Cilj je ovoga rada bio istražiti koje riječi i koje njihove oblike te njihov broj proizvode strani i nasljedni učenici početnoga e-tečaja hrvatskoga kao inoga jezika kako bi se vidjelo koje su razlike među tim dvjema skupinama. Također se željelo utvrditi je li pojedinačni jezični razvoj ispitanika sustavan. Građu čine prijepisi snimljenih razgovora tijekom pojedinačne nastave uživo pomoću Skypea na e-tečaju hrvatskoga kao inoga jezika HiT-1, i to četiriju pojedinačnih razgovora s deset ispitanika (N=10), s pet nasljednih i s pet stranih učenika, što je ukupno 40 sati snimljenih razgovora. Snimljeni su drugoga, šestoga, desetoga i dvanaestoga tjedna tečaja kako bi se utvrdilo što se događa s jezičnom proizvodnjom učenika u svim vremenskim točkama mjerenja. Transkribirani su fonetski kako bi se sačuvale izgovorne razlike, posebno kod nasljednih učenika koji u svojemu govoru rabe standardne, ali i dijalektalne oblike. Ispitanici su polaznici početnoga tromjesečnoga e-tečaja HiT-1 koji dolaze iz različitih zemalja svijeta i govore različite materinske i ine jezike. Budući da tečaj pohađaju osobe različitoga spola i dobi, u istraživanju su sudjelovali učenici muškoga (N=6) i ženskoga (N=4) spola u rasponu od 19 do 62 godine. Najprije je analizirana jezična proizvodnja svakoga učenika pojedinačno, potom su uspoređivani rezultati učenika unutar samih skupina, a onda su međusobno uspoređivane skupine stranih i nasljednih učenika. Iz prijepisa su najprije izdvojene natuknice, njima su pripisane sve proizvedenice (svi padežni i glagolski oblici koje je proizveo svaki od učenika). Popisani su točni i netočni oblici koji su podijeljeni u tri skupine: krivi oblici (postojeći hrvatski oblici upotrijebljeni u krivome kontekstu, npr. živim u Pulu umj. živim u Puli ili nismo pospremala umj. nismo pospremali), nepostojeći oblici (npr. ponejelak umj. ponedjeljak, nijest umj. nije) te dijalektalni oblici (pojavljuju se u hrvatskim dijalektima, npr. oću umj. hoću, poso umj. posao, volu umj. vole). Proizvedene su natuknice kategorizirane, podijeljene po vrstama riječi te prebrojane kako bi se vidjelo povećava li se njihov broj tijekom tečaja i je li to povećanje sustavno kod svakoga učenika i skupine. Potom su analizirane sve zabilježene proizvedenice kako bi se istražila morfološka raznolikost u proizvodnji svakoga učenika ponaosob te kako bi se vidjelo što se događa u morfološkome razvoju svakoga učenika pojedinačno, potom su se usporedili učenici unutar svake skupine, a na kraju su se usporedile sličnosti i razlike među skupinama stranih i nasljednih učenika. Usporedbom dobivenih rezultata pokazalo se da svi učenici do kraja tečaja proizvode veći broj hrvatskih natuknica i proizvedenica u odnosu na početak; da kod većine učenika povećanje proizvodnje riječi nije sustavno, tj. njihova je proizvodnja u pojedinim točkama mjerenja promjenjiva. Izravno poučavani oblici (nominativ, akuzativ, lokativ imenica; nominativ pridjeva te osobnih i posvojnih zamjenica; prezent i perfekt glagola) češći su u govornoj proizvodnji svih učenika od oblika na koje se programom nije sustavno usmjeravalo, a također su kod svih učenika znatno zastupljeniji padežni i glagolski oblici u jednini nego u množini. Sukladno očekivanjima pokazalo se da nasljedni učenici proizvode više hrvatskih natuknica i proizvedenica od stranih te da im je morfološka raslojenost veća i raznovrsnija uslijed veće izloženosti hrvatskomu tijekom usvajanja, no ta se razlika smanjila na kraju tečaja u odnosu na početak. Pokazalo se da su strani učenici točniji od nasljednih u proizvodnji imeničkih, glagolskih, pridjevskih i zamjeničkih oblika. Budući da pojedinčeva svojstva uvelike utječu na jezičnu proizvodnju učenika u ovladavanju inim jezikom, istraživanje je pokazalo da pojedini strani i nasljedni učenici e-tečaja HiT-1 pokazuju neka netipična obilježja u jezičnome razvoju s obzirom na skupinu kojoj pripadaju.
|Abstract (english)|| |
The aim of this thesis is to describe and analyse morphological aspects in spoken production of Croatian as a second and foreign language (L2) learners during the three month beginners’ online course, as well as find differences between foreign and heritage language learners. Although the teaching and learning environment of the course in which the research was conducted was largely the same for all learners (the same content provided for independent work, the same frequency of individual one-to-one live sessions with teacher, all students based abroad), in line with all previous scientific knowledge it was assumed that a) individual language development of foreign and heritage language learners would display certain differences in language production and b) that language development would be distinctive for each individual, allowing learners to exhibit features not typical for the group they belong to. Subjects are ten students (N=10), five of which are foreign and other five heritage language learners, aged 19-62, six males and four females. None of the heritage learners had learned Croatian language in institutional setting before. The research material consists of conversations with teachers recorded in 2016 and 2018 during one-to-one live sessions, with four points of observation (the second, sixth, tenth and twelfth week of the online course). All conversations were phonetically transcribed in order to preserve differences in pronunciation, especially with heritage language learners using standard as well as dialect forms. Transcribing one conversation of 45 to 54 minutes took on average five to six hours. Conversations were conducted in Croatian and English as a common language. Every sound, word and sentence, even laughter, were transcribed in order to provide better understanding of the context, use and accuracy of grammatical forms, hence these transcripts can serve as data for future research, especially for analysing teacher’s talk. During the total of 40 hours of conversations, each student has produced 300 to 500 shorter or longer utterances in Croatian, which is considered sufficient and reliable output for linguistic analysis. From the transcripts, first lemmas were extracted, each with all word produces (Cro. proizvedenice), i.e. related Case and verb forms produced by each of the students, including correct and incorrect forms. Incorrect forms were divided in three groups: existing Croatian forms used in wrong context (for example: živim u Pulu instead of živim u Puli or nismo pospremala instead of nismo pospremali), non-existing forms (for example: ponejelak instead of ponedjeljak, nijest instead of nije) and dialect forms (occurring in Croatian dialects, for example: oću instead of hoću, poso instead of posao, volu instead of vole). Next, the produced lemmas were categorized according to number and parts of speech in order to establish whether the number was increasing during the course and whether the eventual increase was systematic for each student and for both groups. Representation of all parts of speech in the relation to total language production was also analysed for foreign and heritage language learners. Word produces were attached to related lemmas to explore the morphological diversity produced by each student and to examine the morphological development of each student individually. Each student was compared to fellow learners within their group and finally differences and similarities between groups of foreign and heritage language learners were described. Since spoken production of non-native Croatian speakers is not sufficiently examined or described nor analysed and there is hardly any research of morphologic and lexical differences in language development of CL2 foreign and heritage language learners, the results of this research cannot be fully compared with existing findings. However, based on general knowledge on L2 development and on existing findings on morphologic aspects in L1 (first/mother language) and L2 (foreign/second language) texts of Croatian language, seven hypotheses were formulated which proved almost entirely correct. H1: It was assumed that both foreign and heritage language learners would be able to produce a greater number of Croatian lemmas and word produces at the end of the course than at its beginning, but that increase in number would not be systematic for all learners. Both groups have produced more Croatian lemmas and word produces at the end of the course compared to its beginning but the recorded increase was not systematic – foreign language learners have produced more word produces in the second than in the third conversation while heritage language learners have produced less in the second conversation than in the third, proving the accuracy of the hypothesis. H2: It was assumed that the order of the parts of speech represented in spoken production would be similar with both groups of learners. The research supported the second hypothesis as well, showing prevalence of nouns, followed by verbs and adverbs with both groups. Next in line are adjectives and pronouns, whereas prepositions, conjunctions, particles and interjections are least represented in both groups. The single difference was established between the two groups consists in the fact that with foreign language learners particles are more represented than conjunctions and interjections which have the same share, while with heritage language learners conjunctions are more represented than interjections and particles. H3: It was assumed that the directly taught word forms would be more represented than the forms that were not subject of explicit instruction, while production of other forms would remain sporadic and depend on individual learners input and exposure, particularly with heritage language learners. Both groups have indeed produced directly instructed Case forms of nouns (foreign 90%, heritage language learners around 85%). With foreign language learners the nominative Case ranked highest (40%) followed by accusative Case (36%), while with heritage language learners accusative Case (37%) was more frequent than nominative Case (35%). The share of locative Case use was at the same level for both groups. Both groups displayed considerably higher rate of adjective produces in nominative Case in relation to other Case forms (68% vs 35%). Accusative Case is second (30% vs 35%) with locative Case at the third place. Among the pronouns produces in both groups the most represented is the nominative Case (81% vs 65%), followed by the accusative and then the dative Case (with 31 word produces in heritage language learners group). Results for verb production indicate that the present tense is the most represented in both groups (77% vs 67%), followed by perfect (9% vs 12%). Share of imperative use is equal for both groups, while the future tense is more represented with heritage language learners. Use of conditional tense is recorded only with heritage language learners. These data are in line with research outcomes for Croatian as L1 that showed that frequency of use for Case forms depends on parts of speech. Consequently, the research findings proved the third hypothesis correct, since in production of inflected parts of speech the forms the course was focused on are more represented than other forms. H4: It was assumed that noun, adjective, pronoun and verb produces would be considerably more represented in the singular than in the plural form, in all Case forms and verb forms, both on individual level and in total. For nouns, around 80% of word produces was in the singular form (81% vs 79%). With adjectives, share of singular forms lies at 81% (79% vs 82%) and with pronouns at 79% (81% vs 76%), indicating that the proportion remains at a similar level for all declinable word classes. With verbs, the research recorded around 72% of singular forms (70% vs 74%), slightly less than declinable word classes. Therefore, the fourth hypothesis was proved correct, since among all word produces singular was considerably more represented than plural. H5: It was assumed that foreign language learners would produce less and be more accurate in inflected forms than heritage language learners who were expected to produce more output and more errors. The hypothesis was supported. Share of incorrect forms in relation to overall production of all word produces is significantly lower for foreign language learners (12% vs 18%). The difference in representation of incorrect forms between the two groups was most pronounced in verb production (11% vs 20%). H6: It was assumed that, as the course progressed, the initial difference in morphological word variety between foreign and heritage language learners will decline. At the end of the course the language production of all subjects would include more morphologically variety (correct or not) than at the beginning, with foreign language learners displaying higher rate of increase. Since foreign language learners have only just started learning and heritage language learners have been exposed to it, it was expected that at the beginning of the course foreign language learners would mostly produce directly taught forms, while heritage language learners would use Case and verb forms that were not directly taught more frequently than foreign language learners. However, in time, and especially at the end of the course, it was expected that the difference would decline and foreign language learners would produce the forms the course was focused on but also other forms that did not occur at the beginning of the course. Taking into account all word produces, the differences between the two groups decreased significantly (107% vs 68%) due to the fact that foreign language learners showed twice the rate of increase than that recorded in the heritage language learners group (62% vs 32%). Similar results were recorded for other parts of speech – nouns, adjectives and verbs, with the exception of pronouns where the difference was increased. The sixth hypothesis therefore was proved partially correct, supporting the claims on total production and production of nouns, adjectives and verbs but not of pronouns. H7: It was assumed that among foreign and heritage language learners there would be some respondents who would display deviations in L2 acquisition in relation to the group they belong to. Since numerous studies have shown that L2 acquisition process depends on various features of individual learners, it was expected that some of subjects among foreign and heritage language learners would not exhibit typical features the group they belong to, i.e. that spoken production of some foreign language learners would resemble the spoken output of some heritage language learners and vice versa. Three subjects showed characteristics not typical for the group they belong to. The only false beginner among foreign language learners was very different than the group he belongs to in category of morphological word variety while exhibiting more diversity than some of the heritage language learners in other categories. According to the number of produced lemmas and word produces as well as incorrect forms, he resembled some heritage language learners rather than his fellow foreign learners. Two of the heritage language learners showed more resemblance to the mentioned foreign language learner than some members of their group. Additionally, one of them was the only heritage language learner who achieved level of A1.1 at the final spoken examination, just like all foreign language learners but not heritage language learners. This research provided different insights in L2 development in learners of morphologically more developed languages like Croatian and exposed similarities and differences between foreign and heritage language learners. The results show that L2 learners at the beginners’ level during the three month course can produce much larger number of lemmas and word produces, on average 1,5 per lemma. The most represented forms are nouns, then verbs and adverbs followed by adjectives and pronouns, with other parts of speech barely represented. Occurrence of parts of speech and their order in Croatian as an L2 showed that frequency of use for Case forms depends on the parts of speech and that in spoken production singular forms considerably outnumber plural forms (around four fifths of output). Teaching plays an important role in L2 acquisition, given that learners, for the most part, produced Case forms of nouns the course was focused on (around four fifths of output). All learners make errors in production of target forms, mostly replacing them with other forms (more than two thirds of incorrect forms), non-existent or unique forms (one fifth). Therefore, the spoken production of learners should include all word produces, i.e. produced forms because the incorrect forms serve as indicator of their development. Beside insights provided by the research questions, during the analysis of Case and verb forms it proved necessary to introduce a new theoretical term, called word produces, which encompasses all that a learner produces in an attempt to produce a certain Case or verb form, including both correct and incorrect forms. Word produces is a hyperonym to all produced correct and incorrect forms, dialect and non-existent forms as hyponymes. The doctoral thesis opens new possibilities for further research, including research of every individual category frequence of incorrect singular forms in relation to plural, frequence of incorrect forms in individual Case and verb forms etc., frequency ratio of certain lemmas and entire frequency dictionary. Since there has been no systematic research on teacher’s talk at the beginners’ Croatian course, the transcripts containing teacher’s utterances during sessions offer a possibility to explore what happens in her/his spoken production during live teaching session, establish similarities and differences in interaction with foreign and heritage language learners and compare features of teacher’s talk with that used in interaction with L1 learners. Apart from research into linguistic development, the transcripts from this research can provide basis for exploration of some glottodidactic questions – how the application of informationcommunication technologies in teaching Croatian as a second language affects various aspects of language teaching, how the changed of roles of teacher and learners in that setting require alterations and additions of glottodidactic and methodological principles and set new paradigms in teaching process.