Family Language Policies of Multilingual Families during the COVID-19 Pandemic: Evidence from Cyprus, Estonia, Germany, Israel, and Sweden

Abstract

This study explored the language and literacy practices of multilingual families in Cyprus, Estonia, Germany, Israel, and Sweden during the COVID-19 pandemic. The study focuses on the different roles of family members in language transmission in order to understand whether these practices might have been influenced by the COVID-19 pandemic. We aimed to answer two key research questions: RQ1, whether and how the pandemic conditions affected the heritage language, societal language acquisition, and heritage language literacy learning environments in the five countries examined (Cyprus, Estonia, Germany, Israel, and Sweden); and RQ2, what is the nature of child and parental agency in facilitation of the possible changes in the corresponding five countries? Fifty semi-structured interviews (ten in each country) were conducted. The data highlighted the factors that triggered changes in family language policy during the pandemic and the role of the child’s agency, parents, extended family, and social network during this period. Based on our findings, we argue that the pandemic conditions gave the children new opportunities for agency when it comes to language and literacy choice and communication with extended family members. This even facilitated new sources of input and suggested the active role of a child as an agent in shaping family language policy in the family.

1. Introduction

The COVID-19 pandemic has led and is still leading to unprecedented changes worldwide, and it continues to affect societies and communities in a number of different ways. The world crisis is highlighting many challenges, yet it is also providing new opportunities, giving us a possibility to understand the cost of the social and health crisis and what can be learned from it for the future. Multilingual communities were reported to undergo a “dual pandemic” (Lou et al. 2021): in addition to the COVID-related challenges, heritage language (hereafter HL)-speaking groups experienced limitations in the governmental provision of public health information in HLs (e.g., Piller et al. 2020). Furthermore, the stigmatizing of linguistic minority groups has been reported in the recent research, especially against Asian groups (Dovchin 2020; Lou et al. 2021; Zhu 2020). Finally, minority communities were disadvantaged in access to medical services and digital education due to limited resources and the lack of translation services for minority-language-speaking communities (e.g., Zhu 2020).

Despite these challenges, the pandemic may have been a pervasive catalyst for a language practice shift in HL-speaking communities. HL usage in multilingual families may have potentially increased during the periods of social isolation, when there was a decrease not only in educational and leisure activities but also interactions with peers in the societal language (hereafter SL). Therefore, the pandemic has potentially provided new opportunities for maintaining and developing the HL.

However, little is still known about to what extent the pandemic triggered changes in micro-factors such as parents’ views on multilingualism, language choice and use, and everyday interaction practices in the family, as well as macro-factors, such as societal structure, language policies in the country, and the value assigned to multilingualism by the host community (Curdt-Christiansen and La Morgia 2018; Spolsky 2004, 2012).

The current study aimed to contribute to the scarcity of research on how family language policy (FLP) might be changed rapidly as an answer to the challenging circumstances. In order to understand how FLP might have been influenced by the pandemic, this study explored language and literacy practices of multilingual families in Cyprus, Estonia, Germany, Israel, and Sweden during the pandemic, as well as the different roles of family members in language use, maintenance, and transmission. The information on the effect of the pandemic on multilingual communities is very important, especially for countries with large percentages of HL-speaking communities. The choice of HL-Russian was motivated by the fact that maintenance and development of HL-Russian previously has been well documented in the pre-COVID-19 period in diverse contexts (e.g., Karpava et al. 2020; Otwinowska et al. 2021); however, little is known about HL-Russian transmission and maintenance in the COVID-19 period. In the current study, we use the term “HL” to refer to Russian as a language, which is not the dominant SL in each of the five countries.

Our study aimed to answer two key research questions: RQ1, whether and how the pandemic conditions affected the HL, SL acquisition, and HL literacy learning environments in the five countries examined (Cyprus, Estonia, Germany, Israel, and Sweden); and RQ2, what is the nature of child and parental agency in facilitation of the possible changes in the corresponding five countries?

This article is organized as follows. We begin with an overview of the literature on family language policy, child agency, parental agency, HL and SL acquisition, and HL literacy learning environments. Then, we present the contextual background to Russian as an HL in the five countries examined (Cyprus, Estonia, Germany, Israel, and Sweden). Following this, we introduce data on school closure per country during the COVID-19 pandemic that is relevant for the research questions of this comparative study. This is followed by a description of the research methods, after which we describe our results, focusing on similarities and differences regarding the countries under investigation. The conclusion summarizes our improved understanding of the FLP and literacy practices of multilingual families in the five countries during the pandemic.

2. Family Language Policy

The family language policy (FLP) involves ideologies and approaches to how languages are managed, learned, and negotiated within individual families (Schwartz 2010; Curdt-Christiansen and Huang 2020). FLP research is interdisciplinary, and it takes into consideration both internal and external factors (King and Fogle 2013; Spolsky 2019; Hollebeke et al. 2020; Karpava 2022a). In particular, the internal factors include language management strategies in the family and their implicit and explicit language choices; parental expectations, attitudes, and efforts for HL maintenance and transmission; affective domain; the role of child agency and communication with siblings and relatives. As for the external factors, these comprise top-down ideologies, schooling, socioeconomic status of the family, well-being, social network, communication with educators and experts, quality, quantity of input to the HL and the SL outside the home environment, and collaboration with mainstream and community schools.

Family plays the primary role in socialization (Melo-Pfeifer 2015) and is traditionally seen as a driving force in HL maintenance and transmission. A multilingual family can be considered a “community of practice” (Lanza 2007, p. 47). As children grow older, they interact more and more with their environment and develop into independent language agents who influence their parents’ language choices (e.g., Smith-Christmas 2020, 2021). Consequently, all family members are under a constant mutual influence. A number of studies on language use and language maintenance in immigrant families show differences between immigrant generations (Lambert 2008, to name just one). Furthermore, differences in attitudes towards HL maintenance can also be observed within a single generation of immigrants, depending on linguistic ideologies (e.g., Kopeliovich 2013), which are “morally and politically loaded representations of the nature, structure, and use of languages in a social world” (see Woolard 2020: 1 and references therein).

The role of children as actors who can initiate changes has been emphasized in the literature previously (Said and Zhu 2019; Smith-Christmas 2021). Yet, there is still a need for more research to understand the role of the child’s agency in shaping FLP, which is related to language exposure that the child gets and in turn is associated with the child’s linguistic outcomes in both languages. Children’s agency is related to their willingness and desire to or, on the contrary, their reluctance and resistance towards the use of the HL and the SL and, as a result, the development of relevant literacy skills. The quality and quantity of the language that children are exposed to and other aspects of children’s linguistic environments at home and outside are in strong correlation with the success of language acquisition (for an overview see Armon-Lotem and Meir 2019 and numerous studies cited in it). In the general course, all children from multilingual families are reported to eventually learn the SL due to extensive schooling in this language (De Houwer 2020). As for HL maintenance, it is not always the case (Mieszkowska et al. 2017). During the COVID-19 pandemic, the exposure to the SL might have diminished due to lockdown measures (see the subsection below entitled “The COVID-19 pandemic impact on school closure per country”), thereby changing the balance between the SL and the HL in multilingual communities. At the same time, the pandemic has driven new opportunities in the realm of digital communication and digital education (Daniel 2020).

A handful of projects have investigated the effects of the pandemic on language use in multilingual families. The project led by Ludovica Serratrice, in which 700 multilingual families speaking as many as 95 different HLs took part in an online survey in Britain and Ireland (see http://www.education.ox.ac.uk/events/copy-copy-copy/, accessed on 3 November 2023), showed that, before the pandemic, children favored the SL (English in the UK and Ireland). However, during quarantine, there was a shift toward HL in some families. In Norway, a project led by Elizabeth Lanza (see García González et al. 2021) also focused on the effects of the pandemic on the HL and SL use by collecting data from almost 200 multilingual families. The results in Norway also indicated that younger children switched to the HL more often during lockdowns. Recently, Murrmann (2021) conducted an online survey during the month of February 2021 in which data were collected from 157 families residing in 42 different countries who speak at least 35 different languages, having a total of 339 children. The author reported that families’ opinions with respect to the effects of the pandemic on language situation could be split broadly into four main groups. The largest group (43% of the respondents) reported a positive impact of the COVID-19 pandemic on the language situation at home, stating that the pandemic restored the balance between the languages in the family’s repertoire and helped support the acquisition of the previously less used languages. A second group (31%) reported an overall negative effect on the language situation at home, stating that the SL usage dropped dramatically, and its acquisition was not sufficiently supported. And, finally, 21% reported no changes. This trend was mostly observed in families with young children who were not concerned with school duties. The remaining 4% stated that it was too early or impossible to assess the consequences of the COVID-19 pandemic on the language situation at home. Strikingly, the author reported that 65% of the families tried to introduce a different language policy compared to the one they had had prior to the COVID-19 outburst. In addition, a recent study by Sheng et al. (2021) which relied on exploratory analyses reported a change in the linguistic environment of bilingual English–Mandarin-speaking preschoolers. The increased use of Mandarin between the child and his/her parents in the COVID sample was noted as compared to that of the pre-COVID sample. In contrast, a study by Sun et al. (2023) investigating the effects of COVID on language development in English–Mandarin bilingual children in Singapore showed no change in language input among nuclear members; however, the amount and frequency of conventional and digital media materials and activities increased during the pandemic, which is suggested to positively influence language proficiency in English, while weakening abilities in Mandarin in the post-COVID period.

To conclude, the scarce available research seems to indicate that FLP in multilingual families might have undergone changes during the COVID-19 pandemic. The current study aimed to test whether previous findings are generalizable to other multilingual communities. With the help of a qualitative methodology, we investigated language use and language policy in multilingual families with HL-Russian across five countries, which vary in their SLs and with respect to their lockdown policies. Our goal was to further examine the directionality of the change and the driving forces of these changes based on the multilingual families with the same HL; in our case, this is Russian.

3. The Position of Russian in the Five Countries under Study

We investigated experiences of multilingual families raising their children with HL-Russian in five different countries, which differ with respect to the country and family language policy. Historically, these countries have had different migration waves (repatriates vs. old and new migrants vs. old and new minorities), as well as reasons for migration. Furthermore, these countries have differed with respect to HL-Russian maintenance policies at a country level and within a family. Finally, as described below, the five cases differ in the country and HL-Russian community size.

Russian is widely used by the large number of Russian immigrants in three of the countries included in this study: Cyprus, Estonia, and Israel. However, before the war in Ukraine, only in Cyprus was Russian perceived to be a commodity due to tourist flow, immigration, international marriages, cultural and religious ties, political cooperation, investments, and transnational corporations. Russian has developed into a new lingua franca in Cyprus, and even Cypriots are willing to learn it as a foreign language to increase their chances of better employment (Karpava 2022b).

Despite the fact that Russian has had a long history in Estonia and most Russian speakers have lived in Estonia for their entire lives, the official Estonian language is now dominant and prestigious in the country. To get Estonian citizenship, one must pass an Estonian language proficiency examination (Zabrodskaja 2009). In addition, the secondary education is available only in Estonian. This influences parents’ decisions about their children’s (pre-)primary education with respect to the SL/HL/immersion options (Seppik and Zabrodskaja 2022). An increasing number of Russian- and other-language-speaking students are enrolling in Estonian-medium schools. In Estonia, bilingualism that includes non-Estonian and Estonian is currently very much in demand among non-Estonians (Zabrodskaja 2014).

Today, Russian is the most frequently spoken immigrant language in Israel, after Hebrew and Arabic (Meir et al. 2021). Many members of the Russian-speaking community in Israel are interested in maintaining the Russian language and culture and transmitting Russian to the next generations (Otwinowska et al. 2021). Despite an interest in maintenance of the Russian language and cultural heritage, studies show that the ubiquitous presence of Russian vis-à-vis other languages in Israel is gradually decreasing.

In Germany and Sweden, Russian does not work as a lingua franca. The situation is still somewhat similar to the three other countries covered by our study: Russian is the HL for people of many different ethnicities and nationalities. However, the Russian speakers make up a rather heterogeneous group depending on the place of residence and the linguistic environment, thus leading to different possibilities for HL transmission and maintenance.

The community of Russian-speaking immigrants in Germany is one of the largest consisting of people from the Soviet Union or its successor states, who immigrated to Germany in the 1990s or at the beginning of the 2000s (Ritter 2021). Most of them are Russian-speaking people with German heritage (see more in Baur et al. 2019), who, however, mainly had little or no command of the German language by their arrival and considered Russian to be their mother tongue and often their only language (Dietz and Roll 2019).

In Sweden, Russian is one of the non-official minority languages. The Russian-speaking community is rather small (around 0.3% of the country’s population; see Parkvall 2016). However, the number of Russian-speaking immigrants is steadily increasing. Most Russian speakers are concentrated in the major cities, and, just as in Israel, they are interested in transmitting Russian language to the next generation (Otwinowska et al. 2021).

4. The COVID-19 Pandemic Impact on School Closure per Country

According to the UNESCO Global Dataset on the Duration of School Closures (2022) (https://data.humdata.org/dataset/global-school-closures-covid19, accessed on 25 June 2022), during the year of the COVID-19 pandemic, schools were fully and/or partially closed for an average of six months, which is equivalent to two-thirds of an academic year. However, there were large differences among countries with respect to the duration of full and partial lockdowns (see

Table 1. School closures during the pandemic in five countries for the period of March 2020–March 2022 (UNESCO global dataset on the duration of school closures, https://data.humdata.org/dataset/global-school-closures-covid19 (accessed on 25 June 2022).

Country	Duration of Full and Partial School Closures (in Weeks)	Duration of Full Closures (in Weeks)	Duration of Partial Closures (in Weeks)
Cyprus	28	13	15
Estonia	26	15	11
Germany	38	14	24
Israel	33	16	17
Sweden	24	0	24

for data per country). As a response to the COVID-19 lockdowns, education settings were transferred to emergency remote teaching (Hodges et al. 2020). Full lockdowns were imposed in Cyprus, Estonia, Germany, and Israel, and no full lockdowns were introduced in Sweden (for more information on governmental policies, please see UNESCO Methodological Note 2021: https://en.unesco.org/sites/default/files/en_methogological_note_-_unesco_map_on_covid-19_caused_school_closures_reopening_final.pdf, accessed on 3 November 2023). In addition, as can be seen from Table 1, the overall durations of the full and partial lockdown in Germany and Israel were higher than in the other countries participating in this study.

In Cyprus, the in-school operation of all public and private schools at all levels was suspended on 14 March 2020. Within a few days after school closure, the decision regarding distanced learning—both synchronous and asynchronous—was taken by the Government of Cyprus. In Estonia, due to the COVID-19 pandemic, educational institutions switched to emergency remote teaching on 14 March 2020. On May 2020, Estonian schools partially “re-opened”, but online learning continued. Israel imposed several strict lockdowns with a full closure of the in-person schooling while offering online classes. In Israel, schools were completely closed or partially closed for 33 weeks. In Germany, the first lockdown was imposed from March to May 2020 when all classes were held completely online. In November 2020, the light lockdown measures were introduced, during which the schools were re-opened, and students were attending classes in masks. From mid-December 2020 until the end of June 2021, another strict lockdown—the second lockdown—followed. During this time, schools and kindergartens were closed again. However, the duration of the lockdown depended on the epidemiological situation in the respective state, city, and even school. As far as Sweden was concerned, it was the only country that did not shut down all schools. Some restrictions in some schools were imposed. Sweden relied on its citizens to act according to the Government’s recommendations and have solidarity with each other (https://www.folkhalsomyndigheten.se, accessed on 3 November 2023).

5. Methodology

5.1. Participants

Using a qualitative approach, we aimed to address the research questions of the study. Therefore, we conducted in-depth semi-structured individual interviews with 10 Russian-speaking parents in each country—in total, 50 participants—who have children below the age of 18. To ensure the meaningful comparative analysis across the five countries, the researchers invited participants who reported Russian as L1, which was the inclusion criterion for this study. The interviews were conducted in the Russian language. All the researchers used the open-ended questions according to a number of preconceived themes which are related to family language practices, beliefs, and management strategies before and during the COVID-19 pandemic. The interviews allowed us to obtain rich data on an individual’s experiences and beliefs.

All the participants were first-generation immigrants and were recruited using snowball sampling via social media.

Table 2. Demographic data on the participants.

Country		Cyprus	Estonia	Germany	Israel	Sweden
Age (years)	Mean	43.4	39.2	37.7	42	42.1
	Range	38–51	30–46	32–45	32–54	28–53
Education	School	1	0	0	1	0
	College	0	0	1	0	0
	BA	6	4	9	4	5
	MA	2	6	0	5	5
	PhD	1	0	0	0	0
LoR (years)	Mean	13	39	18	22	20.6
	Range	5–20	30–46	5–28	10–31	7–28
Number of children	Mean	1.8	2	2	2	2
	Range	1–3	1–3	1–3	1–5	1–3
Immigrated/Repatriated from	Russia	3	0	6	4	7
	Ukraine	2	0	0	4	3
	Belarus	2	0	1	0	0
	Moldova	2	0	0	0	0
	Latvia	1	0	0	0	0
	Kazakhstan	0	0	2	0	0
	Kyrgyzstan	0	0	1	0	0
	Estonia	0	10 *	0	1	0
	Abkhazia	0	0	0	1	0
Type of family	Endogamous	5	9	8	8	3
	Exogamous	5	1	2	2	7
Structure of the family	Co-living two-parent family	8	9	7	8	4
	Divorced two-parent family (50%–50% parent care)	0	0	2	1	2
	One-parent family	2	1	1	1	4
Home language use before COVID-10	Mainly RU	1	6	5	3	4
	Mainly SL	3	0	1	1	2
	Both	6	4	4	6	4

LoR = length of residence in the host country (years); * born in Estonia.

provides demographic information on the participants of the study which was collected during the interviews. All the participants were females of mid-to-high socioeconomic status, with BA- or MA-level education. Since there were only ten participants from each country, we decided to recruit participants of the same gender to eliminate gender differences. Recent research shows that the closure of childcare facilities and a shift to virtual learning during COVID-19 placed more burdens on mothers and reduced their working hours (Petts et al. 2021). Therefore, mothers were more willing and available to be included in the current study. Furthermore, female participation can be explained by the composition of the mixed-marriage families, with the local father and an immigrant Russian-speaking mother. The current study represents the mother’s perspectives on the family language policy implementation, which has been proven to be valuable in numerous previous studies (e.g., Schwartz 2010; Said 2021; Selleck 2023). The participants’ ages ranged between 28 and 54. On average, the families had two children in the age range from 2 to 18.

5.2. Materials and Procedure

Interviews are considered to be valid and reliable tools for data collection; especially if taking affective and social aspects into consideration, it is possible to create a safe and comfortable environment for the participants (Prior 2016, 2017; Rolland et al. 2020). The interviews were designed to provide a detailed snapshot of experiences of multilingual families: access to and support of remote learning, care of children in multilingual families, communication between close and distant family members, and child and parental involvement practices regarding HL and SL learning (see Appendix A). The interviews also focused on future steps (post-COVID-19) that are pertinent to supporting and maintaining harmonious bilingualism, defined as “a subjectively neutral or positive experience that members of a family in a bilingual setting have with aspects of that setting” (De Houwer 2020, p. 63). In the current study, all mothers were given the opportunity to express their experiences and subjective perspectives during the interviews. They reported socioeconomic challenges, such as space constraints or limited time, language barriers, and lack of access to technology and a non-supportive online learning environment, as well as insufficient parental support and engagement.

In-depth semi-structured interviews were conducted via video meeting platforms (e.g., Zoom, Skype, and Teams) in March–April 2022. The procedure of the study was developed in November–December 2021. The online video meeting platform has been widely used since the COVID-19 pandemic, and it is becoming a useful replacement for face-to-face interviews (see Mirick and Wladkowski 2019). While conducting qualitative interviews as part of the current study, good research practices, which are based on fundamental principles of research integrity, were fully applied with respect to the research environment, research procedures, safeguards, data practices and management, and collaborative working, as per ALLEA (2023). The informed written consent form was obtained from each participant. The interviews were audio-recorded and transcribed. Overall, the interviews were conducted in the Russian language by multilingual interviewers, though occasional code-switching was observed in both the interviewers and interviewees. In this article, we give a broad orthographical transcription. The data were analyzed in consensus, using the grounded theory research method (Bryant and Charmaz 2019; Creswell and Poth 2018; Rolland et al. 2020). Iterative and recursive content analyses of the data were implemented to reveal the thematic patterns (Ward and Wolf-Wendel 2004). The transcripts were attentively reviewed: repeating keywords and phrases were identified; and the themes were created—(1) the effect of the pandemic on HL and SL acquisition (HL increase/SL decrease, HL decrease/SL increase, and no change in HL and SL use) and (2) the nature of child and parental agency in the facilitation of the possible changes.

6. Results

6.1. The Effect of the Pandemic on HL and SL Acquisition

With regard to our first research question of whether and how the pandemic conditions affected the HL acquisition, our study showed that, before the pandemic, multilingual families in the current sample varied greatly in the amount of HL-Russian exposure at home. Some families maintained predominantly HL-Russian home environments (e.g., “Overall, in 90% of cases, the communication is conducted in Russian”). Others showed strong SL-home environments, with very little presence of HL-Russian (e.g., “Russian, maybe, I do not know, takes up 5% of the total time”). Similarly, during the pandemic, the FLP in the families varied greatly.

The pandemic has also crystallized the concept of FLP for many families. While many of the families stated that they did not have any specific FLP before the pandemic and had more of a laissez-faire attitude towards the languages, they became more aware of how they were managing languages at home and understood that their actions regarding the language use (especially the HL) led to specific outcomes.

6.1.1. HL Increase/SL Decrease

In all five countries under investigation, especially in endogamous families, parents reported that the pandemic provided a great opportunity to bring HL-Russian back to the home environment as parents/caregivers (e.g., nannies and grandparents) were spending more time at home with children, while schooling in the SLs decreased. That was noted especially in Israel and Germany because, in these countries, the overall duration of the full and partial lockdown was higher than in other countries under investigation (see Table 1). Some mothers reported that the pandemic triggered a greater opportunity to make their own HL-Russian more active. Furthermore, children have been reported to watch more cartoons and movies in HL-Russian. In families where Russian was encouraged and served as the dominant language for intrafamily communication, with parents actively investing in transmitting the heritage language to their children, the period of isolation during the pandemic had a positive impact. Thus, the COVID-19 pandemic boosted the families’ pro-HL-Russian FLP. The family members were spending time together at home to ensure that they all spoke in more HL-Russian than SL. The contact with peers in the SL due to school closures decreased (see Responses (1) and (2)).

(1) My husband watched movies with her [daughter] and they discussed them a lot afterwards. I noticed that she got very interested in comedies… yes, in Russian… And I tried to watch Estonian channels with her, but she did not show much interest in them…

(Estonia)

(2) We only speak Russian at home. During the lockdown, hardly anything changed. Maybe, we even spoke more Russian because we were all at home, our children had little contact with their friends, and we watched a lot of Russian films and cartoons at home.

(Germany)

The COVID-19 pandemic made the families in the respective countries aware of the fact that the Russian language can be present in their families both in the online and offline modes. The pandemic also revealed the hidden resources of the family and the close family networks.

In Germany and Israel, in families where HL-Russian was not the dominant language for intra-family communication, the mothers saw the pandemic as an opportunity to reintroduce HL-Russian and foster a more favorable attitude towards it, as seen in Response (3).

(3) The children always speak Hebrew among themselves. They use Russian only for me. They understand my reprimands and cursings. They always answer in Hebrew, very rarely in Russian, with some insertions of Russian words. Their Russian has improved in microscopic doses. A little bit, probably, just because they heard me more during the pandemic.

(Israel)

6.1.2. HL Decrease/SL Increase

Alternatively, some families—especially in Cyprus, Estonia, and Sweden—in comparison to other countries, reported an opposite trend, pointing at the decrease of HL-Russian during the pandemic and the SLs taking over due to the closure of afterschool activities in HL-Russian, thus leading to a more SL-oriented HL environment (see Response (4)). This might be explained by the overall duration of the full and partial lockdown in these countries. Please remember that schools were not closed in Sweden, while they were partially and/or fully closed in Germany and Israel (see Table 1).

(4) Swedish started to clearly dominate during COVID. He [the child] is either watching more movies in Swedish or the language [Swedish] is easier? [What the mother meant was: ‘He is watching movies in Swedish because it is easier for him.’]

(Sweden)

In some countries, the mothers were more concerned about the lack (or slow development) of the SL or frequent language mixing. Moreover, some mothers reported that children transferred their home linguistic practices of language mixing to a virtual classroom (see Response (5)).

(5) During lockdown, we were all at home. So, the mixed language use was the same—a mix of Russian and Estonian—and became a problem as children started to mix language at school.

(Estonia)

The exposure to the SL was limited mainly to and associated with the completion of the children’s school homework. In some families, new language practices were introduced, in particular, the use of the SL during (joint) homework activities. The mothers reported that they separated the usage of the two languages, as in Response (6).

(6) My husband helped him with math, as I am not very good in mathematics. I helped him with German, English, and Geography. It was very difficult. I had to explain everything to him in German only.

(Germany)

One-language–one-context and/or one-language–one-parent policies were replaced by the one-language–one-activity policy. The parents introduced the rule that they spoke the SL with the children when they did their homework together for mainstream school, since this was virtually the time for schooling. As soon as this was over, everyone switched back to Russian. However, the mother noted that they had little or no time for special activities in Russian, like reading books, as children needed more rest after SL schooling. In Estonia, there was one family for whom the lack of the SL at home during lockdown led to the decision to seek an Estonian-medium kindergarten to increase the SL exposure.

6.1.3. No Change in HL and SL Use

Some families reported no change in children’s HL acquisition and literacy development. The families that were initially HL-dominant maintained HL-dominant environments, as in Response (7), whereas the families that were SL-dominant preserved the pre-COVID SL-oriented FLP, as in Response (8):

(7) Nothing changed during pandemic. We spoke Russian and we continue speaking Russian.

(Sweden)

(8) The older [child] spoke Hebrew and continued speaking Hebrew, so everyone stayed in their comfort zone.

(Israel)

One of the key factors in Cyprus is the family composition, endogamous or exogamous, and whether both parents are HL-Russian speakers or members of a mixed-marriage family (usually with a mother speaking HL-Russian and father speaking SL-Cypriot Greek). As a result, the type of the family seemed to affect FLP in Cyprus (see also Karpava 2021). Endogamous families tend to use only HL-Russian at home (see Response (9)):

(9) [Did the FLP change during the pandemic] Of course not. We were speaking our native language, Russian before the pandemic. So, we just continued to do it.

(Cyprus)

Some families reported no language shift, focusing instead on COVID-related issues, such as well-being, mental health, and tiredness, as in Response (10):

(10) We had no problems except for one [problem]: we had fear!

(Sweden)

6.2. The Nature of Child and Parental Agency in Facilitation of the Possible Changes

Regarding our second research question, which deals with children’s active role, research on FLP has started to acknowledge children as co-agents in shaping FLP (Uribe de Kellett 2002; Fogle and King 2013; Slavkov 2015). According to Smith-Christmas (2020, p. 221), child agency is of an intersectional, multidimensional, and multilayered nature, which is based on the constant negotiation of language choices and preferences in the family. While multilingual children grow, they experience life in many different contexts outside their home environment, and the older they become, the greater the variety of these contexts (cf. Hickey 2021). Friends, school, and society influence the children’s language practices. However, the parents are and often continue to be the main role models for the children, shaping the preferences for the child’s choices in language use (Ellis and Sims 2022).

Regarding the role of the child’s agency in facilitation of potential changes, our findings indicated that the patterns observed prior to the pandemic became even more pronounced during the health crisis. Child agency has developed, as siblings and peers prefer to seek help from each other rather than from their parents, likely due to being confined to the home environment, as shown in Response (11).

(11) I never interfere until they ask for help. Sometimes the younger one turns to the older one, and the matter does not reach me.

(Cyprus)

The children who displayed an interest in reading prior to the pandemic began devoting even more time to it during the pandemic. The mothers did not mention any new habits that emerged during this period; instead, the existing ones became essential, as children had more time on their own to develop these habits.

Regardless of the country and the country’s lockdown policy, the role of children as active actors in modifying FLP became more apparent than ever before (probably due to the language present in the online mode, and it is known that the children spent much more of their time online) (see Response (12)).

(12) She looks for Russian-speaking videos on YouTube (she can read and write); she watches Russian-speaking YouTube bloggers for children (her own interest); she is eager to watch films in Russian.

(Germany)

During the pandemic, those parents who wished to transmit the HL to their children saw this unique time as a great chance to teach it. It had become the family’s responsibility to give their children confidence in using the HL by giving them opportunities to use it in communication, including reading and writing. Yet, in Estonia, Germany and Sweden, several mothers mentioned the lack of time, even though the pandemic brought the families closer than before. When such situations occurred, the children, as it was reported by their mothers, took the initiative and grasped opportunities to search for the information on their own.

By having to work from home, some parents exposed their children to different registers of the Russian language, which gave the children opportunities to hear the new lexicon and grammatical structures in formal and informal contexts. Furthermore, there was a bidirectional relationship between child linguistic choices and parental language preferences within HL environment, demonstrating how child agency is influenced by the parent’s new behavioral pattern (see Response (13)):

(13) I started watching more Russian programs and she [my daughter] became interested in them too and listened to them and then she even started searching for her own [programs].

(Sweden)

During the pandemic, new literacy and FLP routines were introduced, taking into consideration children’s age and their relevant needs. If the parents were consistent in applying them, the new language use and learning patterns became sustainable, as seen in Response (14).

(14) [About the younger daughter:] We learnt proverbs and read fairy tales in both Russian and Estonian; I gave her more attention than before the pandemic. [About the older daughter:] She knows how to entertain herself: she picks up a book and reads, or just plays by herself. She speaks different languages with her dolls when I listen to her. For example, she plays “In the shop” and her customers (dolls) buy in different languages.

(Estonia)

Due to the pandemic, children were able to develop digital literacy because they had access to different platforms and gadgets. This facilitated children’s agency in terms of their HL use (cf. Lanza 2021; Curdt-Christiansen and Iwaniec 2022). Children opted to communicate more with their grandparents via various digital platforms, which enhanced their HL skills (see Response (15)). This situation enhanced the role of relatives in shaping the FLP and HL environment.

(15) My parents-in-law live in Russia. They don’t speak Hebrew, they speak Russian to the kids. My mother-in-law… she works for a logopedic kindergarten. She is very concerned about the Russian language skills of the kids. The kids phone her via WhatsApp and the grandma teaches them all sorts of tongue twisters, she teaches them to pronounce “r” correctly, they have a Hebrew “r”, not a Russian one. Grandma teaches them, and she is very happy, as they teach her letters in Hebrew.

(Israel)

Some mothers reported that their children showed no interest in developing the HL. It became a particular challenge in one-parent families or in families where a father lives and works abroad. In such cases, the grandparents’ (digital) agency, i.e., active involvement of grandparents in children’s everyday language usage, was extremely helpful, as they took an initiative in helping with HL transmission (see Response (16)):

(16) Their grandmother took care of them while I cooked and cleaned for the whole bunch. The youngest ones read with her [in Russian], while the oldest only sat with a phone in his hand.

(Estonia)

Many mothers observed that their children’s HL proficiency increased due to online Russian classes and enhanced digital literacy and child agency (see Response (17)):

(17) My children have improved their Russian. The teacher just praised my children. And I can say that my children are independent, and I do not interfere in their educational process. They study on their own, do their homework and do all the tasks on their own. And I haven’t interfered in it for a long time. No, no one helped. My children have been learning on their own for a long time without my help. All children are different, my children are very responsible, so it’s not difficult for them to study online. They realized that online lessons are very convenient, you can have lunch and immediately go to the computer and study with the teacher.

(Cyprus)

7. Discussion and Conclusions

The current study was designed to examine language and literacy practices among multilingual families in five different countries (Cyprus, Estonia, Germany, Israel, and Sweden) during the COVID-19 pandemic. It also aimed to investigate whether the pandemic influenced language and literacy practices within HL-Russian-speaking families (10 families per country) through in-depth semi-structured interviews.

Our objective was to explore whether and how the pandemic conditions impacted the HL and the SL acquisition, as well as literacy learning environments, and to understand the role of both child and parental agency in the facilitation of potential changes. While research on the effects of the pandemic has been conducted in various domains, there is still limited research on its impact on language use and policy within multilingual communities (see Murrmann 2021; Sheng et al. 2021; Sun et al. 2023).

The child’s home environment and parental engagement (e.g., children’s schooling, Slavkov 2017) are important factors in shaping HL and SL development, as shown in previous research (Di Pietro et al. 2020; Schleicher 2020; Yilmaz 2021). Our study provided further evidence in this respect. Interestingly, at the beginning of the interview, the mothers were not quite sure whether the pandemic has affected their FLP, or educational and linguistic practices, while then in the process of the interview, different types of questions elicited rich data regarding the research questions of the current study. The analysis of the data suggested that there were certain similarities and differences across the five countries in which the study was conducted regarding the response of the immigrant/minority families to the COVID-19 pandemic and school closure. The duration of schools’ closure and number of phases varied across the five countries, which affected the educational process in different ways and, as a result, the amount of time that the families had to spend at home in isolation. In particular, the overall duration of the full and partial lockdown in Israel and Germany was higher than in Estonia and Cyprus, as well as in Sweden, even though this country had only partial lockdown. This situation created more opportunities for HL-Russian development and transmission in Israel and Germany than in the other countries, where the families were “forced” towards the pro-SL FLP. Home-based teaching/learning in digital settings was the most prevalent form of education in most of the countries under investigation from March 2020 until May 2022. One of the aims of our study was to provide an overview of family reflections regarding their language management, use, and educational experiences. It further explored the effect of the pandemic on HL and SL use, maintenance, and transmission; the literacy environment; and the development of literacy skills, collecting the perspectives of all members (mothers, fathers, children, and extended family members) in explicit and implicit ways.

With respect to our first research question, which aimed to investigate whether and how the pandemic affected HL and SL acquisition and literacy learning environments, the results indicate that the changes in FLP are noticeable, but they seem to depend on the child’s agency rather than only on parental intentions and assumptions. In line with previous studies (Murrmann 2021; Sheng et al. 2021), the families reported that disruptions from the normal schooling routines had both positive and negative consequences for the use/development of the HL and the SL. Young children required more support from their parents, older siblings, and extended families. It should be noted that the type of the family, endogamous (mixed-marriage) vs. exogamous (co-ethnic, where both partners have the same L1 background; see Karpava et al. 2021 for a further discussion), as well as one-parent vs. two-parent family, is an important factor for each country setting. Individual differences play their roles as well. In Cyprus, families with both parents speaking HL-Russian at home mainly continued to use only HL during the pandemic as they did in the pre-COVID period. At the same time, in the mixed-marriage families, the situation was much more complex. Both Greek and Russian were used, and the rate of use was affected by various factors, such as FLP, language status in the society, motivations for integration into the mainstream society, communication and access to social networks, and availability of time and access to technology. In both Estonia and Germany, Russian-speaking families were more concerned about SL development than HL maintenance due to the lack of SL use during the lockdown time. This trend was echoed in Israel and in Sweden, especially with respect to younger children and in families that practiced mainly the HL-Russian only FLP.

Based on our findings, FLP during the pandemic was shaped by access to education in the HL and the SL. Educational inequalities might have been exacerbated due to home-based learning. Socioeconomic status seems to be a key factor in this respect since some of the families with a low income or who are unemployed were not able to offer the same educational opportunities for their children in comparison with some other well-off families. This finding is in line with findings of Sun et al. (2023), who also reported differences between families with high and low socioeconomic statuses. However, in the Swedish data, the socioeconomic differences seemed to be less pronounced (however, it needs to be investigated with statistical data). In addition, it seemed that not all families had enough space at home or were well equipped with information and communication technology to have a comfortable environment for learning and support of HL and SL literacy skills.

In Cyprus, children had more time for their school lessons and extracurricular activities, such as Russian classes, as their classes were online, thus decreasing the time that was previously wasted on commuting from home to school and tutor centers. Under normal circumstances, parents in Cyprus have to drive their children to private classes because there is no well-developed public transportation, which is not the case in Estonia, Germany, and Sweden. In all of the countries, extracurricular activities (either in the HL or in the SL) were largely cancelled during the pandemic or moved to the online format. However, parents noted that the newly developed mode was appropriate. At the same time, it should be noted that not all families could afford extracurricular online Russian classes (with the exception of Sweden and Germany, where the Russian classes are subsidized by the government), and the parents were either involved in joint HL literacy activities with their young children or the teenagers had to cope by themselves and self-study, often without proper emotional and pedagogical support, a finding that is in line with the previous research (Di Pietro et al. 2020; Uro et al. 2020). Not all children and parents were ready for emergency remote teaching (Hodges et al. 2020), which affected the mainstream school and the HL education and literacy development (OECD 2020). Therefore, the children were deprived of stimulating activities and had more time to spend on their own, which, in turn, leads to child agency and child-initiated choices of activity.

Turning to our second research question regarding child agency, the pandemic conditions gave the children new opportunities for agency when it comes to (1) language and literacy development choice, (2) communication with extended family members, and (3) widening their social network and finding their own new sources of input. One of the main outcomes was that the children became more independent and had a chance to develop their digital literacy. In all the countries, the children’s agency has been boosted, as they have become more flexible and equipped with technical knowledge. The children sought alternative activities to fill the void, which was created due to the cancellations of schooling and extra-class activities by using different platforms, software applications, and multimedia. The children found new sources of input in the HL and the SL by implementing 21st-century skills. However, it should be noted that child agency has been affected by various factors, such as the FLP; home language environment; relationship with extended family, relatives, and friends; schooling; and social network.

Furthermore, intergenerational communication was significantly enhanced both quantitatively and qualitatively during the pandemic, reaching new (digital) heights. The children played a key role as initiators of this transformative change in FLP. Further research on larger populations, different countries and various language combinations are necessary to obtain more valid and reliable insights into the development of HL and SL skills within immigrant communities of multilingual Europe.

The role of digital technologies and new learning opportunities are emphasized. It should be noted that the changes in use of the SL and the HL during the pandemic were related to such factors as the child’s age (Response 14), child agency (Responses 11, 12, 15, and 17), the role of relatives/extended family (Responses 15 and 16), (online) HL environment (Responses 4, 7, 8, 12–15, and 17), SL development (Responses 5–6), family composition (Responses 9 and 10), FLP (Responses 1–3, 9, and 14), and online schooling (Response 17). In many instances, both parents and children were working and studying online at the same time. Most of the children at home had to develop their own independence in digital literacy. Enhanced digital literacy had a positive impact on their HL development, as they had more access to various digital platforms and programs in multiple languages, thus increasing the quantity and quality of HL. The children seem to become especially sensitive to the language use preferences of their parents (cf. Said and Zhu 2019) due to the increased time spent together at home, as it has been shown in the relevant examples in the Results section (Response 13).

8. Limitations and Future Research

While the study offered valuable insights into the effects of the COVID-19 pandemic on the development and maintenance of HL and SL in five countries, it is not without limitations.

Starting with the methodological points, further mixed-method research on this topic is important, given the small samples in each country in the present study. Online questionnaires provide data from larger samples, thus enabling solid generalizations. Yet, we also recommend supplementing questionnaire data with interviews employed in the current study, as interviews enable researchers to obtain richer data. In addition, it is beneficial to include both mothers and fathers in the data collection process to gain perspectives from both parents on the effects of the COVID-19 pandemic on FLP. It will be interesting to obtain the child’s perspective on the influence of the COVID-19 pandemic by conducting interviews with the children themselves. Finally, in order to generalize about the influence of COVID on HLs, it is important to extend the current research to a larger number of countries and different continents and any other H, for example, HL-Turkish, HL-Arabic, and HL-Spanish.