A generational arc: early literacy practices among Pakistani and Indian heritage language families

ABSTRACT This paper focuses on the attitudes towards reading in the home, handed down through the generations and experienced by the young children in four families of Pakistani and Indian origin. The children’s families originally arrived in the UK in the 1960s, and this paper unpicks the stories and attitudinal changes in relation to both English and the heritage language, throughout the generations. Adopting a sociocultural perspective through intergenerational family interviews, roles within the family in terms of literacy support, the families’ use of libraries, experiences, and understanding of the education system, and the impact the heritage language has on family support for reading in English, are explored. Through the dual linguistic lens of both English and the heritage language (Gujarati and Urdu), the study traces a generational arc which explores areas of concern and needs for support, seeking to inform both policy and practice in early childhood education.


Introduction and background
The term 'family literacy' describes how parental and other family members' use of written language shapes children's early literacy experiences (Taylor 1983). Several studies have furthered the understanding of existing family literacy practices, including those across a variety of socio-economic backgrounds, and within heritage language communities (see e.g. Gregory 1994;Hannon 1995;Baker, Scher, and Mackler 1997;Wasik and Van Horn 2012;Song 2016). Wasik and Van Horn (2012) explain: The intergenerational transfer of literacy has intrigued educators, researchers and policy makers, and served as a fundamental rationale for family literacy programmes. Children who come into the world without language learn one of thousands of languages, depending upon the family into which they are born.
(3) a drive by educators to create links between home and school, encouraging parents to become more involved in their children's education. The notion of 'family literacy' therefore not only refers to government-driven interventions, but also encompasses the literacy practices that occur within the home (Nutbrown et al. 2016). This original study focuses on family involvement in and encouragement of children's early literacy experiences. Intergenerational family interviews were conducted with four families whose children attended the same multi-ethnic nursery in an urban town in Northern England. These interviews drew out narratives highlighting common attitudes and family literacy practices, as well as differentiated opinions and attitudes towards family literacy and language development in both the heritage languages and English. Links to early childhood education are represented not only among the youngest generation, but also through experiences shared by older generations, especially the grandmothers, most of whom arrived in the UK at an early age. This article traces family literacy development across these four families, concluding that family literacy development in heritage language families can be represented by a generational arc, where the focus shifts, from an integration into English language and culture in the oldest generation, to one of cultural and linguistic heritage, in the youngest generation, resulting in a dual family pathway towards reading and story sharing in both English and the heritage language, but with distinctly different practices regarding oracy and literacy, often at the expense of heritage language literacy development.

A word on terminology
The term 'family literacy' necessitates problematisation, in order to understand its meaning and implementation among families who may have different or multiple heritages (Song 2016). The terminology involving such families is complex, with research spanning aspects of immigration (Rumbaut 2004), bilingualism, multilingualism, and plurilingualism (Baker 2011), and home language (Kenner 2000), to name but a few. Throughout this paper, the term 'heritage language' is used. Blackledge and Creese (2010) critically explore the term 'heritage' in 'heritage language', arguing that it implies cultural and generational connotations beyond language alone, which ties in well with this study's notion of a 'generational arc'.

Family literacy in heritage language families
Responsibility for family reading, story sharing, and family literacy has rested largely with mothers (see e.g. Reay 2002;Hartas 2011). Hartas (2011, using data from the Millennium Cohort Study, identifies that the mother's level of education is more relevant to fiveyear-old children's literacy development than family income. This raises issues regarding perpetuation of social inequality, home-school relationships, and acknowledgement of heritage language contribution to literacy development. Reay (2002) identifies white middle-class mothers as much more likely to have the social capital (Bourdieu 1986) to involve themselves in their children's education. Interaction with school and teachers is part of their 'habitus' (Bourdieu 1986), whereas working-class mothers and those from ethnic minorities are often less confidently engaged in their children's schooling. Both Edwards (2013) and Hartas (2011) advocate the education of mothers to overcome these issues. Hartas (2011, 909), however, warns that parental involvement should not be considered as 'panacea for making up for the effects of socio-economic inequality', instead, it is just one aspect of a coherent policy to address issues of social justice.
As Gregory (1994) states, 'attitudes to reading vary widely' across ethnic and cultural groups (113), it is therefore important to work with local communities to explore attitudes, and to enable the families' voices to be heard. Blackledge (2001) describes Bangladeshi mothers in Birmingham expressing frustration at expectations made by schools to read with their children in English. Gregory (1998) identifies four assumptions that have traditionally been involved in research surrounding family literacy practices. These assumptions include the belief that the Western model of home literacy is in some ways better or more correct than any models practised in families from other cultural backgrounds, that it is suitable for all children, regardless of their background; and that the correct way to read with a child is transmitted to parents via the school's model. Gregory also raised the point that home literacy usually focuses on the immediate parent-child interaction, whereas in fact siblings can have a key role in culture brokering.
Kenner provides evidence (Kenner 2000;Kenner et al. 2004) that opportunities to develop into plurilingual individuals can enhance learning and academic success. However, it is also clear that the permutations of socio-economic and linguistic background require nuanced approaches towards inclusion in 'superdiverse' (Vertovec 2007) contexts. Given the lack of linguistic data in England, where language information was only added to the National Census in 2011, it is difficult to develop a more comprehensive picture of the relationships between the early literacies of those from monolingual Englishspeaking backgrounds and those from different language groups, a gap this study seeks to address. Robertson, Drury and Cable (2014) highlight tensions in the Early Years Foundation Strategy, which continues to view multiple languages in a child's life as contradictory, opposing forces. Tinsley and Board (2016) stress that, in the current UK education system, much support towards heritage language speakers is in the form of encouraging a generic openness towards other cultures, rather than actual support in the heritage language.
In the US, Golash-Boza (2005) shows that children from ethnic backgrounds who do not learn English to a high standard fall behind their peers who do. Zhou and Bankston (1994) report that those children who manage to integrate well into both the host country's language and the heritage cultures have the strongest social capital, and are most likely to achieve academically. Conducting research into three-generational Gujaratiand Urdu-speaking families in north-east London, Sneddon (2000) showed that support for oral language development in the heritage language mainly occurred at home, whereas written literacy was largely developed, in English, through schools. By the age of 11, however, children were performing above the average of monolingual English-speaking children of similar background, whilst also speaking fluently Gujarati and developing literacy in Urdu for religious purposes. Kenner (2000) highlights that home languages often lack the status which is afforded to English, and the curriculum does not present children from other heritage backgrounds with sufficient opportunities to draw on their full linguistic repertoire. According to Portes and Hao (2002), the heritage language has all but disappeared by the third generation. While England has moved beyond the era where heritage languages are banned in schools, towards a time where pluriculturalism is celebrated (Golash-Boza 2005), educational practices continue to favour white middle-class families who remain more confident in negotiating their children's educational journeys (Reay 2008;Crozier, Reay, and James 2011). The attrition of the home language and the complex impact this has on families is a particular concern. Fishman (2001) problematises the 'language shift'the attrition of home or community languages through assimilationand the attempts to reverse it. Robertson, Drury and Cable (2014) argue that, even in education settings which include bilingual practitioners (such as the nursery in this study), the current dominant discourse works against bilingual education.

The role of libraries to immigrant families
The provision of library services for Indian and Pakistani immigrants at the time of the grandmothers' arrivalthat is, the 1960shas been documented by Lambert's (1969) survey of 50 public library authorities in 1967. At the time, although 33 libraries made some provision for speakers of community languages, others commented that providing such books was detrimental to 'the interest of encouraging integration rather than segregation' (42). This view was shared with at least some of the new arrivals at the time (Vaughan College Study Group 1967). Lambert (1969) reports that demands for heritage language books had been low in libraries not providing the service, but that take-up was good in libraries that went out of their way to establish links with local communities of new arrivals. There is an obvious difference between asking for a service not currently provided, and simply using that service once it is available. Retrospectively, it may be argued that new arrivals not speaking English lacked the confidence to make demands of public libraries, to gain access to books in their own language.

Research setting and approach
This paper adopts a sociocultural perspective in its endeavour to explore experiences and changes in attitude towards family literacy over time. Wertsch (1991) warns that sociocultural research may lack a focus on history. In this study, history is related as experienced by individuals, against the backdrop of immigration of the England of the 1960s. While it goes beyond the scope of this article to provide a comprehensive political and social background, the changes in family practices through the generations form an important aspect of the work. With this in mind, a research method that allowed families to share their stories across generations was the most appropriate research method. Tsikata and Darkwah (2014) explore the methodological challenges of intergenerational research in a multilingual environment, arguing that intergenerational interviews allow family narratives to take shape, complementing qualitative research orientations. This research focuses on four multilingual families, drawn from a single multi-ethnic and plurilingual community in an urban Yorkshire setting. Families were recruited via a multilingual nursery, which the youngest children in each family attended, and all families volunteered for the research. The main languages spoken among the families attending the nursery are Gujarati, Urdu, Punjabi, and Arabic, as well as English. For each of the families interviewed, immigration to the UK from India or Pakistan was part of their family history. For three of the families, the grandmother was the original arrival in the UK. In one family, the grandmother was born in the UK, and in another family, an additional female family member (the mother) arrived from India to marry into the family.
Rather than seeking stories that are 'representative', this research instead lends importance to individual experiences, and although many of the stories were similar across the families, the research argues for awareness of families as unique, with their own stories and heritage, rather than as homogenous, replicable groups. The stories the women told of early childhood literacy experiences, against the backdrop of the 1960s, 1980s, and the 2010s, merge the fields of critical pedagogy and storytelling, 'remind[ing] us that we cannot depend on statistical data to illuminate experience and compel change' (Carmona and Luschen 2014, 2).

Participants
In order to get a better picture of the families themselves, what follows is a brief overview of each. All families have been given pseudonyms to protect their identity.
Desai Family: Grandmother (Tanvi), mother (Yara) and grandchild, a four-year-old girl (Payal), were present at interview. Tanvi arrived from India with her parents at the age of seven, and remains more comfortable speaking Gujarati. Yara is the third of four daughters, bilingual in English and Gujarati. Yara has two daughters aged four and seven, and both girls prefer English to Gujarati.
Patel Family: The grandmother (Isha) and her daughter-in-law (Neha) were interviewed together. Isha arrived from India in 1965, aged five. Her parents could not read or write in English and had limited literacy skills in Gujarati. Her daughter-in-law, Neha, was brought up in India and came to the UK to marry. Neha speaks only English with her son Hakesh (three-and-a-half years old).
Abid Family: Fahmida (grandmother who arrived from Pakistan age 6) and Gulnaz (one of Fahmida's four daughters) attended the interview. Reference was made to Gulnaz' three children (five, four and three years old). English is the main language in Gulnaz' household, which includes the three generations of her family.
Mistry Family: Grandmother (Hemal), mother (Amisha) and daughter (aged nearly three) were present at the interview. Hemal was born in the UK, to a British mother and Gujarati father. English is her first language, she picked up Gujarati later through her friends. Hemal has five children and five grandchildren.

Data collection and analysis
Rather than following a strict interview schedule, the research instead focused on asking questions that would elicit stories and memories from the various family members, to be picked up and expanded on by others. Interviews lasted approximately one hour, were audio-recorded, and fully transcribed before coding.
Either two or three generations were present at each interview, and all interviews made reference to four generationsgrandmother, mother, and children, as well as the parents of the grandmother. Adult men were rarely mentioned, and no men were present at any of the interviews (although some of the young children were male).
To counteract Wertsch's (1991) argument that sociocultural research may neglect historical developments, four categories which lent themselves to historical exploration were identified to support suitable questions for interviews, and also formed the basis for initial coding. Themes reported here revolved around (1) early literacy support within the home, (2) navigation of the educational system, and (3) the role of the library. The fourth theme, around technology developments, will not be reported in this paper, due to word constraints. Within each family, a generational, temporal, and narrative thread was established which sought to identify the family context and development over time, and in relation to each theme. All themes were explored from two specific angles, namely English and the family languages, to facilitate coding that took differences and developments into account. Intergenerational interviews allowed for the same narrative to be highlighted from several perspectives (Tsikata and Darkwah 2014), adding to the richness of the data, and weaving a tapestry from the various narrative threads provided by family members.
As Riessman (2008) points out, 'narratives don't speak for themselves' (3), and thus require careful analysis. Asking grandmothers to remember back to their childhood carries potential bias based on selective memory and continuous re-telling, as well as researcher bias in the selection of data for inclusion in the study (McAdams 1993). Interviewing the families separately allowed for an establishment of common threads, and a coding structure of pre-existing categories, expanded by emerging themes within these (such as interpretation of library rules, as well as actual use of the library) allowed for a nevertheless rigorous approach to analysis.

Ethical considerations
Numerous considerations related to the research necessitated the construction of viable and respectful ethical measurements. Damianakis and Woodford (2012) outline numerous ethical issues related to qualitative research in small, distinct communities, ranging from issues surrounding outside researchers entering the community space to confidentiality in such a research environment. Although I am myself part of a heritage language family, my cultural context (Western European) is very different to that of the women and children who participated in the research. The nursery leader therefore functioned as a bridge to the community, and, in collaboration, we answered any questions that arose during the recruitment process. From a cultural, ethical perspective, Dumont (1986) argues that the female Islamic dress code of 'concealment' also implies an act of 'revealment' (277), and Alvi (2013) stresses inherent cultural-, sociological-, and identity-related complexities in relation to female Muslims, which create not only a visual 'otherness' but also demonstrate multiple female identities. For the purpose of this research, being female and a mother of a young child myself was important. Due to being in a fully female environment, the mothers and grandmothers would lower their face veils (niqab) and/or remove their hijab, indicating verbally that they felt at ease to do so. While the children were very young and only participated peripherally, creating an environment that was flexible and enjoyable for them was important, and the interviews were regularly interrupted to accommodate a child's behaviour or question. One grandmother preferred to communicate in her mother tongue, with the daughter translating for her, all other adults spoke very good Englishnevertheless, the information sheet kept to a simple level of English, and understanding was checked in multiple languages before informed consent was sought.
A note on generations Rumbaut (2004) warns against simply categorising immigrants as 'first' or 'second' generation, and identifies issues related to research which focuses on intergenerational relationships with the host country's language and culture. Rumbaut's typology (2004) classifies interim generations, according to the age at arrival in the host country. Those who are born in the host country would therefore be second generation, those arriving between birth and age 5 are considered to be the 1.75 generation, those arriving between the ages of 6 and 12 are the 1.5 generation, and those aged 13-17 are the 1.25 generation. This classification is directly related to the amount of schooling children will have received at various ages. Any attempt at classification is further complicated by additional family members being brought intherefore, while Isha Patel (grandmother) is generation 1.75, Neha Patel (mother) is, generation 1, leading to differentiated cultural experiences and attitudes that are worthy of consideration. Table 1 gives an overview of interview participants.
In all interviews, the stories and narratives told included references to the greatgrandparents' generation, who in three of the four families, were the original 'first' generation.

Findings
The following section addresses the three predetermined themesthe role of family members and the home environment, navigating and engaging with educational contexts, and the role of the library throughout the generations. An overarching theme, revolving around differences regarding language and literacy development in English and the heritage language, concludes the findings section. This reliance on school and teachers can disrupt communication, with parents who spoke little to no English in a position where they had to trust that schools would adequately prepare their children for life in the UK, while they themselves were still acclimatising to new lives, finding jobs, and communicating with local councils. Parents were not unsupportive, as Fahmida Abid (grandmother) explained, they would ask about school and be keen for their children to achieve. However, the language gap made it difficult for parents to know how to help their children best. Lambert's study (1969) may offer another explanation for the lack of shared book readingin any languagewhile books for children in the UK and other Western countries have been well-established for a long time, 'book production for children is only just receiving the attention it deserves in the Indian sub-continent' (46). The contemporary library service local to the families in the study presented here is listed in Lambert's article as having 200 books in Urdu in 1967, but none in Hindi, Punjabi, Bengali or Gujarati, servicing a community of 1800 Pakistani and 600 Indian new arrivals (56). As Hannon and Cuckle (1984) point out, many schools at the time did not subscribe to the idea of sending books home with children, although Tanvi Desai (grandmother) remembers books being sent home in the 1960s.
Although Gregory (1998) highlights the role of siblings in brokering culture and engaging younger children in literacy practices, not all families had such experiences. Yara Desai (mother), who had two older siblings and one younger one, commented: No [I don't remember seeing them read], because […], you know I was at home they were at school so … and then they used to go to obviously the Mosque up there you know […], so at that time I don't think I ever remember seeing them read.
All four families stated that their children attended the mosque daily, and that it formed an important aspect of the family's life within the community.
Isha Patel (grandmother) voiced another discrepancy in support that changed across the generations: But nowadays the grandchildren are lucky that they've got their own parents who can support them and, you know, move them forward, and also grandparents as well, which we didn't have.
She further explained that her own hardships made her determined to help with her grandchildren's education: The early childhood memories of the grandmothers have shaped their attitudes, which they first passed to their children, and now the grandchildren. Unlike the grandmothers' generation, the young children all have somebody who reads with themoften their grandmother and mother both. This doubles or even triples (with two grandmothers and one mother) the amount of support available to the child, compared to the previous generation. The support available in the development of literacy in early childhood is summarised in Figure 1.
In all four families, the person reading with the children is female, which mirrors the findings of previous research (Reay 2002;Hartas 2011). For all but one interviewee, however, reading with children had an instrumental purposeimportance is attached to the 'skill' of reading, and therefore this skill is nurtured. Once achieved, it becomes less important: while Gulnaz (mother) still reads to her three-year-old, she states that 'the other two [aged 4 and 5] can read now', so they read independently. None of the families reported reading with the youngest generation in the heritage language -Urdu and Gujarati, and, as will be further discussed below, the families' heritage languages are disappearing gradually.

Developing confidence in navigating educational systems
The literature review highlighted the need to educate parents, particularly mothers, to provide them with the social capital necessary to navigate the educational system (Hartas 2011;Edwards 2013). Similarly, though, it warns against viewing family literacy practices that do not follow the standardised, Western model, as inappropriate or 'wrong' (Gregory 1994). Through the stories shared, it becomes obvious that parental support has become more strategic and aligned with curriculum policies in the later generations. There is an awareness among the second generation, the grown-up daughters (now mothers themselves) interviewed, about support for literacy development, and how approaches to the teaching of reading have changed over the years: Yes, so Jolly Phonics now and, you know, they do it with the actions as well so it gets into their head really, really quickly. With us it was A, B, C, D, E, F, G and … I mean we still, you know learnt it, we still know how to read, but for them it's more easy. Yara Desai (mother) There is a sense in these four families that today's parents are becoming more confident to help their young children, but also, in an echo of Gregory (1998), that there is a 'correct' way to read with your child: And as parents we can help them as well because at the front of the book they've got suggestions that as parents you can do to help them.
[…] I think that really helps out a lot, because you know at the front you have something like 'read the book with your daughter again and explain this to her', and we never had that in those days, we were just basic reading. Yara Desai (mother) Yara's comments suggest that parents have subscribed to the various prevailing methods and assumptions cultivated via the education system, and that there may be pressure related to 'doing reading the right way', rather than simply reading with your child, echoing Gregory's (1994) concerns. Contrary to Tinsley and Board's (2016) conclusion that teachers struggle to adequately work with multilingual children (which echoes Kenner's (1999) findings that multilingual literacy work remains 'on the margins of classroom literacy activity in Britain' (3)), the nursery where the research was conducted has been involved in a national initiative to promote family literacy in the early years and provides a booklet to families about reading with their children, which includes reassurance that reading at home can occur in any language. The nursery also makes an effort to include bilingual songs in daily activities. Nevertheless, the families interviewed focused their reading efforts almost exclusively on English.

Access to books and the role of the library
The library is accessed regularly (between once a week and once a month) by three of the four families, and, together with books from school and nursery, is a main source of books for the youngest generation: Obviously when you go to the library she'll [daughter, 7 years old] choose about 10 books from the library as well, all at once. And from school I think they can choose up to 6 or 7 books to read at home. So there's a lot of reading. Yara Desai (mother) In all four families, the grandmothers took their children to the library. For all families, the library served multiple purposes, providing not only access to books, but also 'a place to go', as one grandmother explains: Oh yeah, [libraries] were a regular thing, single parent, five kids, you know not much money, so what do you do to entertain kids on a weekend. So it were off to the library on a Saturday morning isn't it, every Saturday we used to go, bring books home. Hemal Mistry (grandmother) Although she was taken to the library every week as a child, and enjoyed her visits, Amisha, Hemal's adult daughter who now has two children of her own, does not frequent the library, fearing that her children will not be quiet enough: I'm scared of her [daughter, 2] making a noise. […] Yeah it's very echo-y, so any slight noise in there is amplified and it does sound a lot louder than it is. And you do feel that imposing … when you go in you used to think 'shhh … .' even I do it. Amisha Mistry (mother) For the other families, the library still forms a regular part of their lives. One family goes approximately monthly, but, as with the Mistry family, there are rules and regulations attached to the notion of the library: she does know what the library is as well, and it's there for reading and, you know, she'll go in all the little corners and get all the books that she wants, and I'm like 'you can't have too many books, have only two' you know, but yeah. Yara Desai (mother) A rule of 'only two books' is conceived by the family, rather than the library. Although Gisolfi (2014) comments on changes that have occurred over the past decade regarding library use, leading (among other aspects) towards more communal use and an increase in technology, it seems that, in at least some families, the traditional perception of the library as a rule-regulated space of quiet and contemplation remains.

Maintaining heritage language and culture
For all families interviewed, the heritage language seems to be reducing in significance and usage. Yara Desai (mother) said about her four-year-old daughter: To be honest she speaks more English. Even though I live with my mum, I mean they speak more Gujarati, but the way she's been brought up it's just English. I mean she'll understand Gujarati but […] she doesn't attempt to speak it at all. […] Nowadays I think most kids do tend to speak in English even though, you know, their first language is Gujarati, so hopefully when she's older maybe she might start speaking Gujarati. Yara Desai (mother) There was an air of regret or perceived parental 'failure' attached to these admissions, when Yara was told that other families were in similar situations, she exclaimed: 'Thank God I'm not the only one who says that'. Neha Patel (mother), however, explained that she deliberately speaks only English, and reads only in English, to her son, who is three-and-a-half years old: Because he has to learn everything in English now, he don't need Gujarati, he can just speak in home Gujarati but not outside after leaving home. So that's why I say just read in English and writing.
Fahmida Abid (grandmother) explains how they spoke Urdu at home, but that English was the language used for reading. She gives an example how one of Aesop's fables became a metaphor for her family's life: My first favourite book would have been Town Mouse and Country Mouse.
[…] And we still tease them all about it, because my mum's from Pakistan and we're from here, and no matter how much […] she is a part of this country now, she still retains all her own country values. So even if we have English food she'll still have Asian food, and we'll say to her Town Mouse and Country Mouse! Although the families may smile about the great-grandmother's attachment to her native country, the time of the family's arrival roughly coincided with the beginnings of the understanding that maintenance of one's own language and culture may be a vital part to successful integration into the host country. Proceedings from a UNESCO conference in 1956 (Borrie 1959) state that continued association with the native culture: … may well encourage a willingness to cross cultural boundaries […]. This apparent paradox is based upon the assumption that unless immigrants can establish some cultural roots quickly after settlement, there will tend to be a recoil away from the new society [… . By contrast, if given a sense of solidarity by association with the area of origin, the immigrant will be psychologically better prepared for adjustment to his [sic] new surroundings. (138) All the above examples illustrate that, in line with Portes and Hao's (2002) findings, the heritage language decreases in the third/fourth generation. Three of the four families are now making a particular effort to maintain their heritage languages, whereas Neha Patel, the one mother who is a new arrival to the UK, believescontrary to the literature (see e.g. Sneddon 2000)that speaking English only at home will help their children succeed in school. Any efforts to maintain the heritage language, however, remain focused on oracy, rather than reading. Thus, the practice of book sharing remains firmly in the English language domain in all four families, despite the booklet provided by the nursery, informing families that home reading can occur in any language. This is exacerbated by the family's notion that there is a 'right' way to go about family reading, with tips in reading books tied to UK reading schemes, book bands, and curriculum targets. This describes the 'language shift' problematised by Fishman (1991), illustrating the loss of the heritage languageand all its advantagesacross the generations in communities, and aligns with Robertson, Drury and Cable's (2014) argument that the presence of bilingual practitioners on its own can do little to help families maintain the heritage language, unless clear policies and practices are in place.

Conclusions
All four families had a positive attitude towards reading throughout the generations, however, lack of English knowledge and pre-occupation with more imminently important matters (e.g. working to provide food for the family) meant that the grandmothers in the study had little parental input into their early English reading experiences as children. For their generation, reading occurred in school, and this experience was often separate from family life, taking place irrespective of and disconnected from heritage language and cultural activities. When these women became mothers themselves, they were more confident in assisting their children, and more likely to utilise available resources, such as the library. Between their own daughters growing up and the current generation of young children, another shift occurred, due to the increasing availability of information regarding early reading, and access to grandparents who had themselves struggled through the British education system. In all four families, the increased availability of English resources and focus on success within the English system have resulted in the marginalisation of the heritage language in family reading, but with a desire to maintain the heritage language orally. Interestingly, the only family which speaks no heritage language at all with the children is the Patels, where the mother herself arrived from India, re-introducing a first-generation arrival to the family, and in this family, there is a strong perception that only English is needed for the children's future. Throughout the interviews, the wish for the children's success was particularly strong in the early generationsthe interviewed grandmothers, with their memories of their own parents' hardships at their time of arrival in the UK. This may illustrate a prevailing belief that success within the English schooling system has to occur at the expense of the mother tongue, despite research highlighting the contrary (Kenner et al. 2004). The loss of the heritage language is a particular concern in the younger generations. Family engagement with literacy practices begins with a focus on English as the language of instruction and assimilation, but, across the generations, describes a 'generational arc', ending in young mothers who feel unequipped to support their child in learning the heritage language. Understanding and tracing family stories and histories can help to understand early childhood family practices, and further research is needed to follow explore connotations around heritage languages and identity, especially at the point of young children beginning formal education. Until national policy clearly outlines benefits of bi-and multilingualism, and 'family literacy' explicitly highlights benefits of reading at home in the heritage language, building on current and new research data, both families and practitioners will likely remain confused, thus preventing children from fully benefiting from their linguistic heritage.

Disclosure statement
No potential conflict of interest was reported by the author.

Funding
While the aspects highlighted in this paper are unrelated to any funding, the initial spark was provided via involvement in study funded by Booktrust, exploring historical engagement with family literacy practices.