Acknowledgements:
The most appreciated base used for the elaboration of the application
was constituted by morphosyntactic tables for Czech elaborated by Hana
Skoumalová. For all errors solely the author is to blame.
1. Czech distinguishes masculine animate (Animate=yes above) and masculine inanimate (Animate=no) Gender. Masculine inani- mate Nouns always have the same form in the nominative and ac- cusative Case, whereas masculine animate Nouns have predomi- nantly the same form in the genitive and accusative Case. Mas- culine animate Nouns and masculine inanimate Nouns differ in accusative singular and nominative plural only.
2. The dual Number manifests itself only in the instrumental Case of several Nouns denoting dual parts of the human body.
3. Czech distinguishes 7 cases, the locative Case being obligato- rily prepositional. With most paradigms, the form of the voca- tive differs from that of the nominative in singular; in plu- ral the form of the vocative is identical to that of the nomi- native for all the paradigms.
4. Verbal nouns are classified as Nouns.
1. The verb "být" (E. "to be") in all its functions is characte- rized as Type=c (i.e. the copula), which clearly is an over- simplification because the verb has more meanings (auxiliary etc.).
2. Auxiliary verbs (Type=a) include neither the verb "být" (see above), nor the modal verbs.
3. The 'past participle' in Czech is used for expressing compound active past Tense and is encoded as: Type=p(articiple), Tense=p(ast), Voice=a(ctive).
4. Passive participles are encoded as: Type=participle, Tense='-', Voice=passive. This encoding is used only for the passive participle in the predicative position, e.g. "byl chycen" (E. "(he) was caught"). The case marked passive participles, both in the predicative and in the attributive position, e.g. "padělaný rukopis" (E. "forged manuscript"), are classified as (qua- lificative) adjectives.
5. Adjectival active and passive participles, e.g. "stojící" (E. "standing") or "udělaný" (E. "performed" or "done", cf. Note 4 above) are classified as adjectives.
6. Negative verbs are marked as Negative=y, whereas non-negative verbs are marked as Negative=n.
7. The term transgressive roughly corresponds to the term 'verbal participle'. The transgressives distinguish: Number (singular, plural), Gender (masculine, non-masculine) Tense (past, present) The Gender opposition of masculine animate vs. masculine ina- nimate is neutralized in transgressives; thus, there is one masculine form and one non-masculine (the form of feminine and neuter is neutralized).
8. Gender and Animate values correspond to those associated with Nouns and are necessary to account properly for agreement.
9. Gender manifests itself in past participles, passive parti- ciples and transgressives only.
10. Normally, Verbs form the future Tense periphrastically by auxiliary "být" (E. "to be") plus infinitive of the main Verb. In addition to the copula, there are, however, some Verbs which form future Tense non-periphrastically, i.e. synthetically (Verbs of motion). Such verbal forms are marked as Tense=f.
11. The 'yes' value of the Clitic_s attribute denotes a verbal form having the clitic morpheme 's' appended as a suffix. This 's' morpheme expresses 2nd Person singular present Tense of the auxiliary Verb "být" (i.e. the form "jsi"). There is no intermediate hyphen between the verbal form and the 's' morpheme.
12. The Clitic_s attribute is specified for VForm=infinitive (VForm=n) and Vform=p(articiple) only.
13. Some modal and auxiliary Verbs do not form imperative and transgressive.
14. The voice value ('a' or 'p') is not specified for VForm=c(onditional) and VForm=t(ransgressive), in which case VForm='-'. \end{verbatim}
1. Three deverbative adjectival participles, i.e. past active participle, passive participle and present active participle are not distinguished. They are conflated in the 'qualifica- tive' value of the Type attribute (Type=f).
2. Only qualificative (and passive participle) Adjectives can be specified for Degree and for nominal (short) Form.
3. The attributes Gender, Number, Case and Animate correspond to the same categories within the Nouns. They are necessary for the proper account of agreement of Adjectives with Nouns.
4. The Formation attribute distinguishes a nominal (short) form from a so-called compound (long) form of an Adjective. The no- minal form can be used in the predicative function only. It is specified for nominative and accusative Case only.
5. The Adjectives which have no degrees of comparison have the Degree value equal to p(ositive).
1. Gender, Number, Case and Animate correspond to the same categories as specified for Nouns. They are necessary for the proper account of agreement of adjectival Pronouns with Nouns.
2. Type=reflexive ecompasses all reflexive pronouns (ªseº, ªsebeº, ªsiº, ªsvůjº) as well as "se" in its role as the obligatory particle of reflexive verbs. Personal and possessive reflexives are further distinguished via the Referent_Type attribute. "se" in all its roles will be marked as the reflexive personal clitic pronoun.
3. Pronouns are distinguished between having a (syntactically) nominal and (syntactically) adjectival function. All pronominal types except the demonstrative and possessive one can be nominal, and all except for the personal one can be adjectival.
5. Referent_Type is used to distinguish personal reflexives (which include "se" in all its functions) from the possesive reflexives ("svůj").
6. Negative and general Pronouns ("general" Pronouns concern the Pronouns like "všichni" [E. "all"], "každý" [E. "every"] etc.) are important from the viewpoint of their syntactic distribution.
7. The Clitic attribute distinguishes clitical vs. nonclitical pronominal forms, e.g. "ti" vs. "tobě".
8. The 'yes' value of the Clitic_s attribute denotes a pronominal form having the clitic morpheme 's' appended as a suffix. This morpheme expresses 2nd Person singular present Tense of the auxiliary Verb "být" (E. "to be"). There is no intermediate hyphen between the Pronoun and the 's' morpheme.
9. Owner_Number and Owner_Gender concern the possessor's Number and Gender, respectively.
1. The classification given above is considerably simplified as
compared to traditional Czech grammars. There can be other
classifications as well, eg.:
(a) syntactic: Adverb modifying a Verb
Adverb modifying an Adjective
Adverb modifying another Adverb
Adverb modifying a Noun from the left
Adverb modifying a Noun from the right
(b) semantic temporal
spatial
qualitative
...
We refrained from this typology here.
2. Particles form a separate part of speech category (see below) as is customary in Czech grammars.
3. The Adjectives which have no degrees of comparison have the Degree value equal to p(ositive) similarly as Adjectives.
1. Czech has only Prepositions, no postpositions.
2. For the disambiguation of word forms belonging to declension parts of speech it seems necessary to include the information about the case which each Preposition requires.
3. A preposition can be contracted with a pronoun; such a preposition has Formation=c(ompound).
1. Conjunctions "abych", "abys", ..., "kdybych", ... are composed of the Conjunction "aby" or "kdyby" and the conditional particle ªbyº ("bych", "bys", "by", "bychom", "byste"). These Conjunctions are therefore specified for Person and Number for the sake of agreement relation.
2. The class of two-part Conjunctions has not been introduced.
1. Numerals have been specified as a separate category because of their specific syntactic distribution. We have specified two syntactic classifications by means of the attributes Type and Class; they concern different syntactic distributions. For instance "několik" (E. "several") will be characterized as: Type: cardinal Class: indefinite
2. Among the definite numbers there are four subclasses (definite1, definite2, definite34, definite) which differ in their syntactic distribution and contain the following Numerals: {1}, {2}, {3,4}, {5,6,...}
3. Gender, Number and Case correspond to the same categories as specified for Nouns. They are necessary for the proper account of agreement of Numerals with Nouns.
4. In direct cases (nominative, accusative) the cardinal numerals belonging to the class with one of the values {fdiqr} (see above) behave like neuter Nouns in singular, whereas in oblique cases they behave like adjectival attributes in plural.
In this section all the attributes presented in the tables are listed in alphabetical order.
The values presented within the tables are, in the following, listed in alphabetical order; the first column gives the name of the value, the second column its code and the third lists attributes for which the value is appropriate.
This index gives the list of morphosyntactic descriptions (MSDs) and their features. In the table below, the first column gives the MSD, the second its expansion into a feature-structure, the third gives the number of entries in the lexicon (184,470 entries), and the fourth gives some examples as word-form/lemma. The list was extracted form the Czech MULTEXT-East lexicon.