Search

The aim of this conference is to continue the research that attempts to identify the precise differences between oral and written language, given that both share the same lexicon, the same morphological and syntactic constraints and of course can convey the same semantic content. But if we can tell the same story in oral and written language, we do not tell it in the same way. The linguistic phenomena that distinguish spoken language from written language are to be found in the construction of the utterances, the "predicates". We compare the oral and written reformulations of pupils at the end of primary school who have had to listen to a narrative text that has been read to them. What are the repetitions of the predications of the source text? Do they change the order of the predicates? What is the degree of complexity of the predicates used? What are the reformulation procedures? We present the first results of this study on a narrative sequence., L’objectif de cette communication est de poursuivre les recherches qui tentent de cerner où se situent précisément les différences entre la langue orale et la langue écrite, sachant que l’une et l’autre partagent le même lexique, les mêmes contraintes morphologiques et syntaxiques et bien entendu peuvent véhiculer le même contenu sémantique. Mais si on peut raconter la même histoire à l’oral et à l’écrit, on ne la raconte pas de la même façon. C’est au niveau de la construction des énoncés, « les prédications » que se trouvent les phénomènes linguistiques qui distinguent la langue parlée de la langue écrite. Nous comparons les reformulations orales et écrites d’élèves de fin d’école primaire qui ont eu à écouter un texte narratif qui leur a été lu. Quelles sont les reprises des prédications du texte source ? Modifient-ils l’ordre des prédications ? Quel est le degré de complexité des prédications employées ? Quelles sont les procédures de reformulations ? Nous présentons les premiers résultats de cette étude portant sur une séquence narrative.

International audience

International audience

International audience

International audience; During the late Miocene (~11-5 Ma), atmospheric pCO2may have been 1 to 3 times pre-industrial values1,2, and mean global temperatures were likely 3−4°C higher than today with reduced latitudinal gradients3. Major ecosystem changes that took place during the late Miocene (e.g., ref 4 and refs therein) are consistent with falling global temperatures and decreasing pCO2, but proxy evidence for these physical climate changes remains sparse because few continuous, orbital-resolution palaeoclimate records cover this interval. Recently, geographically widespread and major (~5-10°C) late Miocene sea surface cooling was revealed for the mid-and high-latitudes based on the Uk'37 sea surface temperature (SST) proxy3. However, existing tropical SST records suggest only minor cooling in this interval (1-2°C). Here we present new multi-proxy geochemical records of surface and deep ocean variability from IODP Site U1443, drilled in the equatorial Indian Ocean (5°N, 90°E) during IODP Expedition 353, which spanthe interval 9 to 5 Ma. Surface ocean records from Site U1443 have the potential to resolve SST change at a “warm pool” tropical site, whilst also providing insight into the long-term evolution in the Indian monsoon subsystem. SST estimates derived from Mg/Ca ratios in the surface-dwelling planktic foraminifer Trilobatus trilobusshow ~4°C of cooling (3.8 -2.5 mmol/mol Mg/Ca) between 8 and 6 Ma. Following a SST minimum at 5.8-6 Ma, reconstructed temperatures increase from 5.8 to 5 Ma, consistent with published extratropical SST records. SST estimates combined with paired foraminiferal δ18O measurements reveal a ~1.5‰ decrease in local seawater δ18O between 8 and 6 Ma, and a subsequent increase between 6 and 5 Ma. Consistent with records from other sites, Site U1443 benthic δ18O data show no long-term increase concurrent with the SST decrease. We compare our results with late Miocene simulations using the IPSL-CM5 earth system model, exploring both pCO2and paleogeographic effects on regional paleoceanography.

The key term of this study is “information structures”, which is described by Harris (1988) in the following way: “The components that go into the making of the structure are the components that go into the making of the information” (p. 57). Thus, the information is neither independent of the sentence’s structure nor additional to it, but is an interpretation of it. The information of a sentence is yielded directly from the word choice and the stated constraints, i.e. grammatical relations” (ibid. 76, 82). During language acquisition, a child is exposed to various sequences of words that are not randomly combined. For example, the probability of exposure to the combination the snow is falling is greater than the probability of hearing the combination the sky is falling. However, these word combinations are organized in a manner that implies the application of certain rules such as the order of constituents and morphosyntactic markers. Language acquisition emerges from the production of possible combinations, further complexification of these combinations (a white, dangerous snow has been falling all week) and finally, their organization in a complete sequence of words that we call predication. Therefore, the information structure itself is a predication. We postulate that the child produces utterances by reformulating the predicative utterances that he/she has previously been exposed to ; first in an incomplete form, then in a complete form and finally in a more complex form, In the framework of a cross- linguistic study of late language acquisition that had been conducted for several years (Martinot et al., 2018), it was shown that the reformulation procedures which children implement in carrying out a specific task (immediate retelling of a story that an adult reads to each child) reveal how children adjust their language. In particular, one can see that the children locate different predications that constitute the source text and transform these predications during their retelling. In the vast majority of cases, the information structure is modified although the information remains equivalent. In this paper, we aimed to analyse how 4- and 6-year- old children, speakers of two different mother tongues - Croatian and French - produce information structures when retelling a story. We use the following protocol: an adult reads individually the story Tom and Julie (original French version, 436 words) and Tom i Julija (Croatian version, 341 words) to 15 children aged 4 and 6. The respective children knew that they had to retell the story to the same adult as soon as he/she has finished reading it (Martinot et al., 2018). One of the advantages of this unique procedure is the possibility to compare language among various children (ibid.: 32-34). The analysis of children's productions is conducted as follows. From the following source utterance (sequence 1), two children (Pauline X, and Chloe X) produce different reformulations: Sequence 1: That morning, the teacher arrived in the school playground later than usual. She was holding a little girl, whom nobody had ever seen before, by the hand (translation of the French source text). Pauline X (4): the teacher *comed very late because she went (XXX) she *holded by the hand a little girl (gloss of the child’s production) Chloé X (6): the teacher she arrives more later than usual with a new little girl In order to analyse the information structure used by each child, the source text has to be sequenced into predications: [P1a] the teacher arrived in the school playground [P1b] (the teacher arrived) later than usual [P2a] she was holding a little girl by the hand [P2b] whom nobody had ever seen before Children productions are also sequenced in predications, which are then matched with the source predications: Ex. [P1b] (the school teacher arrived) later than usual Pauline X (4): the teacher *comed very late Chloé X (6): the teacher she arrives more later than usual Hence, it is possible to label the means used by the child to produce the information [P1b]. Many questions arise from this kind of analysis, such as: Is it repetition (REP)? Change of information (CHG)? Is it semantic or interpretative paraphrase (SIP)? Is it an analytical procedure (paraphrase by lexical analysis: PLA, or by syntactic analysis: PSA)? Is it a synthetic procedure (paraphrase by syntactic synthesis: PSS)? Is it a formal procedure (formal paraphrase by transformation: FPT, or restructuration, FPR)? Our hypothesis is that 6-year-olds are able to act on both the lexicon (SIP) and syntactic construction in order to produce information structures that are compatible with the lexicon of the utterance. Unlikely, 4-year-olds do not often modify lexical words. We will observe in detail, from a quantitative and qualitative point of view, the different reformulation procedures that Croatian- and French- speaking children use to retell the story ; for the present analysis, we will select three source sequences: sequence 1, sequence 6 and sequence 11, all of which have been most frequently reproduced. The reformulation procedures that we have listed above make it possible to describe which means, both syntactic and lexical, 4- and 6-year-old children respectively use to reconstruct the information structures that are necessary for the production of the story.

International audience

International audience

International audience

International audience

International audience

Nous posons que l’acquisition de la langue maternelle se réalise à travers des procédures de reformulation qui permettent à l’enfant de passer de la langue source, celle de son environnement immédiat, à la langue qu’il produit (Martinot, Claire et al, 2018). L’enfant a donc une posture active tout au long du processus acquisitionnel : il agit sur les énoncés de la langue source en les transformant. Les transformations-reformulations évoluent suivant l’âge de l’enfant mais sont toujours mises en œuvre selon le même principe reformulatoire. Chaque énoncé reformulé résulte d’un changement qui s’articule à la reprise d’une partie ou d’un aspect de l’énoncé source. Ainsi, l’enfant peut reformuler un énoncé source en modifiant le sens tout en gardant la construction initiale : Laura (2 ; 5) : il y a du soleil > il y a du tonnerre Ou bien en modifiant le lexique ou la construction tout en gardant la même information : Laura (2 ; 5) : il me torde la joue > il me pince la joue un petit peu Nous proposons de décrire un certain nombre de procédures reformulatoires qui illustrent comment l’enfant agit, au niveau sémantique des énoncés qu’il prend pour point de départ, mais également au niveau syntaxique. Parmi les neuf procédures de reformulation que nous avons répertoriées (ibid.), nous en sélectionnons deux : les reformulations par paraphrase analytique définitoire et les reformulations par paraphrase synthétique syntaxique. Les premières permettent aux enfants de définir les verbes complexes en les décomposant (chuchoter > dire) : Enoncé source : Un jour Julie chuchota à Tom (…) Margot (6 ans) : Un jour Julie disa à Tom (…) les secondes permettent d’exprimer plusieurs informations de façon syntaxiquement condensée (est arrivée (…) tenait une petite fille (…) > elle arrive avec une petite fille) : Enoncé source : la maîtresse est arrivée dans la cour de l’école (…) elle tenait par la main une petite fille Amandine (8 ans) : elle arrive avec une petite fille dans la main Nous analyserons ces deux postures reformulatoires auprès d’enfants français et croates de 6, 8 et 10 ans. L’objectif est de montrer où se situent les points communs, vraisemblablement entre les tranches d’âge, et les différences, vraisemblablement du fait des fonctionnements linguistiques différents du français et du croate. Cette analyse est issue d’une recherche de plus grande envergure (ibid.), elle en reprend le même protocole expérimental : une histoire est lue à un enfant à la fois, celui-ci la restitue immédiatement après la lecture. L’analyse linguistique consiste à décrire le mode de passage entre chaque énoncé source et chaque énoncé reformulé

International audience

International audience

International audience

International audience

International audience

International audience

International audience

International audience

International audience

International audience

International audience

International audience