Online text and concordancer sites for French

 Compiled by Betsy Kerr, University of Minnesota, <bjkerr@umn.edu>

Written texts only:

1. 'The Compleat Lexical Tutor': UQAM (Université de Québec à Montréal) Web Concordancer. http://www.lextutor.ca/concordancers/ Easy-to-use and fairly versatile concordancer, with a choice of several different corpora, including all of Le monde from 1998. Maximum hits: 2001.  Includes composite written and spoken corpora of equal size (Français parlé and Français écrit) which are useful for frequency comparisons.  Allows consultation of larger text segments. 
   

2.  Lexiqum: Concordancier québécois de RALI (Recherche appliquée en linguistique informatique, U. de Montréal).  http://retour.iro.umontreal.ca/cgi-bin/lexiqum  Maximum of 500 examples from a corpus of written texts of Quebec French of various genres.  Be sure to read the page 'Aide' for more information on the corpora, and for more search options.  Maximum context size = 200 words. 

3. ARTFL.

http://humanities.uchicago.edu/ARTFL/ARTFL.html Project for American and French Research on the Treasury of the French Language, University of Chicago. A huge searchable database of texts from 15^th-20^th century French literature, philosophy, arts, sciences. (U of M students can access this controlled-access site through the U of M Librairies' site.) Can do simple or sophisticated searches, but requires a little learning.

Spoken corpora only:

1. ELICOP. http://bach.arts.kuleuven.ac.be/elicop/ Etude LInguistique de la COmmunication Parlée, Département de Linguistique, Université Catholique de Louvain (Belgique). Concordancer and extensive transcripts of several oral corpora, notably 80 hours of the Orléans corpus (1968-71). This and the other sub-corpora of the ELILAP corpus are all spontaneous conversation. The LANCOM corpus presents smaller samples of role-plays by Flemish-speaking learners of French (27 hrs.), and a very small sample (3-4 hrs.) of the same by Belgian Francophones and French Francophones.

Go to 'Consultation des corpus' for complete instructions on how to use the concordancer. 
Use the link 'Recherches KWIC' to obtain a simple KWIC (KeyWord In Context) concordance, i.e. a list of occurrences with just one line of context (maximum hits: 500).
The link entitled 'Nouveau formulaire de recherches' allows one to set the length of context accompanying each occurrence, as well as other settings (maximum hits: 999). 
From the homepage, click on 'Corpus étiqueté' ('Tagged corpus') for another search engine that allows one to specify the syntactic category of each word in a string of words to be searched for. Easy to use. Maximum hits: 999.
All three of these search engines allow the use of what are called  'regular expressions'; see instructions for their use on the 'Help' page accessible from the 'Nouveau formulaire de recherches'. 

2. DELIC.  http://www.up.univ-mrs.fr/delic/crfp  Corpus de Référence du Français Parlé.  Project of the Equipe de recherche DELIC (DEscription Linguistique Informatisée sur Corpus) under the direction of Jean Véronis at the Université de Provence.  This is a demonstartion concordancer, which will give a maximum of 300 examples from the CRFP corpus. The corpus includes a variety of different oral genres, varying in degree of formality. For more information about the corpus, see the related pages: Composition, Enregistrement, and Transcription (cliquez dans le menu à gauche). 
   

NOTE:  The author of this site will make available to other researchers, on request, digitized text and audio files of the Minnesota Corpus, a 10-hour corpus of spontaneous conversation by two groups of three native speakers.  Write to Betsy Kerr at bjkerr@umn.edu. 

Updated, March 2009