Growing up in Greece, a country with a conservative monolingual policy, meant that most of my experiences with other languages came at secondary school or private language institutes. A lot has changed since then: people travel more (or sometimes they are forced to leave their countries), it has become much easier to read and listen to content in other languages, and it is no longer self-evident that your family all share the same linguistic background. What all of this means is that, unlike people in my generation, children today are growing up in a very complex linguistic context. It also means that much of what we know about linguistic development is not as helpful as it used to be, because it tends to reflect simpler societal norms.

This is a research gap that my research colleague, Prof. Eleni Motsiou, and I have tried to address in an article entitled “Family language policy in mixed-language families: An exploratory study of online parental discourses“, which was just accepted for publication by the International Journal of Bilingual Education and Bilingualism. In the article, we take a focused look into the topics that parents in mixed-language families discuss, with a view to mapping out their beliefs, the problems they face, and the strategies they use as they raise their children to be plurilingual.

You can find a link to a free copy of the article by scrolling to the end of this post. In this post, you can also read about:

1. What prompted this study;
2. Some background information on language policy and family language policy;
3. How we used digital ethnography to study mixed-language families;
4. What the data suggests about mixed-language families and their language policy;
5. Some practical implications for mixed-language family parents.

Some background on mixed-language families

What prompted us to do this study was the realisation that mixed-language families are not just more common now than they used to be, but they are also qualitatively different.

New types of mixed-language families

In the past, mixed-language families tended to fall in one of two categories. One family type involved couples where one spouse was an expat living in a more-or-less homogeneous linguistic community other than the one in which they grew up. For example, a Brit might decide to live in Greece, where he or she might marry and start a family. Typically, in such a scenario, parents would follow the so-called One-Parent-One-Language pattern of interaction: each parent would speak to their children in their own native language.

The other type of mixed-language family that used to be dominant in the past usually involved migrant couples moving to a different country with their family, or starting a family there. In the Greek context, a typical example might involve a couple moving to Western Europe with their children. Another example would be if members of a cultural/linguistic group (e.g., Greek-Americans) married within their cultural community and decided to pass on Greek as a heritage language to their children. In this scenario, parents would likely use their heritage language at home and the community language in social situations.

The challenge we are facing is that, as people become more and more transnationally mobile, such clear-cut distinctions are becoming increasingly harder to apply. It is not uncommon nowadays to see families where parents come from different backgrounds (e.g., Greek and Italian), live in a third country (e.g., Germany) and use English as an intra-family lingua franca. We found these situations intriguing because it is not always clear how these families make decisions about how to use the languages at their disposal. Some questions we were keen to find out more about included: which languages do parents want to prioritise (and why?), which languages do children prefer, and what methods do parents use to help their children develop?

Family language policy

Spolsky’s model of language policy

Bernard Spolsky (2004), who put forward an influential model of language policy, described it as having three components:

1. language ideology, i.e. a set of beliefs, including those unconsciously held, about language (e.g., the assumption that one language is ‘better’ or ‘more useful’ than others).
2. language management, i.e. laws, rules and regulations, included unstated ones, about which language or dialect is appropriate to what context (e.g., defining which language teachers and students should use at school). Language management might reflect ideology, but doesn’t always do so.
3. language practices, i.e. the actual linguistic choices that people make when communicating. Language practices may conform to management, but don’t always do so; similarly, they may reflect ideology or they may be at a dissonance with it.

Our model of family language policy

Our own thinking about family language policy was influenced by Spolsky’s model, which we tried to apply to the level of family units. Scaling down the model involved one important modification. Because our focus was on parents of young children, parent-child interaction was both a language practice and an attempt to influence the children’s linguistic development. As a result, it proved hard for us to sustain a rigid management-practice distinction, and we therefore merged the two categories into one.

Our family language policy model thus contained the following two parts, an ideological and a practical one:

• Linguistic Ideology: beliefs, attitudes, concerns etc. about language and linguistic development;
• Language Transmission and Management (LTM): strategies and practices that parents enact in order to foster their children’s linguistic development.

---

The study

To find out more about the linguistic ideology and LTM of mixed-language families, we engaged in a digital ethnography project. Digital ethnography, or cyber-ethnography, is an adaptation of ethnographic work, applied to online communities.

Specifically, we collected a year’s worth of publicly available posts and comments from two online parent communities. One was an international community, from which we only collected content from Greek-speaking parents outside the Greek-speaking world (Community A). The other was a community of parents in mixed-language families living in Greece and Cyprus (Community B). Information about both communities is summarised in the table below. Because Modern Greek was the majority language in Community B, and a minority/heritage language in Community A, we expected that we might find differences in the LTM practices of the two groups.

Online community	Members (as of 3/3/19)	Posts (Comments per post)	Words
Community A	40,315	64 (1-52)	112,184
Community B	1,253	166 (0-128)	129,994
Total		230	242,178

Table 1. Overview of online communities and data corpus

We analysed the data using a procedure called thematic analysis. This involves close reading of the data, grouping them into categories, and working with the patterns that emerged. We used Spolsky’s model as a rough guide in this process, but other than that we kept an open mind: this meant that we generated new categories for diverse problems, ideas, strategies and techniques that parents used, and rearranged them in ways that made sense.

We eventually ended up with much more data than we expected. In addition to our Family Language Policy data, this also included categories about interacting with school and health professionals. We are still in the process of making sense of what these data can tell us; our article in the International Journal of Bilingual Education and Bilingualism, however focuses only on Family Language Policy, which we believe is more practically relevant to many parents.

---

What we found about mixed-language families

Linguistic ideology

• Attitudes towards plurilingualism are very positive in the communities we studied. Many parents spoke very much in favour of knowing many languages in general, and some suggested that such knowledge may also have broader cognitive effects. They also spoke positively about specific languages that they wanted their children to master. On the whole, pluriligualism seemed to be viewed as a ‘new kind of normal’.
• Most parents were very keen on preserving heritage/minority languages. This was true both for people living in Greece and Cyprus and for people living in contexts where Greece was a non-majority language. We believe that this connects to the positive attitudes towards Modern Greek in most international contexts. There are some indications in our data, however, that speakers of ‘low-prestige’ languages (e.g., Asian languages) in Greece may be reluctant to use their native languages.
• Some anxiety was expressed about the possible negative developmental effects of plurilingualism. Perhaps unsurprisingly, many young parents seemed concerned about developmental milestones, and they often expressed uncertainty whether ‘they were getting it right’ when it came to their children’s linguistic development. On other occasions, parents channeled the negative views of teachers and health professionals that were reportedly sceptical about plurilingualism. However, the overall discourse in the communities was highly supportive, and the parents tended to affirm each other’s LTM practices.

Language Transmission and Management

We documented 14 different Language Transmission and Management strategies and practices, which we grouped in five main categories, as follows:

• One Parent One Language (OPOL), and variants: This practice involved the exclusive use of one language by each parent. Typically, this would be the parents’ native language. This was one of the most widespread practice we documented. In fact, many parents very strongly believed that it should be followed without deviations. Despite that, many parents also reported that they would often adapt their practices to communicative situations. For when friends were present who would not understand the parent’s native language, they might use OPOL more flexibly.
• Reinforcing Non-Majority Languages: This category encompassed a broad set of practices, such as Minority Language at Home, or Minority Language Everywhere. These practices involved both parents strategically using a minority or heritage language that they wanted to reinforce. Sometimes this also required them to learn the target language, as it was not necessarily one that they spoke natively.
• Situational selection: Some parents reported that they would adjust to the communicative situation according to some systematic pattern (e.g. One Situation One Language, Two Parents Two Languages or Situational switching). For example, a parent might use a heritage or minority language when engaging in ordinary conversation with their children, but switch to the community language when helping their children with homework.
• Ad hoc use of multiple languages: This category referred to the non-systematic use of multiple languages. One of the participants succinctly described this practice as follows: “There’s no such thing as a must in communication; just choose the languages that come naturally to you”.
• Reinforcement strategies: This category referred to specific, often improvisational, techniques that parents used in order to help children develop one or more languages. Some examples of such strategies involved repeating every sentence in both languages, or parents pretending that they could not understand their children’s dominant (‘best’) languages.

---

What does this mean for parents?

In our article, we took a descriptive approach to family language policy, so we did not really engage with practical implications. We aim to do this in a next step of our project, where we will conduct interviews with mixed-family parents.

However, I do think that there are some takeaways from these findings as well. The first lesson is that there are multiple ways to approach plurilingual development, all of which seem to be equally effective on the long run. This is an important thing to remember, because there is much advice out there which tends to be rather too prescriptive. Some parents in our data, for example, categorically stated that anything other than a strict One-Parent-One-Language policy would confuse children. Others insisted that the minority language should be used exclusively, even around friends and family members who may not understand it. Establishing consistency is definitely helpful, and I do not doubt that this advice is well-meaning. However, I am concerned that this kind of discourse could create unrealistic expectations for young parents. I am also concerned that it may be causing unnecessary feelings of guilt, at a time when they are already experiencing a lot of stress. So, one thing to keep in mind is: as long as you provide children with enough quality linguistic interaction, they will work things out well enough on their own.

Another implication from our data is that, as parents, we could help our children a lot by encouraging them to be linguistically flexible. We live in a world where languages co-exist in much closer proximity than in the past. Children will come into contact with peers from different linguistic background at their schools, and they will engage with input in multiple languages through traditional and social media. One way to prepare them for this fluid situation is to help them to draw from their linguistic repertoire any language code that is appropriate – and to model such behaviour ourselves. The type of systematic language switching that we documented in some families seems to be closer to this goal than less flexible approaches.

One final remark concerns the ways in which we, as parents, interact with education, speech and health professionals. Although the benefits of plurilingualism are more or less uncontested in the scientific literature, we were somewhat disconcerted to notice several reports in our data, where parents were told by teachers and doctors to stick to one language only. In some cases, teachers seemed too keen to blame school difficulties to what they perceived as linguistic deficits associated ‘confusing input’; similarly, some doctors attributed variation in achieving developmental milestones to bilingual experiences. This was especially the case with families who spoke ‘low-prestige’ languages, such as immigrants from Eastern European and Asian countries. These parents were further disadvantaged by lack of social capital that might help them challenge this advice. The thing to remember is that such recommendations are grounded on social attitudes, not scientific evidence; and that your family language policy should reflect the needs and priorities of your family, not social expectations, even when these are expressed by people in authority.

---

The article

If you or your institution subscribes to the International Journal of Bilingual Education and Bilingualism, you can access the article from this link.

Alternatively, you can download a post-print version of the article by clicking on the button below:

This post-print version incorporates changes in response to peer-review, but does not include final copyediting and proofreading. Permission for self-archiving has been granted in accordance with the publishers’ copyright and self-archiving policy. You are welcome to use this document for your own reference. You may also cite the document, but please bear in mind that the pagination is different from the version of record.

For those of you who find this kind of information useful, the suggested (APA) reference for the article is:

Kostoulas, A. & Motsiou, E. (2020). Family language policy in mixed-language families: An exploratory study of online parental discourses. International Journal of Bilingual Education and Bilingualism, 25(2), 696-708. https://doi.org/10.1080/13670050.2020.1715915

---

Whether you are a parent in a mixed-language family or a researcher with an interest in family language policy, I hope that you found this post useful. I am very keen to read any feedback you have in the comments section below, or alternatively you can always contact me directly. Also please feel free to use the social sharing buttons below to forward this information to anyone who might find it interesting.

---