Show Less
Open access

Dimensions of Linguistic Space: Variation – Multilingualism Conceptualisations Dimensionen des sprachlichen Raums: Variation – Mehrsprachigkeit – Konzeptualisierung


Edited By Lars Bülow, Ann Kathrin Fischer and Kristina Herbert

This volume focuses on the use and structure of the German language in Austria. In addition, the aim of the book is to compare the linguistic conditions in Austria with those in other German speaking countries. The 20 articles present current findings from the research fields of variation, contact and perception.

Der Band widmet sich schwerpunktmäßig der Verwendung und Struktur der deutschen Sprache in Österreich. Ziel des Sammelbandes ist es außerdem, die sprachlichen Verhältnisse in Österreich mit denjenigen in anderen deutschsprachigen Ländern zu vergleichen. In 20 Beiträgen werden daher aktuelle Forschungsergebnisse aus den Forschungsbereichen Variation, Kontakt und Perzeption vorgestellt.

Show Summary Details
Open access

Medial Diglossia in Vorarlberg – a Gain or Hindrance for Monolingual Language Acquisition? (Mirja Bohnert-Kraus / Andrea Willi / Katharina Korecky-Kröll / Andrea Haid / Christine Czinglar)

Mirja Bohnert-Kraus, Andrea Willi, Katharina Korecky-Kröll, Andrea Haid, Christine Czinglar

Medial Diglossia in Vorarlberg – a Gain or Hindrance for Monolingual Language Acquisition?

Abstract: Parents of children living in regions with a so-called dialect-standard continuum often worry that the use of dialect might impede the child’s language acquisition and put the child into an unfavourable position with respect to education (cf. Wiesinger 2008: 44). Therefore, the parents often decide to use a more standard-like variety. Even in regions with diglossia (e.g. Vorarlberg, cf. Ammon 2003: 164) there is a growing discussion over whether children in preschool age should be confronted with the national standard variety.

Our article deals with possible differences between children from Vorarlberg and Vienna regarding language acquisition with respect to the influence of a standard- or dialect-speaking environment. Based on data from two cooperating projects (University of Vienna and Schweizer Hochschule für Logopädie Rorschach), we investigate if language acquisition proceeds differently depending on the primary variety: Six children are compared, three living in Vienna – and therefore in a more standard speaking environment – and three children from Vorarlberg who are socialised in Alemannic dialect. Further influencing variables such as socio-economic status, age, gender, number of siblings etc. were controlled so that we have three matched pairs. Their language development was investigated by using speech tests and analyses of spontaneous speech at four points in time from age 2;11 to 4;9.1

In the standard speech tests, there are almost no differences between the two groups considering the variety characteristics. In spontaneous use of grammatical categories, the Viennese children seem to reach higher frequency and diversity at a younger age and a greater progress within one year. Possible variety effects are considered and individual reasons are discussed.←233 | 234→

1 Introduction2

This paper deals with bilectal language acquisition in a situation of medial diglossia as opposed to monolingual acquisition of a standard variety. A much-debated question is whether diglossic language environments (e.g. in Vorarlberg) are a gain or a hindrance for a child’s language development. To date, there has been little agreement on which variety should be used in diglossic language environments (e.g. in Vorarlberg or Switzerland), due to both potentially negative and positive effects on children’s language attainment such as the extra challenge for bilingual children to acquire their second language (cf. Landert 2007) or better executive abilities (cf. Katsos 2016) when speaking a dialect.

The present paper is based upon data from two research projects: EdUS “Influence of the environment on children’s language acquisition” and INPUT “Investigating parental and other caretakers’ utterances to kindergarten children”. The two studies compare children from Vorarlberg with children from Vienna. The children in Vorarlberg grow up with the regional variety “Vorarlberg Alemannic”, whereas the Viennese children are exposed to a more standard-like variety. In the current study, three matched pairs of children were selected and compared with respect to their acquisition of grammatical phenomena. In order to compare the acquired data, variables that have an influence on language acquisition were controlled as much as possible.

The paper is structured as follows: In the next section, we will present the theoretical background of medial diglossia in Vorarlberg, the differences between the standard variety and Vorarlberg Alemannic as well as previous studies on advantages and disadvantages of bilectal language acquisition. Section 3 gives an overview of the two research projects, the research question, the participants and the method. Section 4 presents the results on the lexical and grammatical development of children growing up in a standard- and in a dialect-speaking environment. In section 5, we will discuss our results and relate them to those of previous studies. Finally, we give a short conclusion on implications for child-rearing in a bilectal situation.←234 | 235→

2 Theoretical background

Various studies have discussed the possible effects of a diglossic environment on language acquisition. This section focusses on the situation in Vorarlberg, but also discusses results from other European regions.

2.1 Medial diglossia in Vorarlberg

According to the classification of German dialects, Vorarlberg Alemannic is part of the Middle Alemannic dialects (cf. Wiesinger 1983). However, other dialects such as High Alemannic dialects are also spoken in Vorarlberg. A mitigated form of medial diglossia (cf. Kolde 1981) is found in Vorarlberg (cf. Ammon 1995; Haid 2011). This means that the standard variety is used in media (e.g. television, papers and books) as well as at schools (formal language) and the regional variety is used for daily communication (informal language). Children hear both varieties from an early age onwards. Still, children do not usually speak the standard variety before entering school (see also Schallert 2010: 35).

2.2 Differences between the standard variety and Vorarlberg Alemannic

The following characterisation of Vorarlberg Alemannic is not intended to be exhaustive; we only discuss features that are important for our analysis of functional elements in children’s language data. The two varieties have distinct pronunciation, vocabulary and grammar rules. In particular, differences in nominal and verbal inflection are apparent. The inflectional system of Vorarlberg Alemannic is simpler compared to standard German (cf. Gabriel 1963). In the following, some grammatical characteristics of Vorarlberg Alemannic are described.

Relative clauses with all three genders (masculine dr, feminine d, neutral s) are initiated with the relative particle wo ‘who’, as in dr Maa/ d Frau/ s’Kind, wo z’Vorarlberg wohntthe man/ the woman/ the child who lives in Vorarlberg’ (Gabriel 1963). Moreover, subordinate wh-clauses have two complementisers, as in Ich was ganz genau, warum dass är des kset hot ‘I know for sure, why that he has said that’ (cf. Penner 1993 for Bernese Swiss German).

The inflection of personal pronouns is based on marked forms of either the singular first or second person pronouns, such as ii ‘I’, miar ‘me-DAT’, mii ‘me-ACC, myself’, or du ‘you-NOM’, diar ‘you-DAT’, dii ‘you-ACC, yourself’ and plural miar ‘we’, üüs ‘us’, iar ‘you-NOM-PL’, and eu ‘you-ACC/DAT, yourselves’ (Gabriel 1963). Unmarked forms can be pro- and enclitic (e.g. gang-I ‘go I’, gosch-t ‘go you’, gom-mr ‘go we’ or gib-mr ‘give me’).←235 | 236→

In the standard variety as well as in Vorarlberg Alemannic gender can be expressed via the definite articles (der, die, das ‘the’) and the indefinite articles (ein, eine ‘a’). However, realisations differ (cf. Shrier 1965). The masculine article der (standard) is realised as dr or an (dr/an Maa ‘the/a man’) (dialect), the feminine article die (standard) is realised as d or a (d/a Frau ‘the/a woman’) (dialect), and the neutral article das (standard) is realised as s or a (s/a Huus ‘the/a house’) (dialect). When compared with the standard variety, dialectal realisations of the indefinite feminine and neutral articles show a syncretism (fusion of forms) e.g. a Frau / a Huus (dialect) vs. eine Frau / ein Haus (standard) ‘a woman / a house’. Another speciality of Vorarlberg Alemannic is the compulsory use of the definite article before names, such as dr Thomas goht is Huusthe Thomas goes into the house’ (e.g. cf. Penner/Kölliker Funk 1998 for Swiss German).

Vorarlberg Alemannic has three different cases (nominative, dative and accusative). Unlike the cases in the standard variety, the dialectal nominative and accusative case are identical (syncretism) for all three genders (cf. Shrier 1965): Er legt dr Stift anni (accusative) ‘He puts the pen down’, Dr Stift ist rot (nominative) ‘The pen is red’; vs. Er legt den Stift hin / Der Stift ist rot (standard).

Two dominant plural markers are characteristic for Vorarlberg Alemannic, such as umlaut plural and zero plural (e.g. sg. Hund, pl. Hünd, vs. standard pl. Hunde ‘dogs’; sg. Auto, pl. Auto, vs. standard pl. Autos ‘cars’). The standard plural ending -(e)n corresponds to the dialect plural ending -a(na) (e.g. sg. Häxa, pl. Häxana, vs. standard pl. Hexen ‘witches’). Feminine words ending with -a, such as Bluama ‘flower’ can be either realised with a zero plural or the plural ending -na (Bluama or Bluamana) (cf. Ruoff/Gabriel 1998).

Vorarlberg Alemannic has fewer verb tenses compared to the standard variety, which applies to all spoken Upper German dialects (including Viennese dialect). Preceding events are always expressed in the present perfect tense, e.g. I(ch) bin do gsi ‘I have been here’. Only a few frequent forms such as war ‘was’ are realised in the past simple tense in oral communication. The inflection of verbs is simplified since all three plural form endings are identical (e.g. mir/ihar/sie laufan ‘we/you/they walk’, whereas there are two forms in standard German: wir/sie laufen, ihr lauft).

2.3 Advantages and disadvantages of bilectal language acquisition

So far, hardly any research has been carried out on bilectal language acquisition. However, much uncertainty still exists about dialects having a negative impact on the language and literacy acquisition of the standard variety. In the 1970s, there was a great debate in Germany about the suspected negative influence of speak←236 | 237→ing a dialect on school success (cf. e.g. Ammon 1972; Hasselberg 1976). Studies on the situation in Austria such as the one described by Wiesinger (cf. 2008: 44) have examined parental concerns about complicating effects of dialect on their children’s language development. These concerns have been a controversial and much disputed subject in Switzerland and Vorarlberg. Primarily, it was debated which variety should preferably be spoken in kindergarten. The Swiss Conference of Cantonal Ministers of Education identified dialect as the major cause of the unsatisfactory Swiss results in the PISA tests (Programme for International Student Assessment) (cf. DLV 2010). Preliminary studies carried out by Landert (cf. 2007) and Gyger (cf. 2005) found that the simultaneous acquisition of two varieties can be extra challenging for multilingual children with little knowledge of German. These children benefit when exposed to a standard variety in childcare facilities such as nursery and kindergarten.

Further literature has emerged that offers contradictory findings about dialect. Löffler (cf. 2002) highlighted beneficial effects of dialect on monolingual children’s acquisition of literacy and found that children had no difficulties to align their writing with the standard variety. Another advantage was stressed by Berthele (cf. 2010) who found that dialect speakers had better receptive abilties when acquiring a foreign language. While Antoniou et al. (cf. 2016) assume that bilectal children speaking two closely related dialects and bilingual children have similar advantages in executive functions (e.g. working memory), Katsos (cf. 2016) claims that bilectal and bilingual children with any language mix show better executive performances than monolingual children. Also, Vangsnes et al. (cf. 2015) found that bidialectal literacy improves school achievement. Thus, the proximity of languages and varieties appears to have no influence – this view is also supported by Suter Tufekovic (cf. 2008). The reported evidence suggests that not only bilectalism is responsible for cognitive gains, but also systematic code switching between varieties.

3 The present study

The present study combines the results of two research projects using the same methodology.

3.1 The research projects EdUS and INPUT

The following findings are based upon two cooperative research projects, the project EdUS conducted in Vorarlberg and the project INPUT carried out in←237 | 238→ Vienna.3 Participants consisted of 65 mono- and bilingual children and their parents and caretakers. All participants were aged between 2;11 and 3;6 years at the beginning of the observation period, which was scheduled for a year and a half. At four different data points data was collected at the children’s homes as well as in kindergarten. A battery of language production and comprehension tests were administered to all children, spontaneous speech samples were recorded at home and kindergarten and interviews with all parents and caretakers were conducted.

The objective of both research projects was to identify influencing variables (such as socioeconomic status, amount of input, reading and media consumption in different languages). Furthermore, the influence of parental and caretakers’ language input on children’s language development was analysed. The aim was to investigate whether and how adults modify their child-directed speech when talking to multilingual instead of monolingual children and whether such a difference becomes apparent in the children’s language. We also wanted to investigate to what extent simultaneous language acquisition of a standard and a regional variety influences the children’s language development.

3.2 Research question

The main question of the current analyses was whether medial diglossia had an influence on the monolingual children’s language development aged between three and five. For this purpose, monolingual children from Vorarlberg and Vienna were compared regarding their development of vocabulary and grammar. The children from Vorarlberg had receptive and productive language skills in the regional variety, but only receptive language skills in the standard variety, whereas the children from Vienna had both receptive and productive skills in the standard variety.

3.3 Participants

The findings presented in this paper are based on six monolingual children, who were matched to three pairs according to socioeconomic status (SES), age, gender and birth order and went to kindergarten regularly (cf. Table 1). Each of the three pairs differed regarding their language environment (medial diglossia in←238 | 239→ Vorarlberg vs. standard variety in Vienna). All children that were examined came from families with high SES, which was determined by the highest educational degree of the parent who spends the most time with the child (referred to as the primary caretaker). In all cases, the primary caretaker had either Matura ‘A levels’ or a similar educational qualification (see Czinglar et al. 2015 for details of the SES classification in the present study). The children were three years old (+/- two months) at the beginning of the study and had been attending a childcare facility for at least ten months. Parents’ information about the child’s weekly media consumption (television and audio books) was only partly comparable. However, the numbers (hours/week) showed that all the children from Vorarlberg had regular standard language input.

Table 1: Participants of the present study


←239 | 240→

3.4 Method

The study was designed as follows: All six children were assessed at four sessions during a period of a year and a half. The first two sessions (at the age of approx. 3;0) and the second two sessions (at the age of approx. 4;6) were each three months apart. In between was a break of one year. Additionally, interviews with all parents and caretakers were conducted in order to collect information about the socioeconomic status, language input and the childcare facility, among other details.

Language comprehension and production tests were part of the design to systematically assess and compare the children’s language development. The children’s language skills in receptive vocabulary were tested in the first and the third session. A German research version of the PPVT-4 (Peabody Picture Vocabulary Test, Dunn/Dunn 2007) was administered. After hearing a word spoken in their L1 (standard or dialect, spoken by a native standard or dialect speaker, respectively), the children were instructed to select the picture that best matched the word’s meaning (noun, verb or adjective). The children’s language skills in receptive and productive grammar were tested in the second and fourth session. Therefore, two subtests of LiSe-DaZ (a standardised test of German as a second language, Schulz/Tracy 2011) were again administered by a native standard speaker (for the Viennese children) or a native dialect speaker (for the children from Vorarlberg). In the first subtest, the children’s utterances were elicited using a picture story (language production). In the second subtests, the children were presented with wh-questions they had to answer (language comprehension). All collected data were analysed.

Videos of spontaneous speech interactions (between caretaker and child/children or parent and child) were recorded in all four sessions. Adults were encouraged to act as naturally as possible during interaction. Different settings were recorded (e.g. role plays, card and board games or reading from a picture book) with different quantity of input. Some settings (e.g. card or board games) turned out to be less language stimulating than others (e.g. reading from a picture book or role plays) (see Templ et al. 2018 on situation-dependent differences in caretakers’ input). Thirty minutes of each recording were selected to analyse the spontaneous speech (see Korecky-Kröll/Czinglar 2017). First, the language data was transcribed with the programme CLAN (Computerised Language Analysis) from CHILDES (MacWhinney 2000) by using the CHAT transcription format. Second, another researcher controlled the transcripts by listening to the recordings. Then the data was semi-automatically coded for part of speech information and morphology using an electronic word form list or «lexicon« for Standard German (see Korecky-Kröll 2017 on details regarding the method). An extra word←240 | 241→ form list was created for Vorarlberg Alemannic words that differed beyond pronunciation when compared to the standard variety. The vocabulary varied greatly in all recordings. Therefore, the word form list was primarily consulted to identify the child’s grammatical inventory, since the use of grammatical categories (such as prepositions, articles, modal and auxiliary verbs, subordinating conjunctions, pronouns and wh-questions) is an indicator for language development.

4 Results

In this section, we present test and spontaneous speech data on the lexical and grammatical development of children growing up in a standard- and in a dialect-speaking environment.

4.1 Receptive and productive language tests

The data in Table 2 show that all six children scored age-appropriate results regarding receptive vocabulary in PPVT-4 in both sessions4. No significant differences were found between the children growing up with a regional variety or a standard variety. However, some differences were revealed in the grammar-oriented test LiSe-DaZ: CHR (child from Vorarlberg) scored no age-appropriate results regarding productive language development in LiSe-DaZ in the first session regarding the use of focus particles and lexical verbs. However, CHR caught up until the second session and all six children achieved age-appropriate results in all receptive and productive grammar subtests. Table 2 presents exemplary scores obtained from two subtests of LiSe-DaZ. The variable ESS measures verb placement according to four milestones for the sentence bracket (cf. Schulz/Tracy 2011: 33–35), which are the same in the standard and the dialect variety. All the children reached the highest stage of development in their fifth year of life. They produced correct subordinate clauses with the finite verb placed at the end of the clause in the fourth testing session. Two children (a matched pair) already formed correct subordinate clauses in their fourth year of life. LAR (a child from Vorarlberg) had more difficulties in←241 | 242→ understanding wh-questions (WHQ) than SOS (her Viennese matching partner) in the first testing session. However, both children scored similarly in the second testing session. Again, neither advantages nor disadvantages were found between the children growing up with a regional or a standard variety.

Table 2: Raw scores of PPVT and two LiSe-DaZ measures (ESS, WHQ)


4.2 Analysis of spontaneous speech data

First, we present a comparison of all grammatical and lexical categories, second, the results for the different grammatical categories such as articles, prepositions, modal and auxiliary verbs, followed by personal and reflexive pronouns in the dative case, subordinating conjunctions and finally wh-words.

4.2.1 Grammatical and lexical categories

Initially, all spoken word tokens were divided into lexical and grammatical categories. We excluded pragmatic categories (such as interjections, communicators and onomatopoetics) as they are not relevant for assessing a child’s level of speech and language development (cf. Le Normand et al. 2013). Nouns, lexical verbs, adjectives and adverbs were coded as lexical categories. Functional words such as pronouns, determiners, prepositions, conjunctions, particles, pronominal adverbs, modal and auxiliary verbs were coded as grammatical categories. The recordings 1 and 2 (each 30 minutes, fourth year of life) as well as the recordings 3 and 4 (each 30 minutes, fifth year of life) were taken together for data analysis because it is assumed that the children’s language development does not show much change within three months.

As shown in Figures 1 and 2, two out of three Viennese children spoke more and used more grammatical categories when comparing the earlier and later recordings. This could indicate an improvement in grammar acquisition. Interestingly, no increase in the use of lexical and grammatical categories was detected←242 | 243→ in the spontaneous speech samples of the children from Vorarlberg. Their quantity of categories was relatively stable. They expressed approximately 500 to 1000 grammatical tokens per recording hour (30+30 minutes), whereas the Viennese children produced more utterances with approximately 700 to 1200 grammatical tokens in all recordings. Overall, the Viennese children spoke more than the children from Vorarlberg. Yet, the difference between the use of grammatical and lexical categories in both groups was not significant (χ2 = 0.366, p = 0.545).

We also analysed parental input in order to exclude the possibility that differences in parental input were the cause for different total numbers of tokens (see Figures 3 and 4). The total number of tokens was relatively stable for all parents over the whole testing period. No significant difference between the use of grammatical and lexical categories was evident (χ2 = 3.222, p = 0.073). Interestingly, the Viennese parents talked more than parents from Vorarlberg, meaning that the children reached a greater total number of tokens not on account of more speaking time, but rather due to more talkative parents and therefore larger input. Both groups of parents showed a similar speaking tempo. Therefore, the difference in talking could be attributed to different settings of interaction (see section 4).

Figures 1–4: Lexical and grammatical categories (children Vienna vs. Vorarlberg, parents Vienna vs. Vorarlberg)


←243 | 244→

4.2.2 Articles, prepositions, modal and auxiliary verbs

Overall, the Viennese children uttered more tokens regarding the four main grammatical categories (determiners, prepositions, modal and auxiliary verbs) and made more progress within a year than children from Vorarlberg. Again, this difference was not statistically significant (χ2 = 1.667, p = 0.196). The Viennese parents showed similar numbers of tokens in the mentioned grammatical categories. A positive relation between the number of tokens and the total quantity of words was found.5 Parents from Vorarlberg used significantly more grammatical tokens than the Viennese parents, based on the fact that they all spoke less. Nevertheless, differences in parental input did not show in children’s output. Again, no significant correlation was observed.

4.2.3 Personal and reflexive pronouns in the dative case

The Viennese children used both a greater number of personal and reflexive pronouns in the dative case and showed a greater diversity of forms. Furthermore, a greater increase from the fourth to the fifth year of life was evident. All in all, the Viennese children were more talkative. However, none of these differences were statistically significant (χ2 = 0.077, p = 0.781). Data comparison revealed that also the Viennese parents had used more diverse pronouns in dative case, but again the difference was not significant (χ2 = 0.134, p = 0.714).

4.2.4 Subordinating conjunctions

The Viennese children uttered more subordinating conjunctions and on top more diverse forms. An obvious increase of subordinating conjunctions was detected within the fourth and fifth year of life. There was a significant difference when taking account of the total number of tokens (χ2 = 6.624, p = 0.010). This result suggests a more advanced grammar development of the Viennese children. The parents of the Viennese children also produced more and diverse subordinating conjunctions during one hour of conversation than the parents from Vorarlberg. However, this outcome was not significant in relation to the total number of tokens (χ2 = 0.037, p = 0.847).←244 | 245→

4.2.5 Wh-words

Wh-words (pronouns, determiners, adverbs) were analysed as a last category. A surprising result was that the children from Vorarlberg used significantly more wh-words than the Viennese children (χ2 = 49.850, p < 0.001). They used more wh-words in the earlier recordings. The quantity of produced wh-words decreased notably in later recordings. We know that younger children with less developed speech and language skills ask more wh-questions than older children. The children from Vorarlberg could have a prolonged ‘why-question-period’ which is a typical stage in the speech and language development of children in their third year. However, data analysis showed that not only why-questions were formed. Literature suggests that parents with younger and less advanced children ask more wh-questions (cf. e.g. Newport et al. 1977; Cameron-Faulkner et al. 2003), which in turn could animate their children to also raise more wh-questions. And in fact, parents from Vorarlberg used slightly more wh-words than Viennese parents regarding the total number of tokens, but this difference was not significant (χ2 = 3.514, p = 0.061). Two out of three parents from Vorarlberg and one Viennese mother showed a decrease of wh-words when their children were between 4 and 5 years old. This outcome corresponds with results reported in the literature (cf. Newport et al. 1977; Cameron-Faulkner et al. 2003; see also Korecky-Kröll, submitted, for the entire sample of 29 monolingual Viennese children that participated in the INPUT project).

5 Discussion

The purpose of the current study was to examine and detect differences in children’s speech and language acquisition when growing up in environments with language variation. No significant differences were observed between the children from Vienna growing up with a standard variety and the children from Vorarlberg who are exposed to a regional variety and the standard variety (medial diglossia). Both groups showed comparable results in all speech and language tests. Furthermore, they also showed similar results in spontaneous speech samples. Overall, the children and two parents from Vorarlberg were less talkative than their reference group from Vienna. Going back to the video recordings, this discrepancy was not related to speech tempo, but was rather due to different communicative settings: The interactions where children and parents spoke less involved puzzles, card and board games, which are less language stimulating in generally (cf. Templ et al. 2018). Some observed differences between the distribution of lexical and grammatical categories and most grammatical subcategories (articles, preposi←245 | 246→tions, modal and auxiliary verbs as well as personal and reflexive pronouns in dative case) were not significant.

But we found a significant difference regarding subordinating conjunctions and wh-words: The Viennese children used more subordinating conjunctions, whereas the children from Vorarlberg produced more wh-words. These differences may indicate different levels of speech and language development. On the one hand, subordinate clauses are an indicator of a more advanced grammar development. On the other hand, both groups produced the same number of subordinating conjunctions in the standardised subtest of LiSe-DaZ. The higher frequency of wh-words generally indicates a less advanced speech and language development. Hence, it could be hypothesised that the children from Vorarlberg showed a prolonged phase of why-questions. However, the data revealed that not only why-questions were used and that the parents from Vorarlberg also raised more questions than the parents from Vienna. A possible explanation for this might be that younger and less advanced children are generally exposed to more questions. Furthermore, it has to be taken into account that different settings of interaction offer different opportunities of asking questions (cf. Templ et al. 2018). Whether the higher frequency of wh-words can be attributed to less advanced language development or to individual variation cannot be determined for sure. It is likely, though, that children from Vorarlberg asked more questions than the Viennese children based on their parental input (cf. Cameron-Faulkner et al. 2003), which in turn contained a greater number of questions.

Overall, our study shows no disadvantage for children from Vorarlberg growing up in an environment of medial diglossia.6 Even if some grammatical categories seem less differentiated in the children from Vorarlberg; growing up with medial diglossia has had no negative impact on their L1 acquisition so far. Of course, a follow-up study on their development in primary school would be desirable to verify whether this also holds for their later language development and to analyse if they will have more difficulties in literacy acquisition than their standard-speaking peers from Vienna. However, even a positive impact of bilectal acquisition on children’s general development such as higher cognitive flexibility,←246 | 247→ working memory performance, attention control, metalinguistic awareness and creativity may be expected, but an investigation of these general cognitive abilities would have gone beyond the linguistic focus of the present study. A considerable amount of studies of the last 15 years identified these advantages for bilingual children (for an overview see Bialystok 2016), and recent studies (cf. e.g. Vangsnes et al. 2015; Antoniou et al. 2016) contributed additional evidence suggesting that these findings also apply to bilectal children.

6 Conclusion

This study has raised the crucial question whether parents should favour the regional or standard variety when speaking to their children. The discussed findings and evidence from existing literature indicate that parents should always favour the language or variety they speak the best (cf. Buell et al. 2010). Children will be exposed to other languages or varieties they need to acquire as soon as they enter kindergarten. A rich and natural input in at least one language or variety should be ensured from a child’s birth onwards. This means that parents who are competent dialect speakers and feel at home in the dialect should speak dialect to their children. Parents who are more proficient and at home in a standard variety and struggle for authentic dialect use do not need to switch to dialect, on the other hand. Parents need to ensure that they always talk in the most genuine and natural way with their children.


Ammon, Ulrich (1972): Dialekt, soziale Ungleichheit und Schule. Weinheim: Beltz

Ammon, Ulrich (1995): Die deutsche Sprache in Deutschland, Österreich und der Schweiz. Das Problem der nationalen Varietäten. Berlin: de Gruyter.

Ammon, Ulrich (2003): Dialektschwund, Dialekt-Standard-Kontinuum, Diglossie: Drei Typen des Verhältnisses Dialekt – Standardvarietät im deutschen Sprachgebiet. In: Androutsopoulos, Jannis K. /Ziegler, Evelyn (Hg.) „Standardfragen“. Soziolinguistische Perspektiven auf Geschichte, Sprachkontakt und Sprachvariation. Frankfurt a. M. u. a.: Lang, 163–171.

Antoniou, Kyriakos/Grohmann, Kleanthes K./Kambanaros, Maria/Katsos, Napoleon (2016): The effect of childhood bilectalism and multilingualism on executive control. In: Cognition 149, 18–30.

Berthele, Raphael (2010): Dialekt als Problem oder Potenzial? Überlegungen zur Hochdeutschoffensive in der deutschen Schweiz aus Sicht der Mehrsprachig←247 | 248→keitsforschung. In: Bitter Bättig, Franziska/Tanner, Albert (eds.): Sprachen lernen – durch Sprache lernen. Zürich: Seismo, 37–52.

Bialystok, Ellen (2016): Bilingual education for young children: review of the effects and consequences. In: International Journal of Bilingual Education and Bilingualism. DOI: 10.1080/13670050.2016.1203859.

Buell, Martha J./Gamel-McCormick, Michael/Unger, Donald G. (2010): Why families should matter for early reading first. In: McKenna, Michael C./Walpole, Sharon/Conradi, Kristin (eds.): Promoting early reading: research, resources and best practices. New York, NY: Guilford Press, 164–174.

Cameron-Faulkner, Thea/Lieven, Elena/Tomasello, Michael (2003): A construction based analysis of child directed speech. In: Cognitive Science 27, 6, 843–873.

Czinglar, Christine/Korecky-Kröll, Katharina/Uzunkaya-Sharma, Kumru/Dressler, Wolfgang U. (2015): Wie beeinflusst der sozioökonomische Status den Erwerb der Erst- und Zweitsprache? Wortschatzerwerb und Geschwindigkeit im NP/DP-Erwerb bei Kindergartenkindern im türkisch-deutschen Kontrast. In: Köpcke, Klaus-Michael/Ziegler, Arne (eds.): Deutsche Grammatik in Kontakt. Deutsch als Zweitsprache in Schule und Unterricht. Berlin: de Gruyter, 207–240.

DLV (2010): Die Verwendung von Hochdeutsch (bzw. Standarddeutsch) und Deutschschweizer Dialekten im Kindergarten. Zürich: DLV.

Dunn, Lloyd M./Dunn, Douglas M. (2007): Peabody Picture Vocabulary Test (PPVT). Minneapolis, MN: Pearson Assessments.

Gabriel, Eugen (1963): Die Mundarten an der alten churrätisch-konstanzischen Bistumsgrenze im Vorarlberger Rheintal. Marburg: Elwert.

Gyger, Mathilde (2005): Projekt Standardsprache im Kindergarten. Schlussbericht. Basel. URL: [accessed 14.06.2017]

Haid, Andrea (2011): Früherfassung von sprachlichen Auffälligkeiten im Kindesalter im Fokus des Vorarlberger Dialektes. Universität Wien: Dissertation.

Hasselberg, Joachim (1976): Dialekt und Bildungschancen. Eine empirische Untersuchung an 26 hessischen Gesamtschulen als Beitrag zur soziolinguistischen Sprachbarrierendiskussion. Weinheim: Beltz.

Katsos, Napoleon (2016): Speaking dialects trains the brain in the same way as bilingualism. URL: [accessed 14.06.2017].

Kolde, Gottfried (1981): Sprachkontakte in gemischtsprachigen Städten. Wiesbaden: Steiner.

Korecky-Kröll, Katharina (2017): Kodierung und Analyse mit CHILDES: Erfahrungen mit kindersprachlichen Spontansprachkorpora und erste Arbeiten←248 | 249→ zu einem rein erwachsenensprachlichen Spontansprachkorpus. In: Resch, Claudia/Dressler, Wolfgang U. (eds.): Digitale Methoden der Korpusforschung in Österreich. Wien: Verlag der Österreichischen Akademie der Wissenschaften, 85–113.

Korecky-Kröll, Katharina (submitted): Requests in first language acquisition of German: Evidence from high and low SES families. Submitted to: Stephany Ursula/Aksu-Koç, Ayhan/Voeikova, Maria D. (eds.): Development of Modality in First Language Acquisition. Berlin: De Gruyter.

Korecky-Kröll, Katharina/Czinglar, Christine (2017): Bairisch-alemannischer Sprachkontakt in einem Wiener Kindergarten: Eine Fallstudie. In Lenz, Alexandra N./Breuer, Ludwig M./Kallenborn, Tim/Glauninger, Manfred M./Patocka, Franz (eds.): Bayerisch-österreichische Varietäten zu Beginn des 21. Jahrhunderts – Dynamik, Struktur, Funktion. 12. Bayerisch-Österreichische Dialektologentagung. Stuttgart: Franz Steiner Verlag, 199–219.

Landert, Karin (2007): Hochdeutsch im Kindergarten? Eine empirische Studie zum frühen Hochdeutscherwerb in der Deutschschweiz. Bern: Peter Lang.

Lenhard, Alexandra/Lenhard, Wolfgang/Segerer, Robin/Suggate, Sebastian (2015): Peabody Picture Vocabulary Test. Frankfurt a. M.: Pearson Assessment and Information GmbH.

Le Normand, Marie-Thérèse/Moreno-Torres, Ignacio/Parisse, Christophe/Dellatolas, Georges (2013): How do children acquire early grammar and build multiword utterances? A corpus study of French children aged 2 to 4. In: Child Development 84, 2, 647–661.

Löffler, Cordula (2002): Analphabetismus in Wechselwirkung mit gesprochener Sprache. Zu Sprachentwicklung, Sprachbewusstsein, Variationskompetenz und systematisch fundierterFörderung von Analphabeten. Aachen: Alfa Zentaurus.

MacWhinney, Brian (2000): The CHILDES Project: Tools for Analyzing Talk. Mahwah, NJ: Erlbaum.

Newport, Elissa L./Gleitmann, Henry/Gleitmann, Lila R. (1977): Mother, I`d rather do it myself: some effects and non-effects of maternal speech style. In: Snow, Catherine/Ferguson, Charles (eds.): Talking to children. Language input and acquisition. Cambridge: Cambridge University Press, 109–149.

Penner, Zvi (1993): W-Morphology in the COMP System of Bernese Swiss German and the Licensing of Empty Operators in the Prefield Position. In Abraham, Werner/Bayer, Josef (eds.): Dialektsyntax. Opladen: Westdeutscher Verlag, 201–212.

Penner, Zvi/Kölliker Funk, Meja (1998): Therapie und Diagnose von Grammatikerwerbsstörungen. Ein Arbeitsbuch. Luzern: Edition SZH.←249 | 250→

Ruoff, Arno/Gabriel, Eugen (1998): Die Mundarten Vorarlbergs. Ein Querschnitt durch die Dialekte des Landes. Mit einem Katalog des Tonarchivs der Mundarten Vorarlbergs. Graz: Neugebauer.

Schallert, Oliver (2010): Syntax des Vorarlberger Alemannischen: Ergebnisse eines Forschungsprojekts. In: Montfort. Vierteljahresschrift für Geschichte und Gegenwart Vorarlbergs 62/1, 35–67.

Schulz, Petra/Tracy, Rosemarie (2011): Linguistische Sprachstandserhebung – Deutsch als Zweitsprache (LiSe-DaZ). Göttingen: Hogrefe.

Shrier, Martha (1965): Case Systems in German Dialects. In: Language 41, 3, 420–438.

Suter Tufekovic, Carol (2008): Wie mehrsprachige Kinder in der Deutschschweiz mit Schweizerdeutsch und Hochdeutsch umgehen. Eine empirische Studie. Frankfurt am Main: Peter Lang.

Templ, Viktoria/Weichselbaum, Maria/Korecky-Kröll, Katharina/Dressler, Wolfgang U. (2018): Deutschspracherwerb ein- und zweisprachiger Wiener Kindergartenkinder: Der Einfluss des sozioökonomischen Status der Familie, des sprachlichen Hintergrunds und der Sprechsituationen. In: Carvill Schellenbacher, Jennifer/Dahlvik, Julia/Fassmann, Heinz/Reinprecht, Christoph (eds.): Migration und Integration – wissenschaftliche Perspektiven aus Österreich. Jahrbuch 4/2018. Göttingen: V&R unipress, Vienna University Press, 195–210.

Vangsnes, Øystein A./Söderlund, Göran B. W./Blekesaune, Morten (2015): The effect of bidialectal literacy on school achievement. In: International Journal of Bilingual Education and Bilingualism 20, 3, 346–361.

Wiesinger, Peter (1983): Die Einteilung der deutschen Dialekte. In: Besch, Werner/Knoop, Ulrich/Putschke, Wolfgang/Wiegand, Herbert E. (eds.): Dialektologie. Ein Handbuch zur deutschen und allgemeinen Dialektforschung. 2. Teilband. Berlin: de Gruyter, 807–900.

Wiesinger, Peter (2008): Das österreichische Deutsch in Gegenwart und Geschichte. Wien: LIT-Verlag.←250 | 251→

1 This notation for age (years;months) is common in child language acquisition research: The semicolon is used instead of a decimal point because a year consists of twelve months.

2 The work on this paper was funded by the Vienna Science and Technology Fund (WWTF: SSH11-027) and the Land Vorarlberg (county of Vorarlberg). We presented a former version of this paper at the conference “German in Austria and Other Pluricentric Contexts” (7.–9. July 2016 in Vienna) and published it in the “SAL-Bulletin” (September 2016). We thank Wolfgang U. Dressler and Rudolf de Cillia for their support and feedback, as well as numerous students for their help with transcribing. We would like to express our sincere gratitude to all the families, caretakers and children that took part in the study.

3 The project INPUT was conducted at the University of Vienna at the Department of Linguistics and supervised by Wolfgang U. Dressler. The project EdUS was conducted at the Schweizer Hochschule für Logopädie Rorschach (Swiss University of Applied Sciences of Speech and Language Pathology Rorschach) and supervised by Andrea Haid.

4 As there was no standardised German version of the PPVT at the time of data collection, a research version translated from the original English version was used. Therefore, results cannot be compared to age norms, but only to the entire monolingual sample of the INPUT project consisting of 29 children (15 from high socioeconomic status). The mean raw score of high SES children for the first PPVT session at age 3 was 58.2 (min: 24, max: 84), whereas it was 97.47 (min: 40, max: 149) for the second PPVT session at age. Since 2015, there is a standardised German version of the PPVT (cf. Lenhard et al. 2015), which will be used in subsequent projects.

5 We included pragmatic categories (see footnote 6) in the following data analysis regarding the total number of tokens.

6 The receptive acquisition of the standard variety in Vorarlberg had no impact on the speech and language development of monolingual children aged 3–5 years. However, the productive acquisition of the standard variety could have an effect: A child of the same age from the Viennese sample who was exposed to Vorarlberg Alemannic at home and to the standard variety in kindergarten showed a slightly delayed speech and language development in both dialect and standard German, but the delay was more pronounced in the standard than in the dialect (cf. Korecky-Kröll/Czinglar 2017).