Igor Mel’čuk

Communicative Organization in Natural Language

The book defines the concept of Semantic-Communicative Structure [= Sem-CommS]-a formal object that is imposed on the starting Semantic Structure [= SemS] of a sentence (under text synthesis) in order to turn the selected meaning into a linguistic message. The Sem-CommS is a system of eight logically independent oppositions: 1. Thematicity (Rheme vs. Theme), 2. Givenness (Given vs. Old), 3. Focalization (Focalized vs. Non-Focalized), 4. Perspective (Foregrounded vs. Backgrounded), 5. Emphasis (Emphasized vs. Non-Emphasized), 6. Presupposedness (Presupposed vs. Non-Presupposed), 7. Unitariness (Unitary vs. Articulated), 8. Locutionality (Communicated vs. Signaled). The values of these oppositions mark particular subnetworks of the starting SemS and thus allow for the distinction between sentences such as (a) A man killed a dog vs. The dog was killed by a man, (b) John washed the window vs. It was John who washed the window or (c) It hurts! vs. Ouch! The proposed Sem-Comm-oppositions are conceived as an attempt at sharpening the well-known notions of Topic ~ Comment, Focus, etc. Possible linguistic strategies for expressing the values of the Sem-Comm-oppositions in different languages are discussed at some length, with linguistic illustrations.

Verbes supports sans peine

The paper is aimed at defining the concepts needed in the discussion of so-called ‘support (≈ light) verbs’ and presenting a way of describing them in the lexicon in terms of Lexical Functions [= LFs]. It develops the following six points :1. A genuine support verb is semantically empty (or ‘emptied’ in the context of its keyword).2. There are just three types of ‘pure’ support verbs — Oper, Func, and Labor — distinguished according to the syntactic role fulfilled by their keyword.3. Two sorts of meanings are often combined with support verbs : phasic meanings (‘begin,’ ‘stop,’ ‘continue’) and causative meanings (‘cause’) ; such a meaning plus a support verb form a complex LF.4. There exist other sorts of meanings (especially, intensification) that can bear on the predicative noun but are expressed together with the support verb : they form, with the latter, a configuration of LFs.5. A family of semantically full collocational verbs show the same syntactic behavior as support verbs : these are called realization verbs.6. Using support verbs and their encoding in terms of LFs, it is possible to construct a universal Deep-Syntactic paraphrasing system. Several examples of DSynt-paraphrasing rules are given.The discussion is carried out based on French.

The explanatory combinatorial dictionary as the key tool in machine translation

Lexical and syntactic mismatches between languages are the main challenge for every translation (especially formidable for Machine Translation; see Mel’cuk & Wanner 2001, 2006). An example of mismatches in English-to-Russian translation: a. The demonstrators were brutally beaten and tear-gassed by the police. b. Rus. Demonstranty podverglis’ zverskomu izbieniju so storony policii; protiv nix byl primenen slezotocivyj gaz lit. ‘Demonstrators underwent bestial beating from_side of_police; against them was applied tear-gas’.