The EANC Parser supplies every token (wordform) in EANC with a set of markup tags. This section provides an overview of the lexical morphological markup system used in EANC. It is divided into three subsections:
We also provide a full list of EANC lexical morphological tags for your reference.
EANC uses short tags for representing grammatical categories. Short tags were chosen over full descriptions in order to present information in a compact and easily readable format.
Short tags are used, for example, in the Gram Query Line, a field that is located under the Gram and Lexical Attributes link, where attributes selected in the Gram and Lexical Attributes window appear as a string of grammatical labels. Please note that there are a few cases where the relationship between the name of a checkbox in the Gram and Lexical Attributes window and the corresponding gram tags appearing in the Gram Query Line may not be obvious.
The same compact notation is also employed:
EANC tagging system follows the Leipzig Glossing Rules (LGR) as closely as possible. LGR, developed jointly by Bernard Comrie, Martin Haspelmath and Balthasar Bickel, include a collection of commonly used tags for cross-linguistically widespread grammatical categories. It has become one of the most influential international standards of abbreviations for grammatical category names.
LGR covers a major part of the grammatical structure of Standard Eastern Armenian, although a few non-LGR tags had to be added to the EANC tagging system. An important advantage of the LGR system is that its gram tags are in most cases self-explanatory. Please refer to the full list of EANC tags that provides examples and traditional Armenian tags.
Dative/Genitive. Distinguishing between genitive and dative is probably one of the most controversial issues in Armenian morphology. In nouns the putative cases are distinguished only by their ability to attach a definite article. Generally speaking, dative usually has one, while genitive never does. The EANC Parser does not distinguish genitive from dative and assigns both tags simultaneously (Gen/Dat). The only lexical category which always distinguishes between genitive and dative is personal pronouns. It is possible to interpret genitives of the personal pronouns as separate lexemes - possessive pronouns. EANC morphological model follows the more traditional view, according to which they are members of the case paradigm.
Converbs. From a typological point of view, a typical converb is used non-finitely in a subordinate predication while never or rarely finitely as a predicate of the main predication. The forms that we call converbs in Armenian are very rarely used as main predicates, at least in written discourse. However, they do not necessarily have to be subordinate verbal forms. Some of them are more often used with auxiliaries. We use the term converb to refer to those SEA forms that are used as subordinate predicates and/or as dependent parts in auxiliary verb constructions. We do not distinguish between these two converb functions because the most widely used converbs have both. In other words, converbs are understood broadly as dependent verbal forms lacking person-number agreement.
Destinative converb ( Ապառնիդերբայ apaŕniderbay ). Note that destinative converb and genitive/dative infinitive are always formally identical. The EANC Parser is unable to distinguish between these two functions because it does not use any context information (e.g. presence or absence of an auxiliary verb) when it assigns morphological tags. Typologically, the use of a case form of action nominals such as infinitives to mark future is not uncommon. We decided therefore to analyze such converbs as true case forms of infinitives. However, the more traditional destinative tag is preserved in the Gram and Lexical Attributes window for convenience purposes. Note that when destinative is selected in the Gram and Lexical Attributes window, the corresponding query line appears as ‘Inf, Dat, ~Def’. The search output will contain both destinative converbs and true nominal uses of genitive/dative infinitives (non-definite only).
Associative plural. Associative plural is a nominal plural form indicating a group of people associated with one salient member of this group (more common in spoken language and dialects). Examples are Վարդանանք ‘Vardan and his associates’, Շուշանենք ‘Shushan and her associates’. This category also covers pronominal forms like մերոնք ‘our group, our kin’ which are morphologically the same as nominal associative plural though more problematic from the semantic point of view.
Relational noun . A relational noun is a rare nominal form, mostly found in SEA oral discourse, although it occurs in written discourse as well. Morphologically, a relational noun is a nominal stem followed by a genitive marker followed by a definite article morpheme followed by either another definite article or a genitive/dative ending with a definite article or a non-core case marker.
Functionally, a relational form is a nominalized genitive, possibly suffixed with nominal morphological markers, such as a case ending or a definite article. The resulting meaning is close to 'that of / relating / belonging to N', where N is the nominal stem of the form. There are reasons not to interpret the first of the two article morphemes as an article, because it acts as a nominalizer rather than a determination marker. Therefore, relational nouns in EANC are assigned ‘gen’, ‘nmlz’, and ‘def’ tags; this is how they appear in the Gram Query Line. Note also that if you check the Nominalized checkbox in the gram selection form, relational nouns, being essentially nominalizations of genitive, will be found too.
Nominalized Attributes. This category covers various case-number forms of headless (stand-alone) adjectives, participles, and genitives.
full list of EANC tags). However, there is a small number of cases where this relationship may not be obvious; we provide a list of such cases below.
1. Relational noun (under Nominalizations), when checked in the Gram and Lexical Attributes window, is represented in the Gram Query Line as
Relational = ‘gen, nmlz, def’
2. Any converb, when checked, is provided with an additional ‘cvb’ tag:
Imperfective = ‘cvb, pfv’
Perfective = ‘cvb, pfv’
Connegative = ‘cvb, conneg’
Simultaneous = ‘cvb, sim’
3. The only exception to 2. is the destinative converb:
Destinative = ‘inf, dat, ~def’
4. Any participle, when checked, is provided with a ‘ptcp’ tag.
Subject = ‘ptcp, sbj’
Resultative = ‘ptcp, res’
5. When checked, the ordinal numerals checkbox under the Lexical categories tab in the Gram and Lexical Attributes window, produces ‘Num, A’ in the Gram Query Line. This reflects the dual nature of cardinal numbers, which are adjectives from the morphosyntactic point of view, but semantically are related to numerals.
6. Independent (or nominal) pronouns like դու ‘you.sg’, նա ‘he, she’, ինչ ‘what’, ով ‘who’ are represented by ‘S’ (for substantival pronoun) in the Gram Query Line.