Bilingualism in the Early Years: What the Science Says

Many children in North America and around the world grow up exposed to two languages from an early age. Parents of bilingual infants and toddlers have important questions about the costs and benefits of early bilingualism, and how to best support language acquisition in their children. Here, we separate common myths from scientific findings to answer six of parents’ most common questions about early bilingual development.

Bilingualism in the Early Years: What the Science Says

Bilingual parents are vocal in their desire to raise proficient, dynamic bilingual children. They have questions, and they want answers. But there is a complicated history of positive and negative press about raising children in bilingual households, to the point where some pediatricians—even today—recommend against exposing children to two languages. Attitudes against early bilingualism are often based on myths and misinterpretations, rather than scientific findings. Here, we aim to address the most frequently asked questions about childhood bilingualism using research findings from a variety of scientific fields including developmental psychology, cognitive psychology, education, linguistics, and communication sciences and disorders. This article is intended for parents and the many people who parents turn to for advice about fostering successful bilingual development: preschool teachers, elementary teachers, pediatricians, and speech-language pathologists.

Bilingualism refers to the ability to use two languages in everyday life. Bilingualism is common and is on the rise in many parts of the world, with perhaps one in three people being bilingual or multilingual (Wei, 2000). Contact between two languages is typical in regions of many continents, including Europe (Switzerland, Belgium), Asia (India, Philippines), Africa (Senegal, South Africa), and North America (Canada). In the United States, a large (and growing) number of bilinguals live in California, Texas, Florida, New York, Arizona, and New Mexico. In California, for example, by 2035, it is expected that over 50% of children enrolled in kindergarten will have grown up speaking a language other than English (García, McLaughlin, Spodek, & Saracho, 1995). Similarly, in some urban areas of Canada such as Toronto, up to 50% of students have a native language other than English (Canadian Council on Learning, 2008).

Despite the prevalence of bilingualism, surprisingly little research has been conducted on the topic, particularly on the foundations of bilingual language learning in infants and toddlers. The science of bilingualism is a young field, and definitive answers to many questions are not yet available. Furthermore, other questions are impossible to answer due to vast differences across families, communities, and cultures. But with an accumulation of research studies over the last few decades, we are now equipped to partially answer some of parents’ most pressing questions about early bilingualism.

There are few venues for communicating scientific findings about early bilingualism to the public, and our goal is to distill bilingual and developmental science into practical, accessible information. We are researchers who study bilingual infants and children, and as such, we interact with bilingual families regularly. When we give community talks to preschools and nonprofit organizations about language development in early childhood, the question-and-answer period is invariably dominated by questions about early bilingualism. The consistency in questions is astonishing. Are bilingual children confused? Does bilingualism make children smarter? Is it best for each person to speak only one language with a bilingual child? Should parents avoid mixing languages together? Is earlier better? Are bilingual children more likely to have language difficulties, delays, or disorders? This article is organized around these six common questions.

3. Is it best for each person to speak only one language with a bilingual child?

One popular strategy for raising bilingual children is “one-person-one-language,” a strategy first recommended over 100 years ago (Ronjat, 1913). Theorists originally reasoned that associating each language with a different person was the only way to prevent bilingual children from “confusion and intellectual fatigue.” While appealing, this early notion has been proven false. As discussed above, there is no evidence that bilingual children are confused by early bilingualism, and the cognitive benefits associated with bilingualism run counter to the notion of “intellectual fatigue.”

It is still important to consider what strategies families can use to promote early bilingual development. Research has shown that a one-person-one-language approach can lead to successful acquisition of the two languages (Barron-Hauwaert, 2004), but that it does not necessarily lead to successful acquisition of the two languages (De Houwer, 2007). Further, children who hear both languages from the same bilingual parent often do successfully learn two languages (De Houwer, 2007). A one-person-one-language approach is neither necessary nor sufficient for successful bilingual acquisition.

Several other factors have proven to be important to early bilingual development. These factors might lead some families to use a one-person-one-language strategy, and other families to use other strategies. First, it is important to remember that infants learn language through listening to and interacting with different speakers. Infants need to have a lot of exposure to the sounds, words, and grammars of the languages that they will one day use. Both quality and quantity matter. High quality language exposure involves social interaction—infants do not readily learn language from television (DeLoache et al., 2010; Kuhl, Tsao, & Liu, 2003), and low-quality television viewing in infancy has been linked to smaller vocabulary sizes in bilingual toddlers (Hudon, Fennell, & Hoftyzer, 2013). Opportunities to interact with multiple different speakers has been linked to vocabulary learning in bilingual toddlers (Place & Hoff, 2010).

Quantity can be measured by the number of words that children hear per day in each language. Quantity of early exposure has a profound effect on children’s ongoing language development: hearing more words gives children a greater opportunity to learn a language, which leads to later advantages in school performance (Hart & Risley, 1995). For bilingual children, it is important to consider the quantity of their exposure to each language. While a bilingual’s two languages do influence each other to a certain degree (Döpke, 2000), in many ways they travel on independent developmental paths. Bilingual children who hear a large amount of a particular language learn more words and grammar in that language (Hoff et al., 2012; Pearson & Fernández, 1994), and show more efficient processing of that language (Conboy & Mills, 2006; Hurtado, Grüter, Marchman, & Fernald, 2013; Marchman, Fernald, & Hurtado, 2010). Bilingual parents thus need to ensure that their children have sufficient exposure to the languages they want their children to learn. We return to this topic in the next sections.

Relatively balanced exposure to the two languages is most likely to promote successful acquisition of both of the languages (Thordardottir, 2011). In situations where each parent spends equal time with a child, one-parent-one-language can be a great way to ensure equal exposure. Conversely, exposure to a second language only when grandma and grandpa visit on the weekend, or when a part-time nanny visits on a few weekdays, or when a language class meets on Thursday nights, will not lead to balanced exposure. Imagine an average infant who sleeps about 12 hours a day, and so is awake 84 hours per week. A single afternoon (~ 5 hours) is only about 6% of the child’s waking life, and this exposure alone is unlikely to lead to acquisition of a language. Similarly, in homes where one parent is the primary caregiver, a one-parent-one-language is unlikely to lead to balanced exposure.

Unfortunately, providing perfectly balanced exposure in the early years will not necessarily ensure later bilingualism. As children become older, they become more aware of the language spoken in the community where they live, and are likely to use this language at school. This is known as the majority language, while other languages that are not as widely spoken are known as minority languages. Even if initially learned in preschool, minority languages are much more likely than majority languages to be lost as development continues (De Houwer, 2007). Many experts recommend providing slightly more early input in a minority than in a majority language, and where possible providing children with opportunities to play with other kids in that language (Pearson, 2008). Raising a bilingual child in communities that are largely bilingual such as Miami (Spanish-English), Montreal (French-English), and Barcelona (Catalan-Spanish) provides fewer challenges for ensuring the ongoing use of the two languages.

So what language strategies should parents use? The best answer is that parents should use whatever strategy promotes high-quality and high-quantity exposure to each of their child’s languages. This could include structured approaches such as using different languages as a function of person (one-person-one-language), place (one language at home, one language outside), or time (alternating days of the week, or mornings/afternoons). Some parents insist on speaking only one language with their child, even if they are able to speak the other (Lanza, 2004), to ensure exposure to a particular language. Other families find that flexible use of the two languages, without fixed rules, leads to balanced exposure and positive interactions. Each family should consider the language proficiency of each family member as well as their language preference, in conjunction with their community situation. Families should regularly make an objective appraisal of what their child is actually hearing on a daily basis (rather than what they wish their child was hearing), and consider adjusting language use when necessary.

4. Should parents avoid mixing languages together?

Many parents of bilingual children are bilingual themselves (Byers-Heinlein, 2013). Code mixing—the use of elements from two different languages in the same sentence or conversation—is a normal part of being a bilingual and interacting with other bilingual speakers (Poplack, 1980). Code mixing is relatively frequent amongst bilingual parents as well (Byers-Heinlein, 2013), and even parents who have chosen a one-parent-one-language strategy still code mix from time to time (Goodz, 1989). But what effects does hearing code mixing have on the development of bilingual children?

Research on the impact of code mixing on bilingual children’s development is still quite limited. One study of 18- and 24-month-olds found that high amounts of code mixing by parents was related to smaller vocabulary sizes (Byers-Heinlein, 2013). However, other studies have found no relationship between code-mixed language and early language development (Place & Hoff, 2011). Further, studies are beginning to reveal that bilingual children as young as 20-months are able to understand code-mixed sentences, and show similar processing patterns as bilingual adults (Byers-Heinlein, 2013). This would suggest that bilinguals are able to cope with code mixing from an early age. It has also been suggested that while code mixing might make word learning initially difficult, it is possible that practice switching back and forth between the languages leads to later cognitive benefits (Byers-Heinlein, 2013). Unfortunately, the jury is still out on whether exposure to code mixing has developmental consequences for bilingual children, but we are currently working on several research projects that will help answer this question.

It is important to note that considerations of code mixing also have important social implications. In some communities, code mixing is an important part of being bilingual and being part of a bilingual community. For example, code mixing is the norm in some Spanish-English communities in the U.S., and Afrikaans-English code mixing is the norm in some parts of South Africa. Different communities have different patterns of and rules for code mixing (Poplack, 1984), and children need exposure to these patterns in order to learn them.

6. Are bilingual children more likely to have language difficulties, delays, or disorders?

Bilingual children are not more likely than monolingual children to have difficulties with language, to show delays in learning, or to be diagnosed with a language disorder (see Paradis, Genesee, & Crago, 2010; Petitto & Holowka, 2002). Parents’ perceptions are often otherwise—they feel that their child is behind due to their bilingualism—revealing an interesting disconnect from scientific findings. Science has revealed an important property of early bilingual children’s language knowledge that might explain this misperception: while bilingual children typically know fewer words in each of their languages than do monolingual learners of those languages, this apparent difference disappears when you calculate bilingual children’s “conceptual vocabulary” across both languages (Marchman et al., 2010). That is, if you add together known words in each language, and then make sure you don’t double-count cross-language synonyms (e.g., dog and perro), then bilingual children know approximately the same number of words as monolingual children (Pearson, Fernández, & Oller, 1993; Pearson & Fernández, 1994).

As an example, if a Spanish/English bilingual toddler knows 50 Spanish words and 50 English words, she will probably not appear to be as good at communicating when compared to her monolingual cousin who knows 90 English words. However, assuming 10 of the toddler’s Spanish words are also known in English, then the toddler has a conceptual vocabulary of 90 words, which matches that of her cousin. Even so, knowing 50 vs. 90 English words could result in noticeably different communication abilities, but these differences are likely to become less noticeable with time. This hypothetical example about equivalence in vocabulary is supported by research showing that bilingual and monolingual 14-month-olds are equally good at learning word-object associations (Byers-Heinlein, Fennell, & Werker, 2013). This offers some reassurance that young bilinguals—like young monolinguals—possess learning skills that can successfully get them started on expected vocabulary trajectories. There is also evidence that bilingual children match monolinguals in conversational abilities; for example, when somebody uses a confusing or mispronounced word, or says something ambiguous, bilingual children can repair the conversation with the same skill as monolinguals (Comeau, Genesee, & Mendelson, 2010).

Just like some monolingual children have a language delay or disorder, a similar proportion of bilinguals will have a language delay or disorder. Evidence that one bilingual child has a language difficulty, however, is not evidence that bilingualism leads to language difficulties in general. The challenge for pediatricians and for speech-language pathologists is to decide if a bilingual child does have a problem, or whether her errors are part of normal development and interaction between the sounds, words, and grammars of her two languages. If parents are worried that their bilingual child does have a delay, they should first consult their pediatrician. Pediatricians sometimes have a tendency to say, “Don’t worry, her language is completely normal.” This statement will end up being false for some children who will end up diagnosed with language difficulties, but it is more likely than not to be true, especially considering that parents can be inaccurate when estimating their bilingual child’s language skills. In some other cases, health care providers with concerns about language impairment may recommend against raising a child in a bilingual environment. This recommendation is not supported by the science of bilingualism. Bilingual children with specific language impairments (Paradis, Crago, Genesee, & Rice, 2003), Down syndrome (Kay-Raining Bird et al., 2005), and autism spectrum disorders (Peterson, Marinova-Todd, & Mirenda, 2012) are not more likely to experience additional delays or challenges compared to monolingual children with these impairments.

If parents do not feel comfortable with a pediatrician’s opinion, they should find (or ask for a referral to) a speech-language pathologist with expertise in bilingualism, if at all possible. Early intervention increases the likelihood of a positive outcome. The problem is that few clinicians receive quality training about the learning needs of bilingual children, which in some cases leads to a misdiagnosis of bilingual children as having delayed or disordered language (Bedore & Peña, 2008; Kohnert, 2010; Thordardottir, Rothenberg, Rivard, & Naves, 2006). The time is past due to eliminate such simple misunderstandings in clinical settings. A bilingual clinician, or an individual who has training in bilingualism, will take care in assessing language skills in both languages, in order to measure the child’s entire language profile. Parents should keep in mind that clinicians have a very difficult job when it comes to assessing bilingual children. They have to (1) accurately assess a bilingual child’s language abilities in each of her languages, (2) integrate the child’s problematic and unproblematic abilities in terms of sounds, words, grammar, and conversation in each language into a coherent whole, (3) evaluate whether the child is delayed and/or disordered in one or both languages, (4) weigh the child’s linguistic/cognitive capacities in comparison to typically and atypically developing monolingual children and, when possible, bilingual children of the same age, and (5) develop an effective intervention that targets subareas of linguistic/cognitive competence in one and/or both languages. This is a tangled landscape for intervention, but one that can be assessed thoughtfully. Regardless of whether parents pursue intervention, they can help children gain bilingual proficiency by using both languages as regularly as possible in enriching and engaging contexts. Furthermore, parents should keep in mind that both monolingual and bilingual children can best show off their skills when using language that matches their daily experiences (Mattock, Polka, Rvachew, & Krehm, 2010).

In summary, if you measure bilinguals using a monolingual measure, you are more likely to find false evidence of delay. Fortunately, researchers and clinicians are now developing bilingual-specific measures that paint a more accurate picture of bilinguals’ global language competence.