DI-UMONS : Dépôt institutionnel de l’université de Mons

Recherche transversale
(titres de publication, de périodique et noms de colloque inclus)
2021-09-10 - Colloque/Présentation - communication orale - Anglais - page(s)

Marion Jesse , "Almost all and presque tout: A corpus-based study of quantity modification with English all and French tout" in Using Corpora in Contrastive and Translation Studies (6th edition), Bertinoro, Italie, 2021

  • Codes CREF : Linguistique appliquée (DI5320), Langue et littérature d'expression anglaise (DI536F), Langue et littérature françaises (DI5365), Linguistique cognitive (DI429A), Traduction (DI5326), Syntaxe (DI5314), Linguistique générale (DI5310)
  • Unités de recherche UMONS : Traductologie (T202)
  • Instituts UMONS : Institut de recherche en sciences et technologies du langage (Langage)
  • Centres UMONS : Ciéphumons (CIEPH)
Texte intégral :

Abstract(s) :

(Anglais) Whereas many have investigated the field of quantifiers (e.g. Barwise & Cooper 1981; Gärdenfors 1987; Langacker 1991, 2008; Doetjes 1997; Benninger 1999; Radden & Dirven 2007), few have ventured into that of quantity modification, i.e. the modification of quantifying expressions, as in (1-2) below. Examples (1) and (2) respectively feature the relative quantifiers all and tout, which specify a predicated quantity P in relation to a reference mass RT which “consists [by default] of the maximal instantiation [T] of the pertinent category”, e.g. all students in which P = RT (Langacker 1991: 82-83, 86). In (1) and (2) English all and French tous are modified by almost and presque respectively, which indicate that the boundary of the predicated mass P approximates that of the reference mass RT but that P and RT do not completely coincide. (1) That's why I use him in almost all my answers. (YCCQA_uk) (2) (…) Beaucoup font aussi la lecture de vidéo, comme VLC qui peut lire presque tous les formats vidéo existant (…) (YCCQA_fr) ‘(…) Many can also read videos, such as VLC which can read almost all existing video formats (…)’ This study focuses on one type of quantification, relative quantification, and explores the possible ways in which relative quantifiers can be modified in terms of quantity. The two quantifiers under scrutiny are English all and French tout (in all its inflected forms). The contrastive set-up of the study will allow us to investigate any cross-linguistic differences in the quantity modification potential of relative quantifiers. Building on Paradis (1997, 2000, 2001), Njende et al. (2015, 2017) revealed co-selection restrictions between specific modifier and quantifier types. They found that absolute quantifiers like many and few take scalar modifiers, which modify – upwards or downwards – the range expressed by the quantifier on an implied open quantification scale (e.g. very many, rather few). Relative quantifiers such as all or none take proportional modifiers, which compare the actually predicated quantity or mass to the reference mass, indicating whether it either approximates (e.g. almost all/none) or reaches it (e.g. absolutely all/none). This study aims to verify Njende et al.’s (2015, 2017) findings for all and complement them with contrastive data for French tout. This will contribute to inventorying the different types of quantity modification the items under scrutiny allow. English all and French tout were selected for this study as they can be modified by a variety of adverbs and are relatively frequent in everyday speech. Data were extracted from a multilingual lower-register written corpus and two spoken corpora. Standard corpora such as Collins Wordbanks Online for English and Frantext for French returned too few instances of quantity modification. As quantity modification is in many respects similar to degree modification, lower-register and spoken corpora were consulted as they have proven to be fruitful sources of data on degree modification. The first corpus consulted is the 29-million-word Yahoo-based Contrastive Corpus of Questions and Answers (YCCQA, De Smet 2009), which is available in English, French, German, and Spanish, and covers the time period 2006 to 2009 (De Smet 2009). Datasets of 250 instances were extracted with AntConc (Anthony 2010) from both the English and French YCCQA subcorpora, provided that sufficient data was available. The English YCCQA data are supplemented with another 250 instances taken from the British National Corpus 2014 Spoken (BNC2014, Love et al. 2017), accessed through SketchEngine. The BNC2014 totals 10,495,185 words from 1,251 conversations by 672 speakers between 2012 and 2016. The French YCCQA data are supplemented with spoken data from both the Corpus de Français Parisien Parlé des années 2000 (CFPP2000, Branca-Rosoff et al. 2012), which contains 744,159 words from circa 58 hours of audio files, and the corpus Traitement de Corpus Oraux en Français (TCOF, “Analyse et traitement”), which totals 1,542,562 words from circa 146 hours of audio files. On the basis of the data selected, a monolingual and contrastive analysis can be performed of all and tout in informal registers in written language, and a monolingual study can be carried out for each quantifier in spoken language. Particular attention will go to the contexts in which quantity modification emerges, as well as to the (in)animate nature and semantics of the nouns quantified. Preliminary data analysis seems to indicate that quantity modification is rarer in French than in English. This may find an explanation in the considerable number of instances of not all N in the English data. Not all typically has scope over the subject of the sentence, as in (3). The French data returned no such uses of pas tout. This could be linked to the position markers of negation favour in the sentence in the languages under study. (3) Not all white people are like that. (YCCQA_uk) Therefore, we posit that quantity modification is overall more frequent in English than in French, and that negative quantity modifiers (such as not or pas) are common in English only. In addition, preliminary results seem to suggest that animate nouns modified by one of the quantifiers under study most often serve as subjects in the sentence (as in (3) above). Quantified inanimate nouns, by contrast, act most often as direct objects, as in (4). (4) You had almost all the advice you need by fellow Italians. (YCCQA_uk) This study will thus allow to catalogue the quantity modifiers for all and tout in informal and spoken registers, will contribute to a better understanding of quantity modification in general and of its relation to the type of quantification involved, and will compare and contrast how all and tout in particular undergo quantity modification. More precisely this study will explore the semantics of the nouns quantified and their possible influence on the choice of quantifier, as well as the frequency to which proportional and totality modifiers are involved in the quantity modification of the relative quantifiers all and tout.