1 Temporal Marking in French Belgian Sign Language: Unmarked Tense, Default Tense and Speech Tense Aurélie SINTE The various ways in which temporal information is expressed in French Belgian Sign Language (LSFB) have not previously been studied. Analyses of a 21-hour corpus including three types of discourse (conversations, narratives and descriptions) have brought three things to light: (1) LSFB draws on the same linguistic resources (timelines, lexical signs, time markers, specific markers and non-manual elements) as other SLs to express time. (2) References to time are not equally distributed among the three types of discourse (they are very present in descriptions, less frequent in conversations though still quite common, and are almost completely absent from narratives). (3) What is traditionally called ‘unmarked tense’ requires a closer look in light of what we observed in the narratives and conversations in this corpus. This traditional notion of ‘unmarked tense’ must be distinguished from the default tense as well as from the time when the conversation took place. Keywords: Time, tense, French Belgian Sign Language (LSFB), narratives, temporal marking, discourse variations, unmarked tense 1. Introduction The results presented in this paper are derived from a PhD dissertation successfully defended in November, 2013 and which is to appear in February, 2015 aux Presses Universitaires de Rennes (Sinte 2015). The main topic of the doctoral research was temporal (and, to a lesser extent, aspectual) marking in French Belgian Sign Language (langue des signes de Belgique francophone - LSFB). This article offers a summary of the principal observations set forth in that dissertation. The analyses are based on a corpus consisting of 21 hours of data including narratives (1h10’), descriptions (3h15’) and conversations (16h35’) (Figure 1). The first series of recordings were gathered in June, 2010. They include a series of interviews all of which followed a single protocol (i.e., all of the participants were asked to carry out the same tasks and answer the same questions.) Two individuals interact under the watchful eyes of three cameras. There is one camera in front of each participant, and the third is set with a wider view to take in the whole scene, capturing both parts of the conversation. The data gathered from the 2010 recordings are drawn from two types of protocol, each of which has provided some 90 minutes of footage. First, the participants underwent a semi-guided interview conducted by a deaf person, in which they were asked both personal and professional questions (with the aim of eliciting meta linguistic elements) 1 . In the second part, a series of visual documents (images, short animated sequences, drawings, etc.) was shown to the same group by means of a PowerPoint presentation. The protocol at work here had been specifically designed to elicit temporal information. The participants were asked to explain or describe in detail what was shown to them. 2 In November 2011, a second round of recordings was added to the corpus. Like the previous protocol, this one was also designed to elicit temporal data. The video element was identical but the protocol was altered. Two series of 21 questions were elaborated and then 1 The protocol used is that of the CREAGEST Project (Cuxac 2008). 2 For a complete description of the protocols, see Sinte 2015.