Studiens syfte är att synliggöra hur förskollärare beskriver sitt förhållningssätt och arbetssätt i arbetet med flerspråkiga barn för att stödja och stimulera deras språkutveckling. För att undersöka detta har kvalitativa intervjuer med sex förskollärare genomförts. Förskollärarnas förhållningssätt till flerspråkighet samt vilka metoder och verktyg som används i förskoleverksamheten för att främja flerspråkiga barns språk undersöktes. Upplevda utmaningar och möjligheter som detta arbete skapar beskrevs av förskollärarna. Studien har utgångspunkt i den sociokulturella teorin där utveckling och lärande sker i samspel mellan människor, med hjälp av redskap och med rätt stöttning. Teorin var till hjälp att analysera förskollärarnas uppfattningar och erfarenheter i arbetet med att främja språkutveckling hos flerspråkiga barn. Studiens resultat visar att förskollärarna har ett positivt förhållningssätt till flerspråkighet där majoriteten inte upplever sig själva som flerspråkiga. Dessutom framkommer det att förskollärarna upplever arbete med flerspråkiga barns språkutveckling som utmanade eftersom adekvata kompetenser inom ämnet saknas. Detta arbete beskrivs som en möjlighet där olika användbara strategier och verktyg finns till hjälp. Det lyfts även fram att samverkan med olika aktörer är grundläggande för att kunna stödja och stimulera flerspråkiga barns språkutveckling. Resultatet pekar även på att för att kunna verkställa detta arbete i förskolan behöver förskollärarna få mer kunskap och vägledning. För att barn ska kunna utveckla det svenska språket samt sitt modersmål behöver de rätt stöd, menar förskollärarna. Studien bidrar med en ökad kunskap om vilka strateger, redskap och kompetenser förskollärarna behöver ha för att ge flerspråkiga barn förutsättningar för att utveckla alla sina språk. / The purpose of the study is to make visible how preschool teachers describe their approach and methods while working with multilingual children in order to support and stimulate their language development. Qualitative interviews have been conducted with six preschool teachers in order to achieve this goal. During each interview, the teachers were asked about which methods and tools they applied in activities in order to promote and encourage language development. Perceived challenges and opportunities that this work creates were described by the preschool teachers. This study is based on the sociocultural perspective, in which both learning, and development take place through interaction between people with the help of tools and with the right support. The theory was helpful in analyzing the preschool teachers´ perceptions and experiences in their work. The results of the study show that the preschool teachers have a positive attitude to multilingualism, where the majority do not perceive themselves as multilingual. The results show that the preschool teachers perceive their work with multilingual children´s language development as challenging due to a lack of adequate competence in the subject. This work is also described as an opportunity where various strategies and tools are available to help. In addition, it is highlighted that cooperation with others is fundamental to being able to support and stimulate language development. The result also points to the fact that, in order to carry out this work in the preschool, preschool teachers need more knowledge and guidance. In order for children to be able to develop the Swedish language as well as their mother tongue, they need the right support, the preschool teachers declares. The study hopes to contribute to, an increased knowledge of which strategies, tools and skills are needed in order to create an optimal environment for multilingual development.

In South Africa, the topic of reading has been of great interest on the development of language since it has been estimated that 12 million learners will go through the education system per year. A key concern for all these learners is language. The majority of South African learners do not speak English as their home language, yet the dominant language of learning and teaching (LoLT) is English. It is important for these learners to become fluent in the LoLT as the future of education is dependent upon it. South Africa has been struggling since 2001 to implement a programme in schools that will improve the Literacy abilities of learners. An in-depth study was undertaken to determine whether the THRASS methodology would possibly improve the reading and spelling abilities of Grade 2 English second learners at an inner city school in Pretoria. The research explored the reading and spelling difficulties encountered by these learners. A profile of the Grade 2 English second language learner was compiled. An understanding of how the THRASS programme works has been included. The results from the research indicated that the reading abilities of the Grade 2 learners had improved, however, there was not a significant difference in their spelling abilities. / Inclusive Education / M. Ed. (Inclusive Education)

The research described here was inspired by a national review which concluded that too many children come into our primary schools with unmet speech and language needs (Bercow, 2008). Teachers are in a position to identify language difficulties but many have expressed uncertainty regarding their role in this process. I believed that the LAMP (Linguistic Assessment for Mapped Provision) screen for language and communication which I had developed would help teachers to identify language concerns and would also increase their professional confidence in this complex area. I had developed the LAMP screening instrument and piloted it over a 2 year period prior to this study. It is employed here as a universal screen that is used in a whole school approach in order to enable teachers to identify language need. The use of the LAMP as a universal screen applied to all the children in a school lessened the likelihood of preconceived notions impacting upon teacher’s perceptions of need in the classroom. The LAMP data allowed schools to track the progress of individual children within a class and whole school context. Teachers need to be aware of any pre-conceptions they may have in relation to the performance of children from different socio economic circumstances. The hypothesis that poverty continues to provide the weightiest detrimental effects upon children’s language development was examined and within the parameters of this study was found to be contestable. Study design: A systematic survey was conducted over 4 Primary schools using the LAMP. Rich picture data was accessed from teacher questionnaires and focus groups involving participants from the schools in the study. The repeated measures design provided information on what teachers had learned in the period of reflection between the screenings. The use of a mixed methods repeated measures design helped me to understand what was difficult for teachers and what the teachers felt would help them. The 4 schools in the study were chosen to reflect varied social contexts in order to explore any impacts of SES on the results. Analysis of data: In a repeated measures design, a LAMP screen was completed for every child across the 4 schools by their teachers in February and then June in one school year. Results of screening were analysed and compared on a range of variables using SPSS. Questionnaires were used to collect teacher perceptions before and after using the LAMP screen. Focus groups were held in the schools at the end of the study to add more information on how helpful teachers felt the process had been in raising their awareness, confidence, and skills in the identification of SLCN. Findings: The main trend observed was a decrease in levels of teacher concern related to children’s speech and language needs from the first to second screening survey. Differences were found at a statistically significant level on a range of variables. The expected differences between high and low socio-economic status (SES) schools were not found. Teachers reported increases in their awareness, confidence, and skill in identifying children’s speech language and communication concerns by the end of the study. Some changes to classroom practice were reported. Conclusions: I propose that use of the LAMP screen increased teacher awareness of the nature of language difficulty and that this heightened awareness was a key variable in the observed changes to language concern scores. The LAMP screening process was seen by schools’ staff to have had a positive effect on teacher’s skills and to be relatively easily assimilated into the school system. However some participants identified a number of challenges relating to time constraints and maintaining the use of LAMP as high profile in the context of competing time demands in their schools. It is suggested that EP services would be in a position to support schools to implement and embed the LAMP screening model as part of their Service provision. It is also proposed that economic deprivation or disadvantage did not appear to be the only important factor to consider when making funding decisions intended to support children’s linguistic competency in schools.

Abstract The aim of this study was to research how pedagogues work with communication and technique within language development for children with special needs. It covers children with language disorder and language delay as well as children having a different language than Swedish as their native language. The main questions used are: What means of communication methods are the pedagogues using on their daily work with children with special needs and what methods are used in the preschool to support the children with the language development? What obstacles and possibilities do the pedagogues identify when it comes to use of technical tools and what holdbacks are identified in the preschools when it comes to use of technical tools? The identified conclusions were that the sociocultural perspective became the focus of argument throughout the study as it can be used to connect communication and language development. The method to get answers to the questions was to perform four interviews in different preschools. The preschools were located within different counties close to a major city. The aim was initially to see how preschools work with technical tools and language development with children. This purpose was revisited as several preschools do not use technical tools as portable reading devices as we initially thought. The performed interviews were influenced by the model of Bornholm and knowledge of how this model can be used in the preschool support to support the language development for the children. Keywords: communication in preschool, language development, children with special needs, first language, learning tablet and social context.

Det huvudsakliga syftet med denna uppsats är att undersöka hur föräldrar ställer frågor till barn i en ålder av 2;3- 2;5 år med olika utvecklat ordförråd. Ett annat syfte med uppsatsen är att undersöka om barnen ger respons på föräldrarnas frågor samt om barnen själva ställer frågor. De frågeställningar som står i fokus är: Hur många ord och yttranden säger barn och föräldrar? Vilka typer av frågor ställer föräldrarna? Ställer barnen frågor och i så fall i vilken omfattning? Ger barnen respons på föräldrarnas frågor och i så fall i vilken omfattning? Är frågor överlag effektivt för att få barn i åldersgruppen att tala? Frågeställningarna besvaras genom språkanalyser av transkriberade dialoger mellan föräldrar och barn. Transkriptionerna omfattar tre timmar dialoger totalt och materialet har analyserats med hjälp av CLAN. Deltagarna är nio familjer från SPRINT-projektet vid Stockholms universitet. Deltagarna är utvalda för att representera barn med låg- respektive hög ordförrådsnivå baserat på SECDI data, ett svenskt normerat instrument i syfte att mäta barns ordförråd. Resultatet visar totalt cirka 2 500 yttranden från föräldrarna och drygt 900 från barnen. I ord räknat omfattar materialet totalt cirka 11 000 ord från föräldrarna och drygt 2 000 ord från barnen. Det råder stora procentuella skillnader mellan den högpresterande gruppen (höggruppen) och den lågpresterande gruppen (låggruppen) vad angår barnens antal yttranden och ordförråd till fördel för höggruppens barn. Föräldrarna visar inte på lika tydliga skillnader som barnen, men föräldrarna till barnen i höggruppen gör fler yttranden och använder ett mer varierat ordförråd än i låggruppen. Frågor är vanligt förekommande i föräldrars kommunikation med barn. Det framkommer att totalt för båda grupperna utgör frågor cirka en tredjedel av föräldrarnas yttranden. Vanligt förekommande frågor är ja/nej- frågor, produktiva frågor och frågor med hjälpverb. Barnen i höggruppen ger oftare respons på föräldrarnas frågor än barnen i låggruppen. Likaså ställer barnen i höggruppen oftare egna frågor än barnen i låggruppen. Den tydliga diskrepansen mellan grupperna indikerar att det finns samband mellan kvantitetsfaktorer i föräldrarnas och barnens tal samt att frågor är en metod som kan generera tal från barnen. Det kan finnas anledning att fundera över frågors olika komplexitet i förhållande till barnets ålder och övriga utveckling eftersom det framkommer att föräldrar generellt ställer komplicerade frågor. / The primary purpose of this paper is to explore how parents ask questions to toddlers (in this paper 2;3-2;5 year old children) with different vocabulary development. Another purpose of this paper is to explore if the children respond to questions given by their parents, and if the children ask questions themselves. Questions in focus for this paper are: How many words and utterances are said by the children and the parents? What kind of questions do the parents ask? Do the children ask questions, and if so, to what extent? Do the children respond to the parents’ questions, and if so, to what extent? Are questions generally effective in getting children to talk? The questions are answered through language analysis by transcribed dialogues between parents and children, where the transcriptions cover three hours of dialogues in total. The material has been analyzed in CLAN. The participants are nine families from the SPRINT-project at Stockholm University. The participants are selected to represent children with low- or high vocabulary based on SECDI data, a Swedish standardized instrument for measuring children’s vocabulary. The result shows about 2 500 utterances by the parents, and more than 900 by the children, while the number of words is approximately 11 000 words by the parents and more than 2 000 by the children. Concerning the children’s utterances and vocabulary there are large differences in percentage between the groups, with an advantage for the children in the high performing group (HP group). The parents don’t show such distinct differences, but the parents in the HP group make more utterances and use more variation in their vocabulary than the parents in the low performing group (LP group). Questions are frequently used in parents’ communication with children. In total, for both groups, questions constitute one third of the parents’ utterances. Commonly occurring are yes- no questions, productive questions and questions with auxiliary verbs. The children in the HP group more frequently respond to their parents’ questions than the children in the LP group. Similarly, children in the HP group more frequently ask questions than the children in the LP group. The distinct discrepancy between the two groups indicates causality between quantity factors in parents’ and children’s speech. It also indicates that the use of questions may in general be an effective method for triggering children’s speech. In general parents ask complicated questions, and therefore it may be appropriate to consider variations in question complexity in relation to the child’s age and overall development.

Bilingual children's language can display traits that mimic language disorder symptoms, and a good understanding of bilingual language development is essential in order to avoid misdiagnosis. This study explores the production of subordinate clauses in 46 typically developing Swedish-German speaking children aged 4–6 years. Transcripts of a picture-based narrative test in Swedish and German have been used to manually analyse the (1) number of subordinate clauses per age group, language and individual (2) presence and shape of the subordinator (3) distribution of subordinate clause types and (4) proportion of targetlike/non-targetlike word order in subordinate clauses. It is shown that the production of subordinate clauses is dynamic also in older bilingual pre-schoolers, and both similarities and differences compared to monolingual children are present. Clausal subordination increases across the age range at a group level, but with sizeable individual differences. Development in the two languages does not follow a homogeneous pattern, as e.g. demonstrated by the choice of clause types. Subordinators are largely targetlike and omissions are rare, but code switching occurs to some extent. The bilinguals’ word order exhibits signs of cross-linguistic influence, with a noticeable percentage of incorrect verb placement in German. Given this rather complex picture, a stronger focus on subordinate clauses in formal language assessments applied in Sweden would be desirable. / Flerspråkiga barns språk kan uppvisa drag som liknar språkstörningssymtom, och det är viktigt att ha god kunskap om flerspråkig utveckling för att undvika feldiagnoser. I denna studie undersöks produktionen av bisatser hos 46 typiskt utvecklade svensk-tyskspråkiga barn i åldern 4–6 år på ett explorativt sätt. Transkriptioner från ett bildberättelsetest på svenska och tyska har analyserats manuellt med avseende på (1) antal bisatser per åldersgrupp, språk och individ (2) förekomst och form av bisatsinledare (3) fördelning av bisatstyper och (4) andelen bisatser med målspråklig/icke-målspråklig ordföljd. Resultaten visar att de flerspråkiga barnens bisatsproduktion är dynamisk även i sen förskoleålder och att den har både likheter och skillnader jämfört med enspråkiga barn. Bisatsernas frekvens ökar på gruppnivå över det undersökta åldersspannet, men med stora individuella skillnader. Utvecklingen följer inte ett homogent förlopp i de båda språken, vilket t.ex. syns på preferensen av bisatstyper. Barnens användning av bisatsinledare är övervägande målspråklig och inledaren utelämnas nästan aldrig, men kodväxling förekommer. Ordföljden uppvisar tecken på tvärspråkligt inflytande, med en märkbar förekomst av verbplaceringsfel i tyskan. Med tanke på denna ganska komplexa bild är det önskvärt att bisatserna ges en mer framträdande roll i språktest för användning i Sverige.

Title: Children's aesthetic creation in different environments - A qualitative study of five preschool teachers 'perceptions of children's aesthetic learning processes in pre-schools' creative environments. Author: Mafi Misirhanova Supervisor: Karin Ehrlén Term: Spring 2017  The purpose of my thesis is to examine how the five pre-school teachers who participate in my study reflect about aesthetic expressions in relation to children's learning in preschool. Furthermore, I want to find out about the thoughts of the educators concerning the role of social interaction and the importance of the environment in the aesthetic creation process. My first question is about what view the informants have on social interaction during aesthetic creation processes. My second question is about how the informants reflect on studio and creative rooms at the selected three preschools. The study is based on qualitative interview methods where five pre-school teachers reflected on their internship. The essay is based on a sociocultural perspective based on development and learning. As far as the informants' views on social interaction during aesthetic creation processes are concerned, my study showed that there is a strong connection between aesthetic learning processes, language learning and social interaction as these areas interact during a creative activity at preschool. The informants' reflection on the studio and the creative rooms showed that the preschool with a creative room in the department gave the most positive response because the room was considered to enable continuous creation processes. The studio as a separate room and the playroom with a creative corner had shortcomings in accessibility with regard to the use of the room as well as the material.

<p>Razvojna disleksija je specifičan deficit čitanja uprkos prosečnoj inteligenciji, &scaron;kolovanju i obrazovnom okruženju, a u odsustvu senzornih ili psihijatrijskih poremećaja. Disortografija predstavlja otežano učenje pravopisa maternjeg jezika u vreme kada su deca istog uzrasta, sposobnosti i obuke već savladala pravopisna pravila. Deficit fonolo&scaron;ke obrade obja&scaron;njava disleksiju i disortografiju kao posledicu individualnih smetnji u fonolo&scaron;kom procesiranju koja se ispoljava kao slabost u sposobnosti stvaranja veze između fonema i grafema. Ciljevi ovog istraživanja bili su da se utvrdi nivo razvijenosti fonolo&scaron;ke svesnosti kod dece sa disleksijom i disortografijom, zatim u kojoj meri govorno jezički poremećaji utiču na ispoljavanje disleksije i disortografije, kao i da se utvrdi da li postoji razlika u ispoljavanju ovih smetnji u odnosu na pol dece. Istraživanje je obuhvatilo 689 dece, od prvog do trećeg razreda osnovne &scaron;kole čiji je maternji jezik srpski. Deca su podeljena u grupe u odnosu na razred koji pohađaju. Za procenu fonolo&scaron;ke svesnosti primenjen je test FONT koji obuhvata osam tipova zadataka:spajanje slogova, slogovna segmentacija, identifikovanje početnog fonema, prepoznavanje rime, fonemska segmentacija, identifikovanje zavr&scaron;nog fonema, eliminacija fonema, fonemska supstitucija i spajanje slogova. Brzina, tačnost, fluentnost čitanja i razumevanje sadržaja pročitanog procenjena je testom čitanja The Gray Oral Reading Tests-GORT 5. Za potrebe ovog istraživanja urađena je kros-kulturalna adaptacija testa, prema vodiču za adaptaciju u odnosu na jezičke i kulturalne različitosti. Za procenu razvijenosti govora i poznavanje reči, apstraktne inteligencije, faktora distraktibilnosti i koncentracije, kao i procene vizuomotorne koordinacije, kori&scaron;ćeni su subtestovi Revisk-revidirane skale za merenje inteligencije. Skalom za procenu pravopisnog nivoa rukopisa procenjeno je prepoznavanje semantičkih oblika i njihovo kori&scaron;ćenje u kontekstu saop&scaron;tavanja misaonih sadržaja. Rezultati pokazuju da najveći broj dece, od prvog do trećeg razreda, ima razvijeno čitanje u kategoriji proseka. U drugom i trećem razredu 61 dete (13.7%) ispoljava disleksiju, a 82 (18,4%) dece disortografične smetnje. Deca sa disleksijom i disortografijom če&scaron;će imaju deficit fonolo&scaron;ke svesnosti u odnosu na decu tipičnog razvoja. Deca sa disleksijom i disortografijom ispoljavaju deficit svih elemenata koji čine fonolo&scaron;ku svesnost. Elementi fonolo&scaron;ke svesnosti kod dece mlađeg &scaron;kolskog uzrasta čiji je maternji jezik srpski, predstavljaju značajne prediktore za sticanje čitanja i pravopisa. Govorno jezički poremećaji zastupljeni su kod 137 dece (19.9%), od prvog do trećeg razreda. Kod dece sa govorno jezičkim poremećajima se če&scaron;će ispoljava disleksija i disortografija u odnosu na decu tipičnog razvoja. Najveći broj dece sa disleksijom i disortografijom ima podprosečne rezultate na subtestu Rečnik. Disleksija i disortografija, kao i deficit fonolo&scaron;ke svesnosti su če&scaron;će zastupljeni kod dečaka, ali ta razlika nije značajna. U odnosu na rezultate istraživanja napravljen je Protokol za detekciju, procenu i korekciju disleksije i disortografije.</p> / <p>Developmental dyslexia is a specific reading deficit occurring in spite of average intelligence, appropriate schooling and educational environment, in the absence of any sensory or psychiatric disorders. Dysorthography represents a child`s impaired acquisition of the orthography of his/her mother tongue at the age when children of equal abilities and schooling have already mastered orthographic rules. A deficit in phonological awareness explains both dyslexia and dysorthography as a consequence of individual difficulties in phonological processing displayed as impairment in creating associations between phonemes and graphemes. The aims of this research were to establish the level of development of phonological awareness in children with dyslexia and dysorthography, the extent to which speech disorders affect dyslexia and dysorthography, and if there is a difference in their frequency with regard to gender. The study included 689 children, attending 1st, 2nd and 3rd grade of primary school, whose mother tongue was Serbian. Children were divided into groups according to school grade. For the assessment of phonological awareness the FONT test was used which entails eight types of tasks: syllable blending, syllable segmentation, identifying the initial phoneme, recognizing rime, phoneme segmentation, identifying the ending phoneme, phoneme deletion, and phoneme substitution. Reading accuracy, fluency and comprehension were assessed by The Gray Oral Reading Tests &ndash; GORT 5. For the purposes of this study, a cross-cultural adaptation was performed, in accordance with the guidelines provided in the guide for adaptation regarding linguistic and cultural differences. Subtests of the REVISK &ndash; revised scale for the assessment of intelligence were used to assess the developmental level of speech and vocabulary, attention and distractibility, as well as visual-motor coordination. A scale for the assessment of the orthographic level of handwriting was used to recognize semantic forms and their application in the context of conveying thought content. The results show that the majority of children, attending 1st, 2nd and 3rd grade have an average reading ability. 61 children (13.7%) attending 2nd and 3rd grade have dyslexia, while 82 (18.4%) have difficulties with the orthography. These children more often have a phonological awareness deficit compared to typically developing children. Children with dyslexia and dysorthography display a deficit in all the elements of phonological awareness. The elements of phonological awareness have proven to be significant predictors of mastering reading and writing in children attending lower grades of primary school, whose mother tongue is Serbian. Speaking disorders were found to be present in 137 (19.9%) children attending 1st, 2nd and 3rd grade. These children more often have dyslexia and dysorthography compared to typically developing children. The majority of children with dyslexia and dysorthography have below average results on the Vocabulary subtest. Dyslexia and dysorthography, as well as the phonological awareness deficit are more present in boys but the gender difference is not significant. In accordance with our results, a protocol was developed to detect, assess and correct dyslexia and dysorthography.</p>

A longitudinal study lasting 3.5 years was conducted to investigate complex grammar development, focusing on acquisition of the passive and type 3 conditionals, and its relationship with a number of domain-general, domain-specific and environmental factors. 128 children (M = 5 years 10 months) were tested at the beginning and towards the end of each school year starting from Year 1. The administered measures included established tests of fluid intelligence, short-term and working memory, seriation, grammar, vocabulary, literacy and arithmetic, plus newly-developed tests of passive and conditional sentence acquisition, and arithmetic word problem solving. It was demonstrated that grammar acquisition is not complete even when children start Year 4 of primary school (M = 8 years 7 months), when the current study was completed. At that time, 32% of children have not acquired type 3 conditionals and 89% showed no understanding of centre-embedded sentences. However, only 3% showed no passive sentence acquisition. Fluid intelligence, verbal STM and WM, ability to seriate, vocabulary and parental education level were all found to contribute to individual differences in complex grammar acquisition, independently of age differences and, where relevant, independently of non-verbal ability. There were differences between the passives and the conditionals in their relationship to these variables. Complex grammar development was found to be a significant predictor of reading comprehension, spelling and arithmetic performance, independently of age, non-verbal ability, verbal STM and WM. The findings demonstrate the inter-relatedness of higher cognitive functions, particularly domain-general with domain-specific ones. Modularity in its strictest sense (informational encapsulation, functional isolation) is not present in normally developing brains. Educational applications of the results are discussed.

Brown (1973) a proposé la « longueur moyenne des énoncés » (LME) comme indice standard du développement langagier. La LME se calcule selon le nombre moyen de morphèmes dans 100 énoncés de parole spontanée. L’hypothèse sous-jacente à cet indice est que la complexité syntaxique croît avec le nombre de morphèmes dans les énoncés. Selon Brown, l’indice permet d’estimer le développement d’une « compétence grammaticale » jusqu’à environ quatre morphèmes. Certains auteurs ont toutefois critiqué le manque de fiabilité de la LME et la limite de quatre morphèmes. Des rapports démontrent des variations de la LME avec l’âge, ce qui suggère que des facteurs comme la croissance des capacités respiratoires peuvent influencer l’indice de Brown. La présente étude fait état de ces problèmes et examine comment la LME et certaines mesures de diversité lexicale varient selon le développement des capacités respiratoires. On a calculé la LME et la diversité lexicale dans la parole spontanée de 50 locuteurs mâles âgés de 5 à 27 ans. On a également mesuré, au moyen d’un pneumotachographe, la capacité vitale (CV) des locuteurs. Les résultats démontrent que la LME et des mesures de diversité lexicale corrèlent fortement avec la croissance de la CV. Ainsi, la croissance des fonctions respiratoires contraint le développement morphosyntaxique et lexical. Notre discussion fait valoir la nécessité de réévaluer l’indice de la LME et la conception « linguistique » du développement langagier comme une compétence mentale qui émerge séparément de la croissance des structures de performance. / Brown (1973) proposed the « mean length of utterance » (MLU) as a standard index of language development. MLU is calculated by counting the mean number of morphemes in 100 utterances of spontaneous speech. The underlying hypothesis of this index is that syntactic complexity rises with the number of morphemes in utterances. According to Brown, MLU indexes the development of a « grammatical competence » up to about four morphemes. However, authors have criticized the lack of reliability of the MLU and the limit of four morphemes. Reported variations in MLU with age also suggest that factors such as the growth of respiratory capacities can influence Brown’s index. The present study reviews those problems and examines how the MLU and certain measures of lexical diversity vary according to the development of respiratory capacities. We calculated the MLU and the lexical diversity in the spontaneous speech of 50 male speakers aged 5 to 27 years. We also measured, with a pneumotachograph, the vital capacity (VC) of the speakers. The results show that MLU and measures of lexical diversity strongly correlate with the growth of VC. Thus, the growth of respiratory functions constrains morphosyntactic and lexical development. Our discussion focuses on the necessity to revaluate the MLU index and a « linguistic » conception of language development as a mental competence emerging separately from the growth of performance structures.