Search

This paper compares the written production of students learning Serbian as a foreign language with utterances of children at an early age acquiring Serbian as a first language. In one part of the corpus – the one compared – there are errors in essays by a linguistically heterogeneous group of students who attended an intensive course in Serbian at A1 and A2 levels of knowledge in the Centre for Serbian as a Foreign Language (within the Department of Serbian Language and Linguistics) at the Faculty of Philosophy in Novi Sad. The other part of the corpus – the one the first part is compared with – there are deviations from the standard Serbian language revealed in papers dedicated to the acquisition of Serbian and speech development of children at an early age (one to three years old). By means of comparison, common morphological and syntactic deviations from the standard Serbian norm have been determined and singled out (deviations: 1. in the creation of verb forms, 2. in congruence between the subject and predicate as well as 3. in the use of prepositional-case constructions with verbs) with the aim of showing which students’ errors occur under the influence of the target language itself, i.e. due to intralingual interference. The result of the analysis is a list of the total 43 types and subtypes of errors common for the two groups of respondents, which further leads to the conclusion that the singled out students’ errors can be considered intralinguistic ones. It is established that the total sum of examples by students is 38.1% of their whole corpus in the three observed grammar segments.

This paper carries out a morphological and syntactic analysis of errors in the form of a dependent member of a nominal phrase, specifically – in gender of the congruent word. The corpus has been formed through the excerption of material from 503 assignments written by students who have taken an intensive course in Serbian as a foreign language (at level A1 and A2) at the Centre for Serbian as a Foreign Language at the Faculty of Philosophy in Novi Sad. The aims of the analysis are as follows: identification of errors, systematisation according to types, quantitative and statistical analysis, presentation of analysis results according to types of errors, frequency of occurrence and the students’ first language (since the participants are ethnically and linguistically heterogeneous). Another aim is to compare, as far as possible, the analysis outcome with results of previous analyses of errors in gender of the congruent word in language production by students with Hungarian as their first language learning Serbian as their non-native language. Since there is no category of gender in Hungarian, it is pointed out that the acquiring of this category is especially difficult for speakers of this language. The analysis has shown that: foreign students incorrectly use male gender forms most frequently (45,5%), while the incorrect use of female and neutral gender is completely equal (27,2% per gender). Considerable similarities in errors have been established between foreign students and students with Hungarian as their first language. What was previously determined as the most frequent error among students of Hungarian nationality in Serbia has now been confirmed in the case of foreign students as well, including students whose first language is Hungarian regardless of whether they live here or come from Hungary.

With both Spanish and Romanian descending from a common ancestor, Latin, similarities between the two Romance sisters are a self-evident fact. And indeed, the inherent musicality of Spanish, most appealing to the Romanian ear, acts as an added incentive to an ever larger number of students acquiring it as bilinguals (L2) or as an optional subject (L3) in schools and centres for foreign languages. On the other hand, certain facilitating circumstances make it possible for some children to be actually immersed from quite an early age in the languacultural setting of the target language. A case in point provide families migrating to Spain or Latin America in search for better-paid jobs. Oddly enough, even if, admittedly, migration brings natives and foreigners closer together distance-wise, the remaining linguistic gaps to bridge are anything but few. Lending fresh support to this view is the paper at hand, which, while taking inspiration from one of the errors most frequently encountered in our teaching experience – i.e. mixing up sonar and soñar – , and amassing new evidence adduced by a national project research corpus, proposes to track down Spanish-Romanian interferences and zoom in on their structuralfunctional idiosyncrasies as exhibited by oral narrating skills of Romanian students at home learning SFL, on the one hand, and, on the other, of the former’s fellow-countrymen resuming speaking Romanian after spending/studying several years in Spain, as compared to emigrating Romanian natives, for whom Spanish has been promoted to the status of L1. The ultimate goal of our research, therefore, is to advance new teaching strategies meant to enhance metalinguistic awareness and catalyse acquisition of typologically related languages such as Spanish and Romanian in both major student groups under survey.

The interlanguage – dynamic and systematic language at the same time – is still a challenge for the researchers. Nowadays, we try to build software for the computer based corpus analysis and, in this sense (i.e. Free Text project), tagging tools for the learners’ errors are developed. This paper presents an original error tagging system built on a French corpus written by Romanian and highlights the benefits of such an annotation.

This paper analyzes what the error is, and what is its necessary role, which is not only something to correct or to avoid using containment. The presence of the error, because it forces us to review and argue, teaches us to avoid the extreme superficiality in which Arendt vested the inability to think in which rooted the banality of evil in so far as making us unable to free ourselves from what it tells us we must do.