 |
|
|
|
Dominic Anctil. The notion of lexical error in the framework of Explanatory Combinatorial Lexicology.Lexical
mastery constitutes a central component of language mastery.
Nevertheless, few studies dealing with linguistics errors - notably in
written language - are interested in the lexical aspect. Furthermore,
the actual notion of lexical error seems vague, both in linguistics and
language teaching, and no precise definition of it has been proposed.
In other respects, many errors considered as syntactic or grammatical
are in our opinion attributable to lexical problems.
Using
Explanatory Combinatorial Lexicology·s principles of lexical
description as a guideline, we try to clearly characterize the notion
of lexical error. Through an exhaustive presentation of the proprieties
of the lexical unit, we will observe in which ways they can be violated
or transgressed to cause lexical errors. This approach will provide us
all the necessary tools to propose a descriptive typology of lexical
errors. We will finally put our typology to the test by describing
lexical errors found in a text corpus.
Jurij Apresjan. Prolegomena to Systemic Lexicography.The
principle of integrated linguistic descriptions is stated which
requires a unified (fully coordinated) description of dictionary and
grammar. Lexemes in such a description are assigned all properties that
linguistic rules (semantic rules included) may refer to. The result are
integrated lexicographic representations of lexemes comprising many
novel types of lexicographic information. Lexemes sensitive to the same
rules form a complicated system of intersecting lexicographic types
which are to be uniformly described in the dictionary.
Lucie Barque, Alexis Nasr. Computing semantic relations on structured lexical definitions.The
work described in this paper is based on the BD?F approach and shows
how the BD BD?F definitions can be used to compute semantic relations
between lexical units.
Pascale Bélanger, Richard Kittredge. Paraphrasing with Space Constraints: Linguistic Operations in Journal Abstracting.Recent
research into text summarization provides an opportunity to test,
expand, and eventually implement MTT-based rules for paraphrasing and
other linguistic operations as part of a general model of text
condensation. Our studies of French and English journal articles in
biology and medicine show that lexical functions are important in
modeling the transformation of selected full-text (FT) sentences into
the sentences used for an article·s author-written abstract (AB). A
full model will require an integrated account of anaphor resolution,
paraphrastic operations, entailment rules, and their interaction with
rhetorical rules operating over multiple sentences. We present some
general aspects of this potential long-term application for MTT, and
then focus on several important and frequent sentence-level operations
found in corpus studies of both languages.
Olga Boguslavskaya. Ïðîòèâîïîñòàâëåíèå îáúåêòèâíîé è ñóáúåêòèâíîé ïðè÷èíû â ñåìàíòèêå ïðè÷èííûõ ãëàãîëîâ.
(The "Objective Cause" vs. "Subjective Cause" Opposition in the Semantics of Russian Verbs)The
paper is devoted to semantics of several Russian verbs which include
the component of causation. It is described against a background of
other manifestations of this meaning. It is shown that the major
semantic oppositions of objective vs. subjective cause and immediate
vs. mediate cause, which are characteristic of other types of
causatives and telic words, are also relevant for the words in
question. Constraints on non-actions in the capacity of the initial
situation have been discovered. The format of the semantic definition
is discussed.
Igor Boguslavsky. Actant properties of Adverbial Derivatives.The
actant properties of adverbial derivatives (AdvD) are much more diverse
that those of other types of syntactic derivatives, such as
conversives, Si, Ai, or verbal adverbs. We show that in a general case
a valence slot of an AdvD can be either (a) blocked, or filled (b) by a
dependent of the AdvD itself, or (c) by the dominating verb, or (d)
somewhere within the clause organized by the dominating verb; in the
latter case the AdvD actant may be identified based on its syntactic
position, its semantic role with respect to its mother element, or its
communicative function.
Krystyna Bojałkowska, Zygmunt Saloni. On
the "Subject" of Polish Constructions with "Adverbial Participles" in
the Framework of Meaning-Text Theory.The paper deals with
Polish constructions with forms of verbs called "adverbial
participles". These forms do not combine with the nominative subject in
the surface-syntactic structure, but they have the first actant slot in
the deep-syntactic structure. The first deep-syntactic actant of the
adverbial participle is identified with some actant of the other verb
being a component of sentence structure. The problem of the
identification in various cases is discussed.
Igor Bolshakov. Stable Coordinate Pairs as a Specific Resource of Language.Stable
coordinated pairs (SCPs), e.g., Eng. "comments and suggestions","near
and far", "sooner or later" occur frequently in texts, whereas there
only exist a few thousands of them in various European languages. To
statistically characterize SPCs, we propose a Stability Indicator
similar to Mutual Information Indicator. To structurally categorize
SCPs, we introduce parameters of a lexical, morphological, syntactic,
semantic, and pragmatic nature - for entire SCPs and for their
components. SCPs are compared with collocations (stable word
combinations with subordinate dependencies between components), and
some similarities are found. It is argued that gathering SCPs into a
special database can facilitate a variety of NL processing
applications. The research is based on the gathered and verified
collection of ca. 3700 Russian SCPs.
Alexandre Dikovsky, Boris Smilga. Semantic Roles and Diatheses for Functional Discourse Plans.In
this paper, we elaborate the notion of semantic diathesis as proposed
recently by Paducheva. Our definition differs from Paducheva·s in the
following aspects:
strong semantic typing, including optional types;
roles stemming from proto-situations instead of from lexical definitions;
different system of communicative ranks;
most important, diatheses are used for semantic and discourse planning rather than for lexical derivation.
Dmitry Dobrovol·skij. Semantic Motivation
of Lexical Functions and its Limits (The Combinatorial Profile of the
Russian Degree Modifier črezvyčajnoThe point of
departure is the idea that many combinatorial phenomena which were
traditionally regarded as arbitrary, in fact, are motivated
semantically, at least to a certain extent. The empirical data analysed
here is the use of the Russian degree modifier črezvyčajno
which incarnates the Lexical Function MAGN. The factors that motivate
its selection constraints can be either of a semantic or pragmatic
nature, or its co-occurrence can be driven by usage.
Ronald Feldstein. On the Aspectual Derivation of "Dual Simplexes".It
is well known that the so-called motion verbs of Russian have special
aspectual, derivational, and syncretic properties. This paper concerns
a class of non-motion verbs, which shares many of the dual simplex
properties of the motion verb class (e.g. brosi-/brosaj-, vesi-/vešaj-,
xvati-/xvataj-). Possibly, the form a special class which might be
called "pseudo-motion" verbs, in view of their formal similarity to the
motion verb class. The semantic connections to motion verbs might lie
in a gray area, with some similarities, but without an exact parallel
to the determinate/indeterminate category.
Valentina Gavrilova. Âçàèìîäåéñòâèå
êàòåãîðèé ïåðåõîäíîñòè è çàëîãà â ðóññêîì ÿçûêå (ïðîèçâîäÿùèå
è èíèöèèðóþùèå ïåðåõîäíûå ãëàãîëû è èõ âîçâðàòíûå êîððåëÿòû)
(Interaction
of Transitivity and Voice Categories in Russian (Transformative and
Initiating Transitive Verbs and their Reflexive Correlates))Distinction
between productive and initiating transitivity is considered to be a
covert category. The relations between the subject and the object of
both types of transitive constructions are examined.
Kim Gerdes, Sylvain Kahane, Hi-Yon Yoo. On the Descriptive Adequacy of Topology.This
article attempts to sum up the history of the topological structure in
MTT and to analyze its prospects in the future development of the
theory. We discuss the raison d·?tre for considering the existence of
the topological constituent tree as well as the constraints on its
form. We discuss in particular the necessary information for the
computation of the topology-phonology interface.
Alexey Gladky. Ðàçìûøëåíèÿ î âçàèìîäåéñòâèè ëèíãâèñòèêè è ìàòåìàòèêè
(Deliberations on the Interaction between Linguistics and Mathematics).Pollination
with the "mathematical spirit" and methods turned linguistics toward
more precise and objective lines. Yet there are serious drawbacks on
this way. The author is thinking why the two sciences are coming closer
to each other, how far is mathematics applicable to linguistics, and
which factors oppose understanding between mathematicians and linguists.
Leonid Iomdin. A Hypothesis of Two Syntactic Starts.An
original surface syntax representation is proposed for a class of
semi-idiomatic constructions that defy regular syntactic analysis
within the Meaning-Text framework. Semantic properties and lexical
constraints for this class of constructions are discussed in detail.
Lidija Iordanskaja, Alain Polguère. Hooking up Syntagmatic Lexical Functions to Lexicographic Definitions.This
paper presents a research in progress conducted within the theoretical
and descriptive framework of explanatory combinatorial lexicology. Its
general topic is the modeling of syntagmatic lexical relations in
formal dictionaries. A syntagmatic lexical link between two lexical
units L1 and L2 corresponds to a collocation -in Hausmann (1979) and
Mel·čuk
(1998) sense- of which L1 is the base (the controlling
element) and L2 the collocative (the controlled element). The question
we are trying to answer is how meanings expressed by collocatives
connect to specific semantic components in the definition of bases in
collocations.
Martin Janssen. Between Inflection and Derivation. Paradigmatic Lexical Functions in Morphological Databases.Modern
day morphological theory posits borderline cases between derivation and
inflection, behaving partly like inflections, and partly like
derivations. But in morphological database, such in-between cases pose
a problem. This article shows how lexical functions can be used to
resolve this problem. However, it also shows that the use of lexical
functions in this way alters the nature of lexical functions
considerably.
Sylvain Kahane, François Lareau. Meaning-Text Unification Grammar: Modularity and Polarization.This
article presents the Meaning-Text Unification Grammar·s current state,
now that its formal foundations have been clarified with the
development of Polarized Unification Grammars. Emphasis is put on the
model·s architecture and the role of polarization in linking the
various modules - semantic-syntax interface, syntax-morphotopology
interface and the well-formedness grammars of each representation level.
Andrei Kibrik. On the Use of "Quasi-Grammeme" in Athabaskan.I
demonstrate how the intricate morphological system of Athabaskan
languages can be treated with respect to derivation vs. inflection
dichotomy. Many prefixal position on the Athabaskan verb template can
be attributed stable labels "derivation" or "inflection". A striking
feature of Athabaskan is that inflection is generally closer to the
root than derivation. There is a significant residue of positions in
the Athabaskan template that we cannot readily identify as either
derivational or inflectional. There are at least two reasons for that.
First, there is a large group of quasi-inflectional positions; this was
identified due to the illuminating notion of quasi-grammeme proposed by
Mel·čuk. The second reason for complications is template positions that
can variously be associated with either derivation, or
quasi-inflection, or inflection. Generally, the inflection vs.
quasi-inflection vs. derivation opposition is a key to understanding
the intricate synthesis implemented in the Athabaskan verb.
Mikhail Kopotev. Guljat· tak guljat·: between Grammar and Dictionary.The
presentation is devoted to the analysis of two idiomatical types of
Russian sentences and discussion of the question concerning place of
these phrasemes in the Grammar or in the Dictionary. In the first part
the difference in filling variables will be shown; in the second the
contextual dependence of phrasemes will be analyzed; the third part is
devoted to the analysis of the most frequent realizations of the
specified phrasemes.
Sergey Krylov. Âëèÿíèå ÌÑÒ íà îáùóþ ëèíãâèñòèêó (â ñâåòå ïîäâåäåíèÿ ïðåäâàðèòåëüíûõ èòîãîâ «ñåìàíòè÷åñêîé ðåâîëþöèè»)
(The impact of Meaning-Text Model on General Linguistics (in the Light of Preliminary Results of the “Semantic Revolution”)).Our
subject will be those ideas of MTT which overcame its "borders" and
became part of general linguistics outside the borders of this school.
The main principle of MTT is the treatment of every linguistic
phenomena in the framework of such a description which can be
effectively used (implemented) by an automat (a computer) as a source
of data for the successful work of a machine translation system. Basing
on this nuclear principle, one can explain all specific peculiarities
of MTT.
Leonardo Lesmo, Livio Robaldo. MTT and the quantifier scope ambiguity.Sentences
with more than one quantifier can have different interpretations
depending on the scope order of their quantifiers. To our knowledge,
this type of ambiguity has received a lot of attention in the
Underspecification literature but not in the MTT one. In this paper, we
define a new underspecified SemS of MTT which refers to all possible
readings of a sentence. In order to specify a particular scope order,
we introduce additional dependencies. They roughly correspond, in FOL,
to Skolem constants and functions and are part of a new additional
structure of SemR level, named SemCtxDepS.
Alexander Letuchiy, Natalia
Serdobolskaya. Object or situation: the factors of expressing the
arguments of mental verbs in Russian.The paper deals with the
verbs taking both clausal and noun complements (i.e. mental verbs and
verbs of perception and emotion) in Russian. Empirical data have been
acquired through psycholinguistic experiments with teenagers. It is
maintained that the choice of clause vs. noun as a complement is due to
the cognitive and discourse factors.
Irina Levontina, Alexey Shmelev. The particles one cannot do without.It
is sometimes the case that in a given communicative situation, a
certain particle is pragmatically obligatory even though the speaker
does not seek to express the associated meaning. In the absence of the
relevant particle, the wrong implicatures may appear, or the text may
cease to have coherence, or the utterance is perceived as a
manifestation of arch use of language. The lack of the particle where
the communicative situation calls for it every so often gives the
impression that we deal with a word-for-word translation of an original
text containing no similar marker.
If a particle
is pragmatically obligatory in a certain type of contexts in a given
language, and a context of this type has to be translated from a
language with no such marker into the language in question, the
translator may be confronted with a problem. If s/he translates the
text literally (that is, with no particle), s/he will convey the
precise meaning, but the translation will be inadequate (the wrong
implicature will appear). However, if s/he adds the needed particle,
the translation will be more adequate even though a new semantic
component is added.
In the paper, we put
emphasis on the use of the particle åùå in combination with a verb in
future tense; we have established a type of contexts where åùå is
pragmatically obligatory.
Lennart Lönngren. Valency Analysis of Adjectives in Attributive Position.Valency
analysis is based on a non-componential approach to lexical meaning.
With respect to the position of the predicate uniting the two parts of
the syntagma four types are identified. The predicate may be expressed
by 1) the adjective: novyj dom ·(a) new house·; 2) the noun: golovnaja bol· ·headache·; 3) a segment incorporated in the adjective: < vnutri>partijnye konflikty ·conflicts within the party·; 4) a segment incorporated in the noun: Bozh·i >sozda>nija ·creations by God·.
Olga Lyashevskaya. Íàø ñîñåä ïî ñòàðîé äà÷å: ïðåäëîæíûå àêòàíòû è ìîäèôèêàòîðû ó ïðåäìåòíûõ è ðåëÿöèîííûõ èìåí
(Nash sosed po staroj dache: Prepositional Actants and Modifiers of Names of Things and Relational Nouns).Cases
of prepositional coding of actants and modifiers of names of persons
and things (cf. our old cottage neighbour, meat on coals) are
considered. The status of the prepositional constructions is hard to
define because syntactic criteria standard for verbs are of no use
here. The semantic criterion to distinguish between actants and
modifiers is offered: if a predicate introduced by a prepositional
group and its arguments are identical with the (main) predicate of the
lexical decomposition of the noun, than the preposition group is
considered to be actantial.
Igor A. Mel·čuk. Meaning-Text Theory and
the Conceptual Apparatus of Modern Linguistics: Towards a Better
Description of Case and Voice in Maasai.1. One of central
goals of Meaning-Text theory is to develop a stable and rich
con-ceptual system for linguistics that will allow linguists to
formulate unambiguous analytical and descriptive statements in such a
way that their internal consistency and accuracy can be established by
logical procedures. The first stage is the development of such a system
for linguistic morphology (Mel·čuk 1993-2000, in five volumes).
2.
The proposed system of morphological concepts is tested on the data of
Maasai-as far as Maasai cases and voices are concerned.
3. The
Maasai noun distinguishes (tonally) two case forms. Traditionally one
case (among other things, the case of the Grammatical Subject) is
called the ·nominative· and the other (among other things, the case of
the Direct Object), ·accusative·. How-ever, such a description leads to
many paradoxical statements (for instance, the accusative is the
basic-lexicographic-form). A careful analysis of the facts coupled with
better definitions of concepts allows us to redress the situation: the
traditional nominative turns out to be an oblique, and the traditional
accusative is actually a nominative (which in the new approach is the
case of nomination). Under this analysis, the description of Maasai
appears typologically more plausible. An interesting corollary: Maasai
has an ergative construction typical of all verbs (transitive as well
as intransitive ones, including even the copula).
4. The Maasai passive (as described in Tucker and Mpaayei 1955)
was posing serious problems for a formal description: traditionally,
its Grammatical Subject was said to be in the accusative. Based on a
general calculus of voices within the Meaning-Text theory, we can show
that this form is actually a suppressive (rather than a passive).
5. Interesting parallels are drawn between passives, ·impersonal passives,· and supressives of a number of languages.
Irina Mikaelian, Anna Zalizniak. The Russian Conjunction A revisited.We
refine the semantic description of the conjunction A and propose an
explanation of two interrelated facts: the crucial position of this
conjunction within the Russian system of coordination and the absence
of its direct equivalents in Western languages. The conjunction A
appears as the most "empty" coordinative conjunction in Russian. It
does not imply any particular objective relation (temporal or logical)
between two confronted facts. It rather functions as a meta-textual
operator and can express, in an implicit way, a range of very important
attitudes. We suggest that Russian conjunction A is a language-specific
word because it can render a specific conceptual configuration which is
also present in a number of other language-specific words and
constructions.
Jasmina Milićević. Semantic Paraphrasing in Meaning-Text Linguistic Models.Within
Meaning-Text linguistic Theory, linguistic paraphrase has a place of
honor: it is considered both as an integral part of Speaker·s
linguistic competence and a central tool for language modeling.
Linguistic models based on this theory have a built-in paraphrasing
system-a set of universally valid equivalence rules operating at the
syntactic level of representation of utterances. Meaning-Text
paraphrasing system is capable of treating a wide range of fairly
sophisticated lexical-syntactic paraphrases. However, this system is
unable to treat in a natural and elegant way certain approximate
paraphrases based on deep paraphrasing links, i.e. links requiring
semantic decomposition of lexical meanings in order to be discovered.
This paper proposes a new type of paraphrasing rules capable of
handling such paraphrases-semantic equivalence rules, operating at the
semantic level of representation of utterances.
Elena Paducheva. Ñîîòâåòñòâèå «Ñìûñë—Òåêñò» â èñòîðè÷åñêîé ïåðñïåêòèâå
(Meaning — Text Correspondence in a Historical Perspective).Modern
semantics is unthinkable without those notions and devices of semantic
description that originated in the frame of "Meaning – Text" model -
such as valency, semantic and syntactic argument, diathesis. "Meaning –
Text" model gave rise to the notion of sentential form and lexical
decompositions having predicate-argument structure. This theory is
responsible for the idea of integral description of language and
systemic lexicography.
Anna Panina. Towards a Description of Valency Transfer to Operi, Funci and Laborij –
Type Verbs in Japanese.Towards
a description of valency transfer to Oper, Func and Labor-type verbs in
Japanese Japanese language presents a good opportunity to study
transfer of valencies between nouns and Oper, Func and Labor-type
verbs, because an actant·s marking is a reliable indication of whether
the valency is realized by the noun or transferred to the verb.
Preliminary study of three Japanese verbs - Oper1 OKONAU "to carry
out", Oper2 UKERU "to receive", and Labor13 MOCHIIRU "to use" - in
collocation with nouns shows some tendencies in their behavior. The
valency of the noun that, when transferred to the verb, becomes the
verb·s subject (1st for Oper1 and 2nd for Oper3), is realized by the
verb. On the other hand, the valency of the noun which, if transferred
to the verb, would interfere with the syntactic position which is
already occupied by the noun itself, tends to be realized by the noun.
Barbara Hall Partee. Diathesis alternations and NP Semantics.In
this paper we examine the relationships among diathesis alternations,
the semantics of verbs, and the referential status of NPs. Some
diathesis shifts are argued to involve changes in the semantic type of
NP arguments, including possible alternations between ·referential· NPs
and ·property-type· NPs. We explore applications of this approach to
alternations of Genitive and Accusative, both with intensional verbs
like ?dat· ·wait·and with the Genitive of Negation.
Vladimir Plungian, Ekaterina Rakhilina. Ïðèçåìëèòüñÿ è ïðîìàõíóòüñÿ: ñåìàíòè÷åñêèå ìåõàíèçìû ñèíòàêñè÷åñêèõ îãðàíè÷åíèé
(Russian prizemlit’sja ‘to land’ and promaxnut’sja ‘to miss’: Semantic Reasons for Syntactic Blocking of the Expression of Participants).The
paper reopens a discussion of such cases when an obligatory participant
cannot have a syntactic expression within the corresponding verbal
phrase. MTT always used the Russian verb promaxnut·sja ·to miss· as a
classic example of this kind. We show that the similar restrictions
operate for a considerable number of Russian predicate lexemes of
different nature. A possible explanation for this type of syntactic
behaviour is related to the (poorly elaborated) rules of participant
interaction, which make the meaning of semantic variables "fixed" and
yield so-called "situational constants".
Amanda Pounder. Process-and-Paradigm Morphology: A Separationist Sign-Based Framework.Process
and Paradigm Morphology" is a framework based directly on I.A.
Mel·čuk·s morphological framework as it appears in Mel·čuk 1982, 2000
and elsewhere. It shares with that framework the foundational element
of the linguistic sign and the morphological operation as an extension
of the sign. It departs from it, however, in proposing a more complex
structure for the morphological operation, and more importantly, by
elaborating the notion of the morphological paradigm. The talk will
outline some basic elements of Process and Paradigm Morphology,
focusing first on the structure of the sign and expression of
morphological rules and operations. It will then introduce the
morphological paradigm in both its inflectional and derivational
manifestations, and finally show how the paradigm can be extended to
encompass complex predicates such as auxiliary constructions.
Tilmann Reuther. Metaphor in the Meaning – Text Model.First,
the paper gives an overview on actual means how metaphor is dealt with
in MTT and ECD. Second, it presents approches to metaphor by other
linguists (e.g. Paducheva 2004). Third, several types of metaphoric
meaning and lexicographic definitions are presented, and a literary
example is used to demonstrate the role of context in the
interpretation of metaphor. Forth, perspectives for further work on
metaphor within MTT are listed.
Elena Rudnitskaya. An attempt of implementing the Generative Grammar Island Constraints in the "Meaning-Text" model.The
paper considers major types of Islands and Island Constraints in the
Generative Grammar, their syntactic and information structure
interpretations and possible ways of integrating these constraints into
the "Meaning - Text" Model as filters on the stage of synthesis/
linearization (Deep Morphological Structure).
Begoña Sanromán Vilas. Individual-Level and Stage-Level Predicates: The Spanish Emotion Nouns.The
goal of this paper is to present a set of characteristics of Spanish
emotion nouns, aspectually classified as states, that shows the
possibility to delimitate these stative predicates as individual-level
or stage-level predicates. As a starting point we assume that there are
two groups of Spanish emotion nouns according to the semantic role
played by the cause/origin of the emotion: the internal cause emotion
nouns (IC) and the external cause emotion nouns (EC). In IC nouns, as
respeto ·respect·, the emotion is born in the experiencer as a result
of a judgment made about an object. With EC nouns, as asombro
·amazement·, it is required the existence of an external fact that
triggers the emotion. This semantic difference between both groups is
not isolated. We consider that it goes through all the grammatical
levels in such a way that it is possible to outline morphological,
syntactic and lexical co-occurrence patrons of behavior. We shall use
some of these patrons of behaviour, for instance, their paradigmatic
and syntagmatic co-occurrence as evidence for
individual-level/stage-level contrast between the nouns.
Zoya Shalyapina. The Formalism of Lexical Functions:
Some of its Possible Uses, Applications and Extensions.The
formalism of lexical functions, designed as a means of generalizing and
controlling lexical aspects of periphrastic transformations, is as
multifunctional as such transformations are manifold. In automatic text
processing, this makes possible its employment for more than one
purpose. The paper outlines some applications and extensions of this
formalism and its underlying ideas within an automatic translation
framework.
The
paper presents an overview of a newly developed tool included in
AutoPat, - a patent claim generator, for textual input proofing.
AutoPat relies on joint human-computer content specification in a
textual form and uses MEANING-TEXT THEORY as a theoretical framework.
The proofing tool is a submodule of the system analyzer and relies on
deep knowledge representation. The system has been implemented for the
English language but can be extended to other languages.
Vitali Specht. An MTT-Based Parser for German.In
this paper, we present a MTT-based parser for German. The parser
analyses German sentences from the surface-morphological to the
surface-syntactical level of the MTT. The parsing process is divided
into three stages. In the first stage, a sentence is segmented into
topological fields. In the second stage nominal and prepositional
phrases are recognized in the so far received fields, before in the
third stage the full parse tree is built. We use a topological
representation of the sentence on the deep-morphological level of the
MTT.
Jacques Steinlin, Sylvain Kahane, Alain
Polguère. Compiling a "Classical" Explanatory Combinatorial
Lexicographic Description into a Relational Database.This
paper deals with the architecture of a dictionary based on the formal
description of collocations and the semantic derivation. The purpose of
our project was to obtain a more flexible format from an electronic
dictionary, the DiCo which has a traditional structure. By compiling
the later, we get an object model which in turn can be stored in a SQL
database. We show how the object model allows us to access the data in
a easier way and gives us some new viewpoints on the dictionary. Thus,
it can be completely reorganized following new axles like lexical
functions.
Vladimir Trub. Ïðîòèâîïîñòàâèòåëüíàÿ ñòðóêòóðà êàê ñðåäñòâî ñåìàíòè÷åñêîé èíòåðïðåòàöèè îòðèöàíèÿ
(Contrastive Structure as a Means of Semantic Interpretation of Negation).The
paper deals with a variety of aspects of interaction of negation with
other parts of the utterance. The contrastive structure like not P, but Q
is considered to be a universal means of interpretation of negation in
the function of negative evaluation of the truth of the utterance. The
different types of implicit oppositions, peculiar to negative sentences
as well as to the meanings of beginning and cessation are being
examined.
Agnès Tutin. Annotating Lexical Functions in Corpora: Showing Collocations in Context.Within
the class of frozen expressions, collocations present remarkable
linguistic properties, since they are neither completely frozen nor
fully compositional. Using large corpora and NLP robust tools may help
acquire such syntactic properties which could be integrated in a
learner·s tool. We experiment automatic collocation annotation by
adapting Polgu?re·s DiCo combined with patterns described with finite
state transducers.
Elena Uryson. Ôðàçîâûå ÷àñòèöû, ñîþçû è òåîðèÿ âàëåíòíîñòåé
(Connective Particles, Conjunctions and the Theory of Valence).Connective particles like òåì íå ìåíåå
have two semantic actants. The problem is that strict linguistic rules
are sufficient for specifying the expression of one of the actants, but
regulate the expression of the other only partially. One cannot
represent this case in the context of any formal syntactic model. The
treatment of such connective particles offers a clearer view of
syntactic scopes of a lexeme.
Leo Wanner, Margarita Alonso Ramos. What Type of Entity is a Lexical Function?Lexical
Functions are a universal means for modelling restricted lexical
co-occurrence and semantic derivation. However, in MTT, lexical
functions are interpreted in several different ways:
(a) as a means to encode co-occurrence and derivation,
(b) as a deep lexical item,
(c) as a lexicalization device. This may lead, on the one hand, to
confusion and improper interpretation. On the other hand, this results
in a flawed theoretical model. We propose a differentiated notation of
LFs, giving examples and pointing out the advantages this notation has
for the formal description of an MTM and for NLP.
Victor Xrakovskij. The Diatheses and Voices Concept in Early XXI Century.The
paper reviews the development of the Diatheses and Voices Concept
during the 35 years of its existence (1970–2005). The primary focus is
on the basic postulates of the concept which, in the author·s opinion,
are not always observed in practical language descriptions. The aim of
this publication is to give a detailed analysis of such deviations and
outline ways to avoid them.
Tatiana Yanko. Èíòîíàöèÿ èëëîêóöèé. Ãðàììàòèêà è ñëîâàðü
(Intonation of Illocutions. The Grammar and the Dictionary).Elements
of semantic representation of illocutionary meanings have been
described. It has been demonstrated that certain meanings are
compatible with other meanings and form an illocutionary grammar of
language, while certain other elements of semantic representation do
not combine with any other semantic units and demand for individual
representation in an illocutionary vocabulary. Intonational means of
expression for both types of illocutionary meanings has described. The
main material for the analysis is the Russian language. The elements of
the semantic representation proposed for Russian are used in the
analysis of English, French, and Danish.
Zdenek Žabokrtský. Resemblances between Meaning Text Theory and Functional Generative Description.Functional
Generative Description (FGD) is a stratificational dependency-based
approach to natural language description, which has been developed by
Petr Sgall and his collaborators in Prague since 1960·s. Although FGD
bears surprisingly many resemblances with the Meaning-Text Theory, to
our knowledge there is no reasonably detailed comparative study
available so far. In this paper, we try to point out at least the basic
obvious parallels, which, in our opinion, remain generally unknown.
Ñìîòðèòå òàêæå:  Ïðîãðàììà êîíôåðåíöèè (ïî-ðóññêè) (îïóáëèêîâàíî 20.06.2005)
|
|
|
|
