Literacy development of Roma children in L2: a comparative psycholinguistic study

Размещено на http://www.allbest.ru/

Literacy development of Roma children in L2: a comparative psycholinguistic study

Hristo Kyuchukov

University of Silesia in Katowice, Poland



Abstract

The paper presents results from a comparative empirical L1 / L2 literacy study with first-grade Roma and non-Roma children, resident in Bulgaria and Slovakia. The Roma children growing up in traditional communities are socialized through the rich oral culture surrounding them. In the extended Roma families, the care of the children is responsibility of everyone. In such a rich speech environment, the children learn their mother tongue. However, the oral skills of Roma children acquired in the home environment are not used at all in the process of literacy development in their L2 at primary school, neither in Bulgaria nor in Slovakia. Bialystok (2007) says that the oral skills of bilingual children in their L1 and L2 play an important role in the preparation for literacy. The Roma children's oral skills in their L1 are not further developed at the primary school level and their L2 oral skills are initially often very limited. 60 first-grade children (20 Roma and 40 nonRoma children) were tested using the RAN Test at the end of the school year, in Bulgarian and Slovak. Various aspects, including comparative differences in literacy performance between Roma children tested in Bulgaria and Slovakia, are discussed in detail. The findings show that (1) the time for naming the RAN Test in L2 can be used as a predictor for the children's literacy level in L2; (2) A low level of language competence of the children in their L1 appears congruent with a low literacy proficiency in L2; (3) The place of residence of the children appears to be a factor influencing the performance by Roma children in the RAN Test. The Roma children from the village have better results than Roma children living in the city. The level of oral proficiency in L1 of Roma children influences the timing of the naming of the RAN Test in L2. The Roma children resident in town have a more isolated life and do not possess such a rich vocabulary in their mother tongue. The children living in a village have more daily contact with non-Roma Bulgarians, for example, and thus develop a richer vocabulary in their mother tongue as well.

Keywords: Roma, literacy, Language 1, Language 2, Rapid Automatized Naming Test, narrative.



Кючуков Хрісто. Розвиток писемної грамотності у другій мові в ромських дітей: зіставне психолінгвістичне дослідження.

Анотація. У статті представлено результати порівняльного емпіричного дослідження грамотності рідної та другої мов у першокласників ромських та неромських дітей, які проживають у Болгарії та Словаччині. Ромські діти, які виростають у традиційних громадах, соціалізуються через багату усну культуру, яка їх оточує. У великих ромських сім'ях турбота про дітей є обов'язковою для кожного. У такому насиченому мовленнєвому середовищі діти вивчають рідну мову. Проте навички усного мовлення ромських дітей, набуті в домашньому середовищі, взагалі не використовуються впродовж розвитку грамотності в їхній другій мові під час навчання в початковій школі ні в Болгарії, ні в Словаччині. Бялосток (2007) стверджує, що вміння усного мовлення дітей-білінгвів рідною та другою мовами відіграють важливу роль у підготовці до навчання грамоті. Усні навички ромських дітей рідною мовою не розвиваються далі на рівні початкової школи, а їхні усні навички володіння другої мови спочатку часто дуже обмежені. 60 дітей першого класу (20 ромських і 40 неромських дітей) пройшли тестування за допомогою тесту RAN наприкінці навчального року болгарською та словацькою мовами. Детально обговорюються різні аспекти, включаючи порівняльні відмінності в успішності ромських дітей, які тестувалися в Болгарії та Словаччині. Результати показують, що (1) час для називання в тесті RAN на другій мові може слугувати предиктором рівня грамотності дітей у другій мові; (2) Низький рівень мовної компетенції дітей у рідній мові збігається з низьким рівнем володіння грамотністю в другій мові; (3) Місце проживання дітей є чинником, який впливає на результативність ромських дітей у тесті RAN. Ромські діти із сільської місцевості мають кращі результати, ніж ромські діти, які живуть у місті. Рівень усного мовлення рідною мовою дітей ромів впливає на швидкість називання в тесті RAN другою мовою. Ромські діти, які проживають у місті, живуть більш ізольованим і не володіють таким багатим словником рідною мовою. Наприклад, діти, які живуть у селі, мають більше щоденних контактів з болгарами-неромами, розвиваючи таким чином багатший словниковий запас рідною мовою. literacy native language roma

Ключові слова: роми, грамотність, рідна мова, друга мова, тест на швидке автоматичне називання, наратив.



Introduction

Language development of Roma children

The social aspect

In different cultures the language socialization of the young children depends on the cultural rules and practices of the communities. According to Leaper (20142017), in western cultures where the nuclear family with mother-father-child is primary, the socialization depends on the modes and frequency of communication between the parents and their child. The parents are the role models for the socialization and the contacts with the community are limited. Here the younger children learn almost everything through reading books, playing with toys and they also acquire also their different gender roles from an early age from their parents as models.

In a study of 50 parents and 50 children, Baferani (2015) has shown that the family plays an important role in the language socialization of the children. The author finds that factors such as “amount of love”, “warm-cold relationship” and “control” affect a child's socialization within a family.

According to Maccoby (1992), the parents play the role of “teachers”. The non - symmetrical distribution of power and competence between adults and children imply that the parent-child relationship has a unique role in childhood socialization.

Everywhere in Europe, many Roma still lead a traditional life. They live in extended families, where the care of the children is something all family members and members of the community are concerned with. There are different strategies for language socialization of the Roma children. The development of Roma children's speech is strictly determined by the Roma culture and traditions. As is known, Roma culture originates from India, and there are several parallels and close parallels with modern Indian culture. During my visits to India in 2001 and 2002 in observations and research among some Indian communities, whose lifestyle and culture have many common features with everyday life of European Roma, I had the impression that supervision and education of the children, and the ways they teach their native language (L1) have many similarities with Roma customs and practices (Kyuchukov, 2005).

Adult Roma in their linguistic relationship with children struggle in the same instinctive and spontaneous way and formulate linguistic samples, easy for children to understand, awakening and supporting their interest in the communication process. Both Indian and Roma communities use identical strategies. Conversations and language games with young children are part of the daily communication in traditional Roma communities and they obey the rules governing this kind of communication shaping Roma linguistic competence.

Traditional children's games, which are accompanied by comments addressed to children, are paid much attention in Roma culture. In the ethnographic literature, Popov (1988) has investigated the socializing function of children's games (1988). He analyzes socio-cultural factors in mastering the native language and gives examples, emphasizing the important role o f the traditional game culture for children's speech development.

According to my observations, teaching language to young children is associated with cultural traditions and with some features defining their distinctively oral culture. Oral Roma folklore includes the following genres: fairy tales, songs, short poems, fictional stories in which a child is the protagonist. Such a story is `language input”, it serves to shape children's speech competence in a native language. The way of expressive performance of stories for children and babies is characterized by a slow tempo in pronunciation. The contours of exaggerated intonation, leading to melodiousness, indicate that a tale or a story are being addressed to a very young child.

The model of a story offered to a child is manifested in preference for a dialogue. In these cases, the adult (narrator) answers the questions posed himself, or any other family member (aside from another child) can do it. These two models of conversation with a child used by Roma adults can consist of many parts, a fact intriguing for psycholinguists. Another characteristic feature of such a story is that its plot reflects the events of future but not present or past. The type of a conversation most often encountered in a dialogue with a Roma child is really a question, a kind of “test”. In some of our data collected, almost half of the sentences addressed by a mother to a 2-year-old child belong to this category. Using the questions an adult controls all the elements of the stock of a certain knowledge and ideas required by the community (Reger, n.d; 1991;1999; Reger & Berko-Gleason, 1999)

The predominance of question-tests is directly linked with features of Romani culture, as in certain oral folklore genres listeners ask the narrator the same kind of questions. Characteristic of oral replacement of an adult's and child's words shows that in a traditional Romani community, a small child is considered to be a member of a society from the very beginning. The speech patterns addressed to him are characterized not only by the elements and rules of the native language, but also by the fact that a child is at the same time introduced to the traditional oral genres of Roma culture and their characteristic linguistic use.

Cvorovic (2005) stats that “these children [referring to Roma children -HK] not only don't know Serbian, they don't know their language either. Their parents are usually illiterate and the have absolutely no appreciation of education” (p. 39). According to the author, “in this settlement, Gypsy children acquire hardly any skills and knowledge based on written text. The sets of objects that surround these Gypsy children includes neither children's books (nor any books) nor usually any toys. Children are expected to help around the house, fetch water and look after their younger siblings” (p. 42).

For someone who is familiar with Roma culture, this could perhaps sound unserious. Roma parents often are criticized because the activities with their children are not taken seriously by the outside world as a learning process. Very often the role of other traditions, cultural peculiarities of the Roma are not understood and not considered as something serious; but in the traditional families these activities are important. Presenting Roma children and parents in such a humiliated way is absolutely incorrect and does not present the Roma culture in the right light. It is more than obvious that such an author does not have any idea about Roma culture and such a statement is a form of racism. One cannot trust it and cannot use it for any purpose.

According to Stoyanova (2009), children growing up in the conditions of bilingualism, as many Roma children do, develop language consciousness earlier and more intensively than children maturing as monolinguals. Mastering the language, children communicate with adults and thereby they learn the language of everyday communication. This language is the basis of information present ed to a child. This is a special style of communication and it is termed “adult language with respect to children” in psycholinguistics (Stoyanova, 2009, p. 43).

Roma children, like any other children, follow psychological stages of development. In this case, it does not matter which speech community they grow up in. It is known that social and cognitive development of a child is adaptive and constructive (theory of J. Piaget), while Vygotskyan theory postulates that social and cognitive development is culturally shaped and grounded.

The Roma children grow up in a rich environment of oral tradition, hearing stories, songs, fairytales, proverbs, riddles and sayings in their mother tongue Romani. Romani is an orally used and transmitted language, although lately there have been attempts to develop a standard written norm of the language. Each Roma community around the world has its own oral traditions, but in general they are very similar to each other. The children learn the everyday language of communication from their parents, siblings and extended family members from a very early age. They follow the universal paths of language development of any language around the world. The way the parents introduce the words to the children (naming the objects, repeating and correcting the errors of the children) are the same as the parents from any other culture introducing the mother tongue to their children.

The observations presented here have been partially collected in different parts of the world at different times, and partly have become the result of a long-term observation of one Roma boy from his birth to 6 months, a boy named Kotse from Sofia (Bulgaria). He was brought up in a traditional Roma family in a Roma settlement. He was carefully looked after during his first 6 months of his life. Interaction between child - adult was observed.

As has already been mentioned, Roma children go through all stages of language development like other normally developing children. They have a child's linguistic level typical for the first year of life, where `moo” refers to a cow, a “gey- gey” to a horse, “woof - woof" to a dog. The Roma mother from Sofia talking to her daughter asks if she wants some water, using precisely this child register with the term “brum” as lexeme

Mother Manges li brum mi chaj?

Want some water, daughter?

In reality, the word water in Sofia-Roma dialect is pani, but the mother doesn't use this word because she knows that her daughter hasn't acquired it yet in her vocabulary.

A Roma boy, (1;6), whose family emigrated from the former Yugoslavia to Amsterdam, answers his uncle's question:

Adult: Kaj si opapu?

Where's grandpa?

Child: Pas ipapin.

Near grandma

The word papin is created by the child himself. There is such a word in the Roma language, and it means goose, but a child does not know this meaning and creates a word from the word grandpa papu, here adding the feminine suffix -in. In fact, a child follows the usual process of creation of new nouns of feminine gender in the Romani language, adding a specific ending to a noun of masculine gender.

Children from an early age have some notions about the verbs of mental state such as happiness, anger, fear, wish, etc. A boy (1;2) from a village in northwestern Bulgaria understands when an adult uses the verbs denoting mental state in order to ask him to do something:

Adult: De tut xoli!

Make an angry face!

The child expresses anger on his face. The same child understands all kinds of questions. He's just gotten his first teeth and his uncle checks whether the child knows where they are.

Adult: Danda, danda te kaj?

Teeth, where are your teeth?

Another boy (8 months old) was asked by his grandpa to welcome the guests.

Adult: Ker lenge “zdravey”!

Say “hello” to them

The child raises his hand for greetings and says hallo.

Oks (1983) reports that in some non-traditional societies, when a child is three years old, his elder brothers and sisters start looking after him and the communication between mother and a child is realized through the elder children. For example, if a child is thirsty and asks the mother to give him some water, she redirects his request to the elder son or daughter.

Mother: De lespani mi chaj!

Give him some water, my daughter!

Another important feature in the communication adult -- child in the Romani language is that children are presented with a very complex language from early childhood on.

For example: grandmother tries to warn three-month-old Kotse about the danger that he can injure himself by putting different items in his mouth.

Adult: Ma, kapharaves te zlebinja be mocho!

No, you're gonna hurt your gum, son!

The child hears negation and the verb in the future tense. It is not a simplified language because a child is small and can't understand the meaning of the sentence. Adult Roma often use sentences of this kind and even more syntactically complicated structures speaking with children.

All examples noted show that Roma children follow the stages of normal childhood development. The influence of Roma culture and traditions on language socialization of children is also obvious. A subsequent part of the study is focused on the aspects of communication within the Roma communities where oral folklore takes place in communication adult-child as a strategy of language development and socialization.

In various publications Kyuchukov (1994; 2000; 2002) has described the language socialization among Muslim Roma children from Bulgaria and the place of phenomena like code-switching and borrowing during the process of language socialization of Roma children acquiring simultaneously three languages - Romani, Turkish and Bulgarian.

In the literature on language acquisition and language socialization, one can find information on monolingual language socialization in illiterate, traditional societies. Schieffelin (1986) reported on language socialization of Kaluli children in Papua New Guinea, using an ethnographic approach for her observations. Schieffelin reported that the Kaluli children learn the language when the adult utters the message and after that adds the word: “elema” ("say like that").

In traditional Roma communities, the acquisition of linguistic and communicative competence by young speakers is a task of all members of the community who are competent in language use. According to Smith (1997), traditional Romani education is a community sort of education. Children learn to understand and read the verbal and non-verbal communication signals utilized by adults in their community at a much earlier age than their non -Roma counterparts. They participate in the day-to day activities of the community and it is here that through watching, listening and observing, they learn the economic, social, linguistic, political and moral codes of their society.

Everywhere around the world Roma children grow up as bilinguals. Crystal (1997) distinguishes `societal bilingualism' from `individual bilingualism', and the term `bilingualism' is used for people who acquire two or more languages simultaneously. In many East and Central European countries Roma children grow up bilingually or multilingually. However, upon entering the school the bilingualism or multilingualism of Roma children is not considered to be an asset. The preschool institutions very often do not offer any possibilities for Roma children to develop their mother tongue. In primary classes in most countries the children are forbidden to use their mother tongue. Due to these actions, the children develop the phenomenon of semi-lingualism, a situation where they do not know their mother tongue well and they cannot learn the official language of the country either. Kyuchukov (2009) showed that the knowledge of children 10 years old in Romani and in Bulgarian is equal to the knowledge of 6-year-old Roma monolingual children and this deterioration in competence is the reason why many children do not understand the school subjects and they drop out of school at the end of primary classes.

According to Eckert (2017), recognizing the role of the home language leads to successful socialization. Being disadvantaged in the Czech society, Roma children do not get any support in their mother tongue at school. The author suggests that by not demanding the schools to consider the home language in the process of teaching, the Roma have implicitly agreed that knowing Czech is the only natural way through which children from different cultures can gain access to education. The author compares the situation of Roma in Czech Republic with the situation of Afro- Americans who must study Standard American English, without considering their ability to speak Black English as a common, oral vernacular among many Afro- Americans.

The educational aspect

Discussing the situation of Roma in East Europe, Miskovis (2009) concludes that Roma have suffered racial discrimination and exclusion ever since they migrated to Europe. The discrimination against Roma and their exclusion are also part of the educational process nowadays. In a study with Czech teachers, Jarkovska et al. (2015) found out that the teachers are “blind” to Roma children's differences. They have the same expectations from them as from the Czech children. In order to justify their own behaviors, the teachers created lists with the cultural and genetic differences of Roma children in order to legitimize the differential treatment accorded Roma pupils at school.

A recent study in Poland (Grzymala-Moszczynska et al., 2019) showed that very often Roma children are placed in special schools after the children have initially been accepted in mainstream schools, because the teachers and headmasters put pressure on the parents, giving different “arguments” such as “ the child is constantly crying”, “it will be better for the child to be in special school”, etc. A psychological study with Roma children in special schools in a comparison with Roma children in mainstream schools showed that the there are no intellectual disabilities involved comparing both groups of children, there are no IQ differences between the two groups and there are no differences in their knowledge of Polish. The parents of the Roma children in mainstream schools complain that they and their children are treated differently by the teachers, the teachers will not believe the parents if they send a written message that the child is sick, and there is no respect for their Roma ethnicity.

All these studies used tests which culturally are not suitable for use with Roma children and they are biased towards the Romani language and culture, a serious obstacle to literacy development. Moreover, the tests are conducted in languages which the children very often do not know well. Even today in Slovakia and in the Czech Republic the children are tested with psychological tests administered in the official national languages. Based on their weaker results, the Roma children are placed in special schools for children with mental disabilities. However, there are no developed tests in the mother tongue of Roma children which can measure their abilities and knowledge of grammatical categories in their mother tongue, or any other kind of tests in Romani.

Recently, Romani language assessment test developed by Kyuchukov and de Villiers was used for testing the pre-school Roma children's knowledge of Romani grammatical categories. Children from Bulgaria, Macedonia, Serbia, Croatia, Czech Republic, Slovakia and Sweden were given a test measuring 10 grammatical categories for children aged 3-6 in Romani. The test measures both perception and production: wh questions, wh compliments, passive verbs, possessiveness, tense, aspect, fast-mapping nouns, fast-mapping adjectives, sentence repetition and number repetition. 80 % of the children tested throughout Europe showed very high results in the knowledge of all grammatical categories. In all selected countries, the number in the Roma population is high, and the number of pre-school and school children is also large. Unfortunately, the governments and the education ministries in these countries do not have any adequate policies towards the proper education of Roma children (Kyuchukov et al., 2017)

The language of instruction, communication and learning dominant at the school is very often an obstacle for the successful education of indigenous, migrant, and minority children. These children are significantly disadvantaged as a result of not growing up with the standard dialect of whatever language is valued by the dominant social group(s), and by the low status of whatever “foreign” language, or dialect of the standard language, serves as these children's mother tongue. We know from examples around the world that this is not simply a matter of competence, even though linguistic nationalists tend to insist that it is, but rather that even the smallest deviations in usage and pronunciation, deviations that have nothing to do with competence, are sufficient to stigmatize already stigmatized minority speakers ( see May, 2011; Adesope et al. 2010; Skutnabb-Kangas, 2000). Additionally, the languages of ethnic minorities are very often also the languages of people in poverty, exacerbating the disadvantage that their speakers experience in mainstream institutions like schools. One such linguistic group in Europe, the Romani, constitutes an ideal case of educational injustice when encountering the multiple obstacles of linguistic difference, racism, social marginalization, and poverty.

Historically, and up to the present, Romani children have had limited access to formal schooling. In the second half of the 20th century, most European Romani lived in socialist countries, where children were routinely diagnose d as “deficient”, i.e. mentally and socially retarded, and placed in special schools where they most often constituted the majority. Romani children were also placed in segregated, Romani-only catch-up classes in regular schools at the beginning of their primary education (Cahn & Petrova, 2004; Sokolova, 2008). Despite European Court of Human Rights (ECtHR) decisions in cases such as D.H. and others v. Czech Republic (2008), which determined that this system violated the rights of Romani students, and required change, these same students continue to be diagnosed in large numbers, peremptorily, as “mentally handicapped” and placed in segregated classrooms and schools, where their educational prospects are dim (for Slovakia, see New, 2011; Miskolci et al., 2017). In this context, Roma students' language ability in Romanes, the mother-tongue or heritage language of Roma is judged to be a liability rather than an asset, and their bilingualism is often interpreted as an inability to speak any language (Hubschmannova, 1979; Kyuchukov, 2014a, b; Kyuchukov & de Villiers, 2009; 2014a, b, c; New, 2014).

In 2010, the ECtHR (ECtHR) ruled in favor of Romani students, who claimed that their placement in special Romani-only classes because of purported deficiencies in Croatian language violated their human rights. This case originated in the late 1990s in two rural villages in Medimurje county in the northeast ern corner of Croatia, adjacent to Slovenia and Hungary, only 30 kilometers from Austria. While the data concerning these fourteen students was typically bleak -- none of them finished primary school and many were absent from school as much as they were present -- the ECtHR decided only by the narrowest of margins that unequal treatment had occurred (ECtHR, 2010). While advocates of the Romani cause claimed that the decision made discrimination based on language difference illegal (e.g. Memedov, 2010), that seems a wishful representation of what the majority actually said. In fact, separation of students into separate classes on the basis of language difference or deficiency was not found to be illegitimate or discriminatory. Rather the Court found that two village schools in Croatia had not taken adequate measures to see that the Romani children were given the opportunity to actually learn Croatian - and consequently the rest of the mainstream curriculum - in their segregated classes. Had the schools done a better job of teaching the Romani children how to speak good Croatian, their separation might have been justified, regardless of the educational outcome.

Literacy in the second language (L2)

In her early work, Gudschinsky (1968) gives a definition of the literacy process and tries to answer the question as to what a literate person is. In most countries, the ethnic minorities learn to read and write in the official language of the country. Even if the minority language is used in primary schools, it is used to master the literacy in the official language. There are different literacy methods but the most traditional one known everywhere around the world was developed by Frank Laubach in the 1930s in the Philippines. It is known as the “Laubach method” - “pictures are used to serve as keys to words and the words in turn serve as keys to their initial sounds” (Gudschinsky, 1982).

Another traditional method is the one developed by Townsend (1948; 1952 [cited in: Gudschinsky, 1982]), and this method consists of recognizing whole words by comparison with each other and eventually using these words in sentences and stories. It differs from a typical look-say method in that the words are very carefully controlled in terms of letters. The textbook starts with three- to five-letter words and adds new ones one at the time. It is expected that in the process of comparison and word recognition, the pupil will learn to associate the letters with the phonemes they represent (Gudschinsky, 1982).

A third method described by Guschinsky (1982) is the one developed by Beckman (1950). Beckman uses syllables in the process of teaching to read.

In most cases the minority children who are bilingual become literate through one of the above-mentioned methods, and often their mother tongue is not taken into account. The methods deal mainly with the pedagogical side of the literacy development process. There is no attention to the linguistic side of it. Guschinsky (1982) developed a linguistic literacy development method taking the differences in the cultures into account. “Different cultures apparently see different sets of letters as similar enough to be confused” (p. 56). Therefore, the author suggests that the literacy development process has to start with introducing the letters most distant from each other and sounds first in order to avoid their misuse. This method pays attention to comprehension. The introduced words and short sentences, which the students must read, have to be comprehensive.

In their classical work on psychology of literacy, Scribner and Cole (1981) stated that the “studies of the effects of schooling on cognitive development need guidance from developmental theories regarding what aspect of schooling will have what effect and why” (p.13). Schooling is a setting in which the children are exposed to large amounts of information and it does not rely only on the technology of reading. “Studies on the psychological effects of literacy converge with and can contribute to the broader area of interest on the effect of education obtained through schooling” (p.13).

Goswami (2009) says that the act of reading is a linguistic act and it is important to be understood that the visual symbols must be learned, they represent a spoken language. “To become a fluent reader, therefore the cognitive skills involved in the proceeding spoken language may be expected to be important” (p.135), Bialystok (2007) identifies three prerequisites for acquisition of literacy: competence with the oral language, understanding of symbolic concepts of print, and establishment of metalinguistic awareness. The relationship between bilingualism and the development of each of the mentioned three skills is different, sometimes indicating an advantage (concept of print), sometimes a disadvantage (oral language competence).

The Socio-Cognitive Theory (SCT), based on the socio-cultural theory of Frawley and Lantolf, (1985) and Vygotsky's theory of cultural-historical psychology “focus on social and cultural factors in L2 learning and use” (Lantolf, 2006, p. 69). The human psychological process, as Pantolf notes, is organized by three fundamental cultural factors: activities, artifacts and concepts. Bruner (1966) (cit. in Lantolf, 2006) proposed that activities, artifacts and concepts function as cultural amplifiers and suggested that as with physical tools, symbolic artifacts (e.g. literacy) amplify memory and increase our capacity to organize and communicate knowledge.

The process of learning of the symbols in each culture from a very early age is important and plays a role in the cognitive development of the child. One such important symbol complex is the language and the way the children learn it from very early age. Being part of and integral to the culture, the language plays an important role in the process of literacy. In order to understand the literacy development process, one should study the whole environment of the child, and the opportunities which s/he has in order to become prepared for the literacy process. The environment includes not only the physical surroundings but also human relationships (McLane & McNamee, 1990). These relationships are the ones which introduce to the children the tools and materials used for the literacy development process. In different cultures, the relationships between children and adults differ and the cultural symbols used for preparation for literacy are different. All this forms the basis for the socio- cognitive theory - through the culture the child develops his/her cognition and that relates to the preparation for literacy and the process of literacy and its acquisition.

Roma children acquire literacy in their second language (L2), which is often their weaker language. Studies in the research literature shed light on the problems of bilingual children who are literate in a language at school which is different from their home language.

Bialystok (2007) says that there are sociolinguistic and cognitive factors which support the literacy in the L2 of bilingual children: these are knowledge of the oral first language and knowledge of the oral second language. The studies investigating children to read in a weak language highlight the importance of oral language proficiency.

However, in many countries in Europe, the Roma children's knowledge of their mother tongue orally is not taken as an advantage supporting the literacy development process in L2; rather, it is assumed to be precisely the opposite. Many researchers, educators and psychologists think that the biggest problem the Roma children must cointend with is in fact their bilingualism. They suppose that the educational process and the literacy of Roma children will be more successful if the Roma children are monolingual.

In most European countries, there is no knowledge about how to organize the literacy development process with bilingual children. Consequently, in many countries Roma children are placed in special schools. There is no research focusing on Roma children examining in depth the literacy development process in their L2, the results from which could be used for changing the educational system with Roma children and particularly the literacy process. In this respect, the study presented here tries to bring new knowledge to bear, seeking to shed fresh empirical light on the literacy problems of Roma children when they must be educated in L2.

Research questions

In the study we take as a point of departure the idea of Bialystok (2007) about the importance of the oral knowledge of the mother tongue, and based on that the research questions are formulated:

Q1: How does the naming speed in the RAN Test in L2 predict children's literacy?

Q2: How do the levels of oral language competence in Romani influence literacy in L2?

Q3: Is there any difference between the performance of the RAN Test by Roma children depending on their residential locality, namely town or village?

Hypotheses

Our hypotheses are the following:

H1: Oral proficiency in L2 will influence the speed of the RAN Test in L2.

H2: The performance in the RAN Test in L2 will predict the level of literacy in L2.

H3: The level of oral proficiency in L1 of Roma children influences the speed of naming in the RAN Test in L2.

H4: Roma children from the city will perform better in the tests than the Roma children from the rural area.

Method

The Tests

The children in the study must perform the RAN (Rapid Automatized Naming) Test, which includes 6 subtests, each of them with 50 items:

- colors;

- numbers;

- letters;

- objects;

- combinations of letters and numbers

- combinations of colors, letters and numbers.

The second task, which only the Roma children have, is an oral narrative in Romani based on a picture story. The story contents 7 pictures, telling the story of a small dog which gets lost and meets different animals on the way, and finally finds his house.

The children in the comparative study

All children involved in the study were in 1 st grade in Bulgaria and Slovakia.

Table 1

Number of the Children Involved in the Study

In Bulgaria	Residence
Experimental group	10 Roma children	Village
Control group	20 Bulgarian children	Town
In Slovakia
Experimental group	10 Roma children	Town
Control group	20 Slovak children	Town



All together 60 children were tested with the RAN Test in the official languages of the countries, Bulgarian and Slovak. The Roma children were also tested in their mother tongue, asked to tell a narrative based on a picture story. All Roma children in both countries are in segregated classes.

The Roma children in the study belong to two different groups. The Roma in Slovakia are more conservative as a community. They are Lovara (Vlax group). The Roma group in Bulgaria belongs to a non-Vlax group. It used to be a Muslim community but over the last approx. 20 years many such Roma were converted to evangelical Christianity and they have a rich religious life together with the Bulgarians in the village. The children are also involved in all religious activities from an early age. They also attend kindergarten in the village, while the Roma children in Slovakia do not attend kindergarten.

Procedure

All children were tested in a separate room individually in the official language of the country where they live and their performance was measured in terms of time (in seconds) using a timer. The Roma children's narratives were audio -recorded and then transcribed and analyzed (using the TTR - Type Token Ratio) for each child and per group.

Results

The RAN Test

The results from the testing are given in tables, separately for the Slovak and for the Bulgarian children. Table 2 gives the results of the Slovak children. The ethnic Slovak children have much better performance than the ethnic Roma children. In most cases the differences between the groups are statistically significant. However, in naming the colors (battery 1: t=4.8; p<.001) and naming the numbers and letters (battery 5: t=3.7; p<.001), the results between the two groups are not statistically significant, although the Slovak children perform much better on those batteries. In all other batteries, the Roma children's performance in the tests is poorer than that of the Slovak children - they need much more time per group for performing a certain test than the non-Roma Slovak children do. All the results are shown in Table 2.

Table 2

Results from the Testing of the Slovak Children

Subtests	Slovak children (time in sec.)	Roma children (time in sec.)	Statistically significant differences
colors	62	90	t=4.8; p<.001
numbers	43	62	t=2.7; p>.01
letters	37	58	t=3.0; p>.001
objects	62	79	t=2.6; p>.001
numbers	50	75	t=3.7; p<.001
and letters
colors,	62	81	t=3.0; p>.001
numbers
and letters

Let us look at the results of the Bulgarian children. The results are presented in Table 3. As is shown in the table, the two groups of children have large differences in their performance of the batteries. Only their performance on battery 6 - naming the colors, numbers and letters - shows no statistical differences between the groups (t=3, 4; p<.01). In all other batteries, the non-Roma Bulgarian children need less time as a group for performing the subtests than the Roma children, although when performing battery 2 (the numbers), the Roma children perform it in a shorter time than the ethnic Bulgarian children. However, the differences are not statistically significant.



Table 3

Results from the Testing of the Bulgarian Children

Subtests	Bulgarian children (time in sec.)	Roma children (time in sec.)	Statistically significant differences
colors	50	62	t=1.7; p>.01
numbers	52	42	t=1.4; p>.01
letters	26	43	t=2.5; p>.001
objects	51	64	t=1.8; p>.01
numbers and letters	37	51	t=2.0; p>.01
colors, numbers and letters	44	67	t=3.4; p<.01

The results in both countries are similar - the children from the experimental groups (Roma children) show lower results than the children from the control groups (Bulgarian and Slovak children). Why is that so? It seems that in both countries that until the end of the first school year, the Roma children do not learn well enough the letters, numbers, names of the colors and objects in the official language of the country so that they can automatize them when they have to name them. However, looking at the results of the children from the experimental groups, one can see that the Roma children in Bulgaria have better results than Roma children in Slovakia and it is the same with the control groups - the ethnic Bulgarian children have much better results than the Slovak children. It seems the educational system in Bulgaria and particularly the literacy development process during the first school year is much better organized in Bulgaria than in Slovakia, although in both countries the literacy method used is the same, the Laubach method.

Is residence a factor for better performance in the RAN Test? The results between the Roma children from a village (in Bulgaria) and the Roma children from a town (in Slovakia) are given in Table 3.



Table 4

Results from the Testing of Slovak Roma and Bulgarian Roma Children

Subtests	Roma from Slovakia (time in sec.)	Roma from Bulgaria (time in sec.)	Statistically significant differences
colors	90	62	t=4.8; p<.001
numbers	62	42	t=2.8; p>.001
letters	58	43	t=2.3; p>.01
objects	79	64	t=2.3; p>.01
numbers and letters	75	51	t=3.6; p<.001
colors, numbers and letters	81	67	t=2.2; p>.01

As can be seen from Table 4, in general the Bulgarian Roma children, although they live in a village, perform most of the subtests much better than the Slovak Roma children. The differences between the groups are significant. In general, the Roma children from Bulgaria perform the Test better than the Slovak Roma children. Only in two subtests are there no significant differences - naming the colors and naming the combination of numbers and letters. It seems these two subtests are equally difficult for both groups. Our expectations were that the Slovak Roma children living in a town would have better results because the town gives more possibilities for communication from an early age in the official language. However, the Roma children in Slovakia belong to a group of so-called Vlax Roma, which is more conservative and traditional. By contrast, the Roma children from Bulgaria are more open, from less traditional families, and attend kindergarten from an early age. Moreover, the Roma in the village are more open and ready to engage in communication with non-Roma Bulgarians in the village, which helps the children to acquire some knowledge in the official language of the country. Consequently, it can be stated that the residential, locality is not a factor - but rather that the Roma community and its characteristics are what influences the children's knowledge.

The narratives

The second task performed only with the Roma children was the narrative task. The children had to tell a story in Romani language based on pictures. The stories were recorded and analyzed using the Type Token Ratio (TTR) method. This method shows in general the richness of vocabulary of the children involved in the study. The results from this task are presented in Figure 1.

Figure 1

TTR of Roma Children in Romanes

As one can see, the two groups have different TTR (vocabulary richness) in their mother tongue. Although the children are speakers of different Romani dialects, that does not influence the results. The Slovak Roma children have 44.5 words per story and the Bulgarian Roma children 50.3 words. The differences between the groups are statistically significant (t=.8; p=2.6 (p> .01). One of the problems of the Slovak Roma children is that poor performance in the narratives is due to the lack of knowledge to develop a cohesive narrative - rather they simply name the protagonists and their actions but don't narrate a coherent story.



Discussion

The first year in the school is extremely important for the children's cognitive development. How successful they are in learning the names of the colors, numbers, letters, objects and their combinations, as well as their automatized naming, serve to predict how successful their literacy development process during the second year will be and their success in school in the subsequent years in primary school. In both countries the Roma children show lower results than the children from the majority population. It seems that the level of knowledge of the mother tongue plays an important role in the process of preparation for literacy acquisition and development. The Roma children in Bulgaria achieve better results than the Roma children in Slovakia, because there are factors which influence their success, such as: the level of knowledge of the mother tongue, attendance at kindergarten and the type of community they live in. The Bulgarian Roma children attended kindergarten before entering primary school, they also participate in the evangelical church activities of their families. The Bulgarian Roma children live in a village where the communication with ethnic Bulgarians is more intense. The Slovak Roma children are living in an isolated community. They do not have any contacts with non -Roma Slovaks. And their Romani is not as developed as much as the Romani of the Bulgarian Roma children.

The results of the majority children indicate that the ethnic Bulgarian children show better results in the RAN Test than the ethnic Slovak children. This has to do with the textbooks and with the methods used in the process of literacy, although in both countries the children become literate using the same method.



Conclusions

In conclusion, it is possible to say that the research questions were answered: (1) The time for naming the RAN Test in L2 predicts the children's literacy in L2; (2) The low level of language competence of the children in their mother tongue predicts the low literacy capacity in L2 (the Slovak Roma children); (3) The place of residence of the children appears to be a factor influencing the performance in the RAN Test by Roma children. The Roma children from the village have better results than the Roma children the children living in the town, locality an affordance here.

Regarding the hypothesis, H1 was confirmed. The children with low oral proficiency in L2 have lower results in timing (speed) in the performance of the RAN Test in L2. H2 is also confirmed. The performance of the RAN Test in L2 is a predictor for the level of literacy in L2. H3 is confirmed as well. The level of oral proficiency in L1 of Roma children influences the timing of the naming of the RAN Test in L2. H4, namely that the Roma children from the town will perform better in the tests than the Roma children from the village is not confirmed. The Roma children from the town have a more isolated life and do not possess such a rich vocabulary in their mother tongue. The children living in a village have more contact with non-Roma Bulgarians and thus develop a richer vocabulary in their mother tongue.

The poor performance in the RAN Test by Slovak Roma children predicts significant difficulties in acquisition of literacy in L2. The Roma children are far behind the non-Roma children because they do not have possibility to develop their mother tongue further in a kindergarten. In Bulgaria, it is even forbidden for teachers to use the mother tongue of their pupils in communication with the minority children. Moreover, non-attendance of kindergarten by Roma minority children and lack of communication with native majority children who are speakers of the official language of the country as L1 is also an important obstacle for learning the L2 in natural way. This predicts that the Roma children will become literate in the official language of the country at a slower pace of progress than the non -Roma children, because oral knowledge of the mother tongue and oral knowledge of the official language, as have been shown here, are predictors of good and fast literacy. This also confirms the theory of Bialystok (2007).

In conclusion it can be stressed that the state policies should create conditions to protect and implement the right of Roma children to receive literacy development education in Romani as a mother tongue, which is a prerequisite for better acquisition of literacy in the second language, as has been shown in this study.

...