Associative Semantics & Meta-Informative Centering Theory

   The Associative Semantics (AS) theory and the Meta-Informative Centering (MIC) theory modify and - in a consistent way - unify two sorts of problems: Predicate Argument Structure (models of semantic relations as expressed by natural languages but with some confusion concerning the proper differentiation of semantic and syntactic domains) and the theories of Information Structure (in fact, concerning the presentation of informative contents as conveyed by linguistic utterances but leaving aside an important part of meta-informative devices of natural languages which are usually treated as belonging to the the more or less formal syntax).

   The ASMIC theories are ontology-based language-dependent approaches. Thus they can be used to identify similar informative and meta-informative contents which are conveyed by expressions of different languages in various morpho-syntactic and prosodic forms.

Essentials of Associative Semantics (AS) Theory

   The semantic situation is defined as a schema and its individuation which contain three parts: relation(s), role(s) and anchor(s). The individuation of  relations, roles and anchor are named frames, participants and supports respectively. NB: the frames of semantic situations are closed with respect to inclusion and can be ordered (state < event < ordinary process < granular process).

   We distinguish various abstract, ontological levels. Instead of using such concepts as agentive, objective, experiencer, etc., on the highest (meta-ontological) level, we postulate generalized agents which are defined by the following ontological features among others: (1) features of control (autonomy): goal and feedback; (2) features of emotion (character): desire and intention, (3) epistemic features (reason): belief and cognition, (4) communication features (language faculty): verbal and visual.

   In  this theory, roles and anchors are seen as isomorphic abstract tuples of concepts. Roles are defined ontologically as neutral, active and passive unary relations of associated semantic situations. The associative combination of their realizations gives rise to median (instrumental, means etc.) roles through derivation. Here are examples of agent meta-roles: (1) active role (Initiator, Causer, Enabler, Benefactor, Executor, Stimulant, Source, Instigator etc.), (2) passive role (Terminator, Causee, Affect, Enabled, Beneficiary, Executed, Experiencer, Goal, etc.) and (3) median role (Mediator, Instrument, Benefice, Motor, Mean etc.).

   Importantly, the same participant (role filler) may play more than one role in the given semantic situation defined as a bunch of associated situations. Note also that in the process of meaning construction the three following levels are distinguished: shallow, standard and deep. There is an important feature of situation participants which characterizes the shallow level: in active roles, the inanimate entities (figures as opposed to agents) are semantically interpreted as if they were animate. Such figures are said to play quasi-active roles (Q-roles).

Essentials of Meta-Informative Centering (MIC) Theory

   Meta-information concerns the way information is presented: to achieve the ordering of non-linear representations in the form of texts (sequences of linguistic utterances), the speaker must select a centre of attention (CA) and “predicate” about it. Hence, in our centering theory, the term “predicate” is reserved for the meta-informative level.

   The main concept introduced here as a terminological device is the Centre of Attention (CA) which refers to meta-information as conveyed by utterances. Global and Local CAs are distinguished on the three following levels:

(a) Subject and Object (in base utterances),
(b) Topic and Focus (in extended utterances) and
(c) General Themes and Particular Themes (in discourse sessions, dialogues or texts).

Note that meta-information is more appropriate than information to describe the base and extended utterances.