or ) ) or ( % Nouns in complement position of an ezAfe construct {} & [] % Square Brackets are ugly! Please get rid of me. & {( or )} ) or ( % Plural Nouns in nominal case & {( or )} & {} & {} & ( or ( & ) ) ) or ( % Singular Nouns in nominal case {} & {( or )} & {} & {} & ( or ( & ) ) ) or ( % Nouns in accusative case {} & {( or )} & {} & (( {} & {} & ( or ) ) or ( ) ) ) ; %({NMS+} & S+) or ({NMS+} & O+); % Similar to above stuff "hA.nms" "An.nms" "gAn.nms" "At.nms": % Noun morphemes, plural suffix NMSp-; %NMSp- & {NMSi+} & Spn3p+; % experimental "i.nms" "|i.nms": % Noun morphemes, indefinite suffix NMSi-; /fa/words/words.n.compound /fa/words/words.adj.compound UNKNOWN-WORD.nk: % Noun component of compound verbs %? ( % Rule out non-compounding possiblities, like 'yek jAru zad' or 'jAru rA zad' % % & % ) %or ( % Real compounds K+ ) ; % II. Pronouns : ( {Wd-} & Spn1+ ); : ( {Wd-} & Spn2+ ); : ( {Wd-} & Spn3s+ ); : ( {Wd-} & Spn3p+ ); : Opn+; : M-; : [M-]; : ( or or or ( ( & ) or () ) or ); marA.pnacc mrA.pnacc torA.pnacc trA.pnacc: % Acc. pronouns 'me' & 'thee' ; "man.pn" "mn.pn" "mA.pn" : % 1st person pronouns or ; "tu.pn" "ComA.pn" "CmA.pn" : % 2nd person pronouns or ; "|u.pn3s" "ce.wh" "ceh.wh" "ch.wh" : % 3rd person singular pronoun or ; "]nhA.pn" "]nAn.pn" "|iCAn.pn" : % 3rd person plural pronoun or ; "CAn.pme" "tAn.pme" "mAn.pme" "C.pme" "t.pme" "m.pme" : % Nominal enclitics eg. ketAbam 'my book' ; % III. Determiners "ik.d" "iek.d" "yek.d" "kmi.d" "hic.d" "kmtr.d" "bEDi.d" "hmh.d" "hm.d" "hmdigr.d" "]n.d" "|in.d" "cndin.d" : D+ or [] or [] or [[]] ; /fa/words/words.numbers: {ORDM+} & ( D+ or [] or [] or [[]] ) ; m.ordm om.ordm: % Ordinal suffix morpheme ORDM-; % IV. Verbs % MACROS : ( ( {VMdur-} & {VMneg-} ) or {VMbe-} ); : ( VMP+ % for normal verbs with personal morpheme suffixes or CCF+ % for verbs with no personal ending, followed by conjunction or VMPP+ % for past participle verbs; in perfect tense or [()] % for verbs with no personal ending (eg. 3rd person singular past) or [RW+] % for verbs with no personal ending (eg. 3rd person singular past) ); : ( VMP+ % for normal verbs with personal morpheme suffixes or CCF+ % for verbs with no personal ending, followed by conjunction or (SUB+ or C+) or VMPP+ % for past participle verbs; in perfect tense or [()] % for verbs with no personal ending (eg. 3rd person singular past) or [RW+] % for verbs with no personal ending (eg. 3rd person singular past) ); : {@AV-} & {( O- or CCOB-)} % for one Object, or multiple ones & {@AV-} & {PP-} & {@AV-} & {[Wi- or [[C-]]]} ; : {( P- or CCAJP-)} % for one Predicate noun/adj., or multiple ones & {@AV-} & {PP-} & {@AV-} & {[ Spn3s- or Sns- ]} & {[(Wi- or C- or CC-)]} ; /fa/words/words.v UNKNOWN-WORD.v: ( % For Non-Future tense (Normal) & {@AV-} % Adverbs can occur almost anywhere, at a phrasal level & & {VMT+} & ) or ( % For Future tense I- & {@AV-} & & VMT+ & {( CCF+ % verbs with no personal ending, followed by conjunction or VMPP+ % past participle verbs; in perfect tense or RW+ % verbs with no personal ending (eg. 3rd person singular past) )} ) ; % IV. Verbs B. Intransitive Verbs /fa/words/words.v.intransitive: ( % for non-future tense & {@AV-} & & {VMT+} & ( or ) ) or ( % For future tense I- & {@AV-} & & VMT+ & {( CCF+ % verbs with no personal ending, followed by conjunction or VMPP+ % past participle verbs; in perfect tense or RW+ % verbs with no personal ending (eg. 3rd person singular past) )} ) % For Future tense ; % IV. Verbs C. Compound Verbs /fa/words/words.v.compound: (( & {( [[PP-]] or VMO- )} % For prep. phrases or acc. enclitics between noun.k and light verb & K- % "K"ompound verbs. This gives false positives (eg. German "*Du sprichst auf") & ) or IK- ) & {VMT+} & ; % IV. Verbs D. Auxiliary Verbs % See subordinating verbs % IV. Verbs E. Subordinating Verbs /fa/words/words.v.sub "bAid.auxpr" "bAis.auxpa" "tuAn.auxpr" "tuAns.auxpa" "xuAs.auxpa" "AyA.q" "AiA.q" "]iA.q" : % Simple subordinating verbs and Y/N & {@AV-} & & {VMT+} & ; /fa/words/words.v.compound.sub: % Compound subordinating verbs & {@AV-} & K- & & {VMT+} & ; % IV. Verbs F. Passive Verb Cu.vpas C.vpas: & {I-} & VPAS- & {@AV-} & & {VMT+} & ; % IV. Verbs F. Past Perfect Verb bu.vpperf: & {I-} & VPPP- & {@AV-} & & {VMT+} & ; % IV. Verbs G. Future Verbs xuAh.fut: % Verb that marks future tense ( VMP+ or VMPK+ ) & {I+} % ( K- % future compound verbs % & {VMO-} % optional accusative enclitic % & VMP+ % & IK+ % ) %or % ( VMP+ & I+ ) ; % IV. Verbs H. Verb Prefixes na.vmn n.vmn: % Negative verbal prefix VMneg+; mi.vmd: % Durative verbal prefix VMdur+; be.vms b.vms: % Subjunctive or imperative verbal prefix VMbe+; % IV. Verbs I. Verb Suffixes "d.vmt" "t.vmt" : % Verb morpheme, tense VMT-; "emAn.vmo" "mAn.vmo" "etAn.vmo" "tAn.vmo" "eCAn.vmo" "CAn.vmo" "am.vmo" "m.vmo" "at.vmo" "et.vmo" "t.vmo" "aC.vmo" "eC.vmo" "C.vmo" : % Verb morpheme, object ( [VMO-] or [VMO+] ); : ({C-} or {Wi-}) & ( ( {CC-} & {VMO+} ) or ( {CC-} & {VMO+} & { CCF+ or ( SUB+ or [C+] ) or RW+ } ) ) ; : % Optional subject can only be 1st person pronoun {Spn1- or CCS-} & ; : % Optional subject can only be 2nd person pronoun {Spn2- or CCS-} & ; : % Optional pronoun can only be 3s; liberal linkage to singular nouns {Spn3s- or Sns-} & {[]} ; : % Optional subject must be a plural link -- either pronoun or noun or conjunction {Spn3p- or CCS- or Sn3p-} & ; : % Workaround for complex predicates separated by future verb VMP- % Normal situation - no split complex predicate or ( % Situation with complex predicates separated by future verb VMPK- % & IK- & {( [[PP-]] or VMO- )} & K- & IK+ ) ; "im.vmp" "m.vmp" : % Verb morpheme, 1st person & ; "id.vmp" "i.vmp" : % Verb morpheme, 2nd person & ; "d.vmp" "O.vmp" : % Verb morpheme, 3rd person singular & ; "nd.vmp" : % Verb morpheme, 3rd person plural & ; h.per eh.per: % Verb Past Particple Morpheme Suffix %dh.perf th.perf deh.perf teh.perf: % Verb Past Particple Morpheme Suffix VMPP- & {( VPPP+ or VPAS+ % For 'to become', 'kardeh Chodan' )} ; "|im.perf" "|m.perf" : % Verb Past Participle Person Particle :-) 1st person VPPP- & ; "|id.perf" "|i.perf" : % Verb Past Participle Person Particle :-) 2nd person VPPP- & ; "|st.perf" : % Verb Past Participle Person Particle :-) 3rd person singular VPPP- & ; "|nd.perf" : % Verb Past Participle Person Particle :-) 3rd person plural VPPP- & ; % V. Prepositions and Case Particles /fa/words/words.pp: % Preposition, normal ( {EZ+} & ( PO+ % eg. 'barA ye1 man2' or CCPP+ % eg. 'barA ye1 man va2 to' ) ) & PP+; rA.acc: % Postposition, accusative PA-; % VI. Adjectives : (CCAMr- or CCAML+); : (CCAPR- or CCAPl+); %/fa/words/words.adj UNKNOWN-WORD.adj: % Adjectives, normal /fa/words/words.adj: % Adjectives, normal ( { AJM+ } &( Pa+ % Predicate adjectives or ) ) or ( ( M- % Attributive adjectives or ) & ( {} or { AJM+ } ) ) ; "tarin.ajm" "trin.ajm" "tar.ajm" "tr.ajm" : % Adjectival suffixes - comparative & superlative AJM-; % VII. Adverbs /fa/words/words.adv "kojA.wh" "kjA.wh": % Adverbs, normal AV+; % A.K.A. in English as MVa, CO, Em (Quickly I go, I quickly go, I go quickly) % VIII. Other Grammatical Words "ke.sub" "keh.sub" "kh.sub" ":" : % Complementizer/relativizer (the ":" is temporary) {SUBP-} & ( SUB- or REL- ) & C+; "agar.sub" "|gr.sub" "magar.sub" "mgr.sub" : % Complementizer "if" NOTE: it's kinda in between complementizer and coordinating conjuntion (moreso complementizer). hmmmm C+ & ({ SUB- or SUB+ }) ; "e.ez" "ie.ez" "ye.ez": % ezAfe -- for Gen., Adj. EZ- % Warning: This is not like the EZ link for English & M+; "brAi.subp" "brA.subp" "vaqti.subp" "uaqti.subp" "uqti.subp": % Complementizer Phrase % question: I know that you know. I know when you know. I know what you know. I know and you know. SUBP+ ; "va.cc" "ua.cc" "u.cc" "yA.cc" "iA.cc" "fqT.cc" "faqaT.cc" "vali.cc" "uali.cc" "uli.cc" : % coordinating conjunctions 'and, or, but' ( ( % for Prep Phrases {COMMA-} & CCPL- & (CCP- or CCPP-) & (CCPr+) ) % for Prep Phrases or ( % for Noun Phrases ( % for noun-nom {COMMA-} & CCNl- & CCNR+ & (CCN+ or CCS+) & {Wd-} ) % for noun-nom or ( % for noun-acc {COMMA-} & CCOl- & CCOR+ & {} & (CCO+ or CCOB+) ) % for noun-acc ) % for Noun Phrases or ( % Adjectives ( % Pred Adj % {COMMA-} CCAPl- & CCAPR+ & (CCAP+ or CCAJP+) ) % Pred Adj or ( % Attr Adj CCAMr+ & CCAML- & (CCAM- or CCAJM-) ) % Attr Adj ) % Adjectives ) or ( % for CP's {COMMA-} & {CCcc-} % CCcc is for multiple CC's & CC+ & {CCcc+} ) % for CP's or ( % for CP's in 1st position {COMMA-} & CCF- & {Wcc-} & CC+ & {CCcc+} % CCcc is for multiple CC's ) % for CP's in 1st position ; "," : ( ( ( % for Prep Phrases CCPL- & (CCP- or CCPP-) & (CCPr+) ) % for Prep Phrases or ( % for Noun Phrases ( % noun-nom CCNl- & CCNR+ & (CCN+ or CCS+) ) % noun-nom or ( ) % noun-acc ) % for Noun Phrases ) or ( {CCcc-} % CCcc is for multiple CC's & CC+ & {CCcc+} ) % for CP's or ( CCF- & CC+ & {CCcc+} % CCcc is for multiple CC's ) % for CP's in the first position or COMMA+ ) % needs to look for another CC to the right of it, also ; %":" ";" : % Colon and Semicolon %; %NUMBERS: ; LEFT-WALL: ( Wcc+ or [Wi+] or [[Wd+]] ) & {Xp+}; % Connects to the Subject in declarative sentence, or Verb in subjectless or imperative sentence RIGHT-WALL: RW- ; "." "?" "!" : % Period, Question Mark, and Exclamation Xp- & RW+ ; % The Junkyard % & ( % CCPr+ % or % CCcc+ % ) % ) % for the first CC in a PP % or % ( % CCcc- % & CCPr- % & ( % CCPr+ % or % CCcc+ % ) % ) % for multiple CC's in a PP, not the first CC %% For verbs %& {S-} %& {@AV-} %& {C-} %%% Formerly external files that have been incorporated into this file: % words.v.m.person.1 words.v.m.person.2 words.v.m.person.3s words.v.m.person.3p % words.pronoun.1 words.pronoun.2 words.pronoun.3s words.pronoun.3p % words.v.perfect.person.1 words.v.perfect.person.2 words.v.perfect.person.3s words.v.perfect.person.3p % words.v.m.object words.pronoun.m.enclitic words.v.m.tense words.adj.m.suffix words.q.aya words.comp.if words.n.m.suffix % words.v.aux words.det