The premise that idioms are an important feature in second language learning and teaching is what underpinned the present study. The aim is to investigate what teaching method is the most effective for idiom retention; etymological elaboration or context elaboration. It is a small scale study run focusing on two groups of 9th grade students were approximately 80% of the students had Swedish as their mother tongue, whilst the remaining 20% had other languages. One group of 18 students where taught 15 opaque idioms. The idioms were presented with their etymology. The preference group consisted of 19 students and the same idioms were presented to this group in context. A pre-test was given to both groups in order to establish what idioms they already knew. A post-test was run immediately after the ecture, where the idioms were presented either in context or with their etymology, in order to determine the methods’ effect on immediate retention. After three weeks, a second post-test was run in order to discover the degree to which the idioms had reached the students’ longterm memory and compare the two teaching techniques accordingly. A questionnaire was also conducted in order to gauge out the students’ idiom awareness and to what degree they believed that the teaching method helped them to remember the idioms. The results of the study show that both teaching techniques are beneficial on idiom retention. Context elaboration, however, turned out to be most effective on immediate- and long-term retention.

L’objectif de ce mémoire est de vérifier si la dernière réforme de l’Académie française (1990) a changé l’orthographe des mots, afin de les rendre moins étymologiques au profit d’une orthographe plus phonétique. Au XIVème siècle, des lettres étymologiques ont été insérées dans la langue française pour lui conférer une stabilité. Dès la naissance de l’Académie française en 1694, celle-ci a voulu les préserver. Mais depuis 1990, elle est accusée d'appauvrir la langue française parce qu’elle aurait ôté ces lettres étymologiques muettes des mots réformés. Par l'intermédiaire d’un corpus dictionnairique (CNRTL, 1-9 éditions de l’Académie) nous évaluerons la véracité de ces accusations en analysant une dizaine de mots rectifiés, en guise d’exemples, pour examiner si une perte étymologique est constatée. Cette méthodologie se rapproche de celle d’André Goosse dans La « nouvelle » orthographe : Exposé et commentaires. Suite à l’analyse nous pouvons en déduire que 50% des mots étudiés ont perdu des lettres étymologiques pour une graphie plus phonétique. Étant donné que ce pourcentage est élevé et que ces lettres n’ont plus de fonction linguistique, l’hypothèse selon laquelle l’Académie s’est débarrassée d’un maximum des lettres étymologiques est plausible mais une étude sur l’ensemble des 2 400 mots réformés est nécessaire pour la vérifier. / The objective of this study is to verify whether the last reform of the French Academy (1990) changed the spelling of words in order to make them less etymological in favor of a more phonetic spelling. In the 14th century, etymological letters were inserted into the French language to give it stability and since the creation of the French Academy in 1634, it has wanted to preserve them. But since 1990 the Academy has been accused of impoverishing the French language because it would have removed these silent etymological letters from reformed words. Through a dictionary corpus (CNRTL, 1-9 éditions de l'Académie), we will assess the veracity of these accusations by analyzing ten rectified words, as examples, to examine whether an etymological loss is noted. This methodology is similar to that of André Goosse in La « nouvelle » orthographe: Exposé and commentaires Following the analysis we can deduce that 50% of the words studied have lost etymological letters for a more phonetic spelling. Given that this percentage is high and that these letters no longer have a linguistic function, the hypothesis according to which the Academy got rid of a maximum of the etymological letters is plausible but a study on all 2400 words reformed is necessary to verify it.