Although secondary school L2 learners might represent the largest group of foreign language learners in developed countries, researchers have so far given very little attention to this type of population (Collins & Muñoz, 2016), with regards to their grammatical development within the CEFR task-based approach. The present study will explore this issue within the frame of spoken peer-interactions, among 14-year-old French pupils in their fourth year of studying English as a foreign language. The data was collected over a school-year span. Forty-eight students were asked to perform an information-gap task (Ellis, 2003) in pairs, at the beginning and at the end of the school year. They were video-recorded and the interactions were transcribed using CLAN software (MacWhinney, 2000) to analyse their language development. We showed (Manoïlov, 2017) that allowing strategic planning greatly enhances young learners’ performance in peer-interactions in terms of fluency (as measured by the number of tokens and the number of turns per minute), between the first collection of data and the last one, showing a trend for improvement. In this study, we will focus on the capacity of the learners to refer to past events, and look more in detail into the data in order to characterise how reference to the past is construed and constructed by students. We coded the transcriptions with a semantic oriented set of parameters to outline all the linguistic markers the learners use when they speak about the past. This coding was devised within a twofold theoretical framework: discourse analysis and enunciative linguistics. Our first findings suggest a classification from an exclusive use of adverbials and lexical items interspersed with a few occurrences of was, to their using the preterit, first with irregular and then with regular verbs, according to the levels of the students. With this work, we contribute to the sketching of the grammatical abilities of students engaged in spoken interactions, at a given stage of their learning process. We then discuss these findings in relation to the French curriculum and gauge their implications for second language research and teaching in secondary schools.