Diachronic study of the use, semantics and discourse values of a number of simple prepositions from Old French (9th century) to the present.
In the course of the past twenty years the literature on prepositions has contributed to a renewed approach to their syntax, semantics and distribution. Yet to date no statistical analyses have been carried out on large corpora to detect diachronic variations in the prepositions’ preferred combinations. This is the general goal of the present project. <br />This general goal breaks down into more specific ones: i) improve our knowledge of preposition combinations in French, from Old French (9th century) to the present; ii) observe and describe the emergence, reorganisation and fall of set structures; iii) draw relationships between the variations in frequency of preposition combinations, the search for added semantics or discourse effects, and a given society’s communication needs; iv) draw relationships between some reorganisations in the linguistic system and extralinguistic issues; v) extend the notion of combination profile to patterns of co-occurrence in larger textual units. <br />The project is also committed to making as accessible as possible, through downloading via an open licence, the following elements, which will be set up or improved thanks to the project: lexicons and learning corpora, natural language processing software and associated resources, software for corpus analysis. <br />
For the 9th-20th centuries, the method used will combine quantitative and qualitative approaches to corpora, with a study of the diachronic evolution of the combination profile of the French prepositions à, en, par, contre, dès, devant, entre, pour, sans, sur, sous, vers and dans. It will more specifically seek to make out the highly specific combinations of these prepositions. Among the significant combinations which statistical methods will enable us to establish there will be more or less fossilised prepositional phrases. One goal will be to observe their evolution closely: emergence and evolution of fossilisations, semantic evolution, changes in their preferred immediate environment within sentences, changes in position, etc.
The notion of combination profile will also be extended to patterns of co-occurrence in larger textual units. The study will draw on the theoretical frameworks of lexical priming (Hoey 2005) and discourse frames (Charolles 1997).
Another innovative aspect of the project will be its search for a relationship between the evolution of fossilised structures (Durkheim’s ’social facts’) and other social facts of an extralinguistic nature. It will try to go beyond the contrast between ‘internal history’ and ‘external history’ and propose a middle way, with influences from both kinds of factors.
A one-day conference will be held in 2014 in Cologne. An international symposium with a call for papers will be organised in Lyon (Université Lyon 2 & ENS-Lyon) in 2015, marking the end of the project. A collective book will be published, based on the research carried out within the Presto project.
The results from linguistic analyses will be made accessible to the general public on the project’s website via an open-access query interface and downloadable SQL sources with a licence of the Creative Commons type.
Software developments carried out within the TXM platform will be made available as open source using the free software licence GPL v.3. from the free software website SourceForge.
Get an innovative view of the historical evolution of prepositions thanks to analyses which combine two sets of methodological approaches: (1) a study of the interdependence between the quantitative and qualitative dimensions (uses and values of prepositions respectively); (2) a study of the principles which, in the course of time, led to a search for added semantic and discourse values - such a search being a true driving force of language change.
Publications by 1st October 2013 (TO + 6 months):
Blumenthal, P. (forthcoming 2014), « Caractéristiques et effets de la complexité sémantique de noms d’affect » in Nouvelles perspectives en sémantique lexicale et en organisation du discours,
The aim of this project is a diachronic study of the uses, semantics and discourse values of the French prepositions à, en, par, contre, dès, devant, entre, pour, sans, sur, sous, vers and dans from Old French to the present.
The study, which will be instrumented and based on a statistical and distributional approach, will bear on the variations in the combinational behaviour of prepositions depending on date (diachronic evolution), genre and author. One area of special interest will be the detection of the highly specific combinations into which the prepositions enter. We will study the rise, reorganization and fall of fossilized prepositional structures. We expect to get an innovative view of diachronic evolution thanks to a model of analysis based on two sets of methodological procedures: (1) a study of the interdependence between the quantitative (uses of prepositions) and qualitative (values) dimensions, and (2) a study of the principles which have underlain the search for added semantic and discourse values in diachrony – a veritable driving force of language change
One of the requisites of the project – which will also be one of its major contributions – will be the construction of an annotated and lemmatized corpus spanning every stage of the history of French and balancing genres and text types. This corpus, set up jointly by the French and German teams, will be based partly on resources already available in Lyon (mediaeval era), Tours (16th century), Nancy (ATILF) and Cologne (modern French, 17th - 20th centuries). A proactive approach will be taken to give open access to as much of the corpus as possible.
Monsieur Denis VIGIER (Interactions Corpus Apprentissages Représentations) – firstname.lastname@example.org
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
Universität zu Köln Département des langues romanes de l'Université de Cologne
ICAR Interactions Corpus Apprentissages Représentations
Help of the ANR 238,135 euros
Beginning and duration of the scientific project: March 2013 - 36 Months