Methodological framework

  • Home
  • Methodological framework

The resources that form the basis of the present classification have been developed using the methodological framework of Lexicon-Grammar (LG) theory (Gross, 1975) based in Harris distributional approach (Harris, 1946; 1955; 1968; 1976). The latter advocates the identification of syntax with regularities of components’ distributional properties leading to the first typical use of transformational mechanisms. Being a model of syntax limited to the elementary sentences of a natural language, the theory argues that the unit of meaning is not located at the level of the word, but at the level of elementary sentences of the form Subject – Verb – Object. To this end, the elementary sentence is being transformed to its predicate-argument structure, and the main complements (subject, object) are separated from other complements (adjuncts) on the basis of formal criteria.

Distributional properties associated with words, i.e., types of prepositions, features (i.e. human, -human) attached to nouns in subject and object positions are also taken into account, resulting in a more fine-grained classification, and to the creation of homogenous word classes. Finally, transformation rules as pronominalisation, construed as equivalence relations between sentences, further generate equivalent structures as the equivalence relation constitutes the substitution of a word from an equivalent one.

All this information is formally encoded in the so-called Lexicon-Grammar (LG) tables. Each table is defined by a set of distinct properties (syntactic, distributional, and semantic) and includes all lexical items sharing these characteristics; entries in one table are therefore, considered to form a homogenous class. Verbs with more than one usage or meaning, have been treated as separate lexical items (possibly represented in different tables) and syntactic and semantic properties are assigned to each meaning thereof.

A set of features that are appropriate for the description of a syntactic category concerning grammatical (i.e., past participle form) or syntactic information (i.e., passive transformation, causative-inchoative alternation, etc.) and lexical choice (i.e., preposition a given predicate is subcategorized for) is applied to all entries under consideration, and their linguistic validity is checked. At the intersection of a row corresponding to a lexical item and a column corresponding to a feature, we set to ‘+’ if it is valid or ‘-‘ if it is not.

Therefore, MWEs are treated as

  • elementary sentences (Subject – Verb – Object), in which the formalism provides the mechanism for encoding uniform features and processing MWEs that are represented as Part-of-Speech sequences e.g Ν denotes a non-fixed nominal, whereas C signifies a fixed one (see Symbol Table)
  • structural patterns with distributional properties, selectional restrictions over the non-fixed or variable elements as well as syntactic phenomena (i.e., clitic and passive alternation, etc) that are encoded

The MWE in (1) below comprises two fixed elements, a verb, and a noun; two variable elements, namely a nominal phrase in accusative and a possessive pronoun (Poss) that modifies the fixed nominal constituent; agreement between these variable elements is mandatory:

με πιάνουν τα διαόλια μου/*σου / *του Γιάννη

me pianun ta diaolia mu/*su / *tu Jani

me1sg.acc catch.3pl the.pl.nom devils.nom.pl my.1sg.poss / * your.1sg.poss / *of-the    John.sg.gen

„my/*your /*John’s devils catch me” (=„to become very angry”)

In this case, a generic syntactic pattern (Ppv V C0 Poss) is used to describe a class in a LG table. An example of MWE representation in accordance with the LG framework (Fotopoulou, 1993), that corresponds to the structure N0 V Prep C1 is depicted in (Table 1).

It becomes evident that the LG framework leads to a uniform and consistent description of elementary sentences and the formal encoding of properties across languages in a comparable manner. In this respect, one of the main advantages of LG is that it allows comparisons between languages and facilitates the construction of cross-language resources.

Relevant bibliography

  1. Constant, M. & Fotopoulou, A. (2021). Verbal multiword expressions: a preliminary study on the fixedness degree, application to Modern Greek and French. In Gavriilidou, Z, Mitits L., Kiosses S., 2021, Proceedings of XIX EURALEX Congress: Lexicography for Inclusion, Vol. II, Democritus University of Thrace.Pages-from-EURALEX2021_ProceedingsBook-Vol2-p713-719.pdf (euralex2020.gr)
  2. Fotopoulou, A. & Giouli, V. 2018. MWEs and the Emotion Lexicon: typological and cross-lingual considerations, in M. Seiler & S. Markantonatou (eds). Mutliword Expressions: Insights from a Multi-lingual Perspective). Berlin : Language Science Press series. Em­pir­i­cal­ly Ori­ent­ed The­o­ret­i­cal Mor­phol­o­gy and Syn­tax.)
  3. Fotopoulou, A and Giouli, V. 2016. “MWEs: Support/light Verb Constructions vs Frozen Constructions in Modern Greek and French Patterns”. In Gloria Corpas Pastor, Johanna Monti, Violeta Seretan, Ruslan Mitkov (eds.) Workshop Proceedings Multi-word units in Machine Translation and Translation Technologies – MUMTTT2015, 1-2 July 2015, Malaga Spain, Geneva, Editions Tradulex, 2016
  4. Φωτοπουλου, Α. & Μίνη, Μ., 2007. «Τυπολογία των πολυλεκτικών ρηματικών εκφράσεων στα λεξικά της Νέας Ελληνικής: όρια και διαφοροποιήσεις», Proceedings of International Symposium on Theoretical & Applied Linguistics (ISTAL) 2007, Θεσ/κη.
  5. Φωτοπούλου, Α. 1993. Ανάλυση των συστατικών των στερεότυπων προτάσεων στα νέα ελληνικά. Παρατηρήσεις σχετικά με την ταξινόμησή τους. Μελέτες για την ελληνική γλώσσα, Πρακτικά της 14ης Συνάντησης του Τομέα Γλωσσολογίας της Φιλοσοφικής σχολής του ΑΠΘ, Θεσ/κη
  6. Fotopoulou, A.,1997. L’ordre des mots dans les phrases figées à un complément libre en grec moderne, in La locution: entre lexique, syntaxe et pragmatique, Publication de l’INALF, Collection “Saint-Cloud”, Kliencksieck, Paris (7).
  7. Fotopoulou, A., 1993. Traitement du cas génitif dans une classification des phrases à compléments figés du grec moderne, Lingvisticae Investigationes, no XVII: 2, John Benjamins, Paris, France.
  8. Fotopoulou, A., 1993. Une classification des phrases à compléments figés en grec moderne. Etude morphosyntaxique des phrases figées, (Ταξινόμηση των προτάσεων με στερεότυπα συμπληρώματα στα Νέα Ελληνικά – μορφοσυντακτική μελέτη των στερεότυπων προτάσεων), Παρουσίαση του Διδακτορικού, Lingvisticae Investigationes, no XVII: 1, John Benjamins, Paris, France
  9. Fotopoulou, A. 1993. Une classification des phrases à compléments figés en grec moderne – étude morphosyntaxique des phrases figées. Thèse de Doctorat. Université Paris VIII – St. Denis, Paris.
  10. Gross, M. 1982. Une classification des phrases “figées” du français. In Revue Québecoise de Linguistique 11 (2): 151-185.
  11. Gross, M. 1988. Les limites de la phrase figée. In Langages 90 : 7-22. Paris: Larousse.
  12. Gross M. (1988b). «Sur les phrases figées complexes du français». Langue française 77: 47-70.
  13. Gross M. (1989a). «Les expressions figées». Στο Programme des Recherches Coordonnées: Rapport no 8. Paris.
  14. Gross, M. 1975. Méthodes en syntaxe. Paris: Hermann.Gross, Μ. 1982. Simple sentences. Discussion of F.W. Householder’s contribution, http://infolingu.univ-mlv.fr/english/Bibliographie/Simple%20sentences.zip
  15. Gross M. (1981a). «Les bases empiriques de la notion de prédicat sémantique». Langages 63: 7-51.
  16. Gross M. (1981b). «La formalisation des langues naturelles». Pour la Science 47: 96-104.
  17. Harris Z.S. (1964). «The Elementary Transformations». Στο Z. Harris (1970) Papers in Structural and Transformational Linguistics, Dordrecht: Reidel.
  18. Harris Z.S. (1976). Notes du cours de syntaxe. Paris: Editions du Seuil.
  19. Mini, M. 2009. Linguistic and Psycholinguistic study of fixed verbal expressions with fixed subject in Modern Greek: A morphosyntactic analysis, lexicosemantic gradation and processing by elementary school children. Unpublished Doctoral Dissertation, University of Patras.