Baby, talk to me: How children get their accents

“When we look at how it is that kids learn not just a language, but the dialect of their environment and its socially relevant variation, we find that adults are the equivalent of linguistic Yodas, subtly providing life lessons through unconscious modeling in how we say what we say to kids.”

Accents don’t just arrive on the heels of another language; they also arise as a by-product of becoming the socially attuned creatures that we are. Even within our first language, there is a lot of variation in how things are said. Some of this variation comes to be recognized as important, because either it points to a specific group (like a Southerner saying “far” for fire) or it points to a social meaning, like knowing that “got to” spoken with all its “t”s means business, while “gotta” comes off a lot more chill. But how is it that kids learn not just the sounds of their language, but which pronunciations are the ones they should use as well as when and where?

Well, moms and dads, you might have thought you were off the hook in this whole language learning thing, given the natural predisposition of infants to pick up language, but where would the fun in that be? You still have a linguistic role to play, at least until the teen years, and much of that involves teaching your kiddos the ins and outs of the social meanings associated with the sounds around them. I know, throw this on top of the need to feed them, clothe them, and monitor their social media and you’re exhausted already, right? Never fear, you and your kid’s natural language instincts are already hard at work, and I bet you barely even noticed.

When we look at how it is that kids learn not just a language, but the dialect of their environment and its socially relevant variation, we find that adults are the equivalent of linguistic Yodas, subtly providing life lessons through unconscious modeling in how we say what we say to kids. The secret sauce lies in what we fancy linguists call child-directed speech (CDS), or what people who actually have friends call baby talk or caregiverese. Yup, that elongated, repetitious, slowed down, singsongy drivel that we patty-cake to our babies seems to be part of the magic recipe for helping kids learn not just how to divide our speech into separate sounds and words, but also to understand the social patterning of those sounds and when they are important.

Now, I know there are often glossy parenting magazines or internet blogs that debate the value of baby talk, but, scientifically speaking, baby talk does seem to offer linguistic pointers to infants, helping them break down the stream of speech coming at them in various ways, and there is no evidence it holds babies back at all. Do we absolutely need to do it? Probably not, as babies in cultures that do not use as much baby talk are able to acquire language without issue. Instead, it merely speeds up a process that happens naturally and it makes us feel like we are connecting with our infants.

Not surprisingly, much of the research on baby talk focuses on mothers, possibly because male scientists are never home long enough to chat up their kids. Still, that research has pointed to some interesting patterns in the types of speech moms use with their babies despite pigeonholing them as baby raisers. For one, mothers tend to be hyperarticulate in terms of how they pronounce their vowels when talking to their baby, which is likely a way of helping the little bean learn to recognize the ones that are important in their language. They also have been found to emphasize contrasts between consonants like “t” and “d” or “k” and “g” more than when talking to adults. Again, this makes sense as an assist to babies in figuring out the basic categories of their language.

But the role of moms doesn’t appear to stop with the highlighting of what sounds their language will have. Research suggests mom also models the latest in linguistic finery—early on using the most standard versions of sounds with her pumpkins, even if she rarely uses them when chatting with other adults. Linguist Julie Roberts looked at Southern pronunciations like those of “bah” for bye and “todd” for tide in the speech of moms in Memphis, Tennessee. She found that all the moms she studied used the more standard pronunciations when talking to their kids compared to when they were talking with the interviewer, a pattern echoed in other studies. For example, in a working-class town called Tyneside in Northern England, where residents are known as Geordies, researchers similarly found mothers using fewer local forms with their kids, especially in the earliest years, than they did when speaking to other adults.

Why would this be? It might be driven by mom’s desire to teach her kiddo the speech forms that are most widely accepted, but since kids are often there when mom talks to adults using her more typical speech, it is also a way of introducing babies to the types of variability they will encounter and showing that “friend talk” involves the more local or casual variants—in other words, hinting at values that have become attached to forms. We also find studies that suggest that when moms repeat something, they often use a more reduced form than in their original saying of that word or phrase—“Is that your little toe? Where’s ya’ li’ toe?”—which is a signal of intimacy and closeness. In contrast, when moms are in teaching or explanatory modes, they tend to use more full or standard forms. All of this modeling helps teach kids the basic sounds they need to command and provides clues as to how to interpret the different pronunciations they hear around them.

Interestingly, what moms highlight in terms of usage norms also seems to change over time. Research finds that moms decrease the use of hyperarticulate forms and standard pronunciations as babies get older. Going back to that study looking at mother/child interactions in Tyneside, moms started using more of the local Geordie accent with their kids as they aged up from two to four years old. This suggests that the earlier pronunciations are focused on helping kids figure out the basic sounds their language has, and once that is accomplished, moms shift to using the specific accents that will prove socially valuable to their kids in their local environment.

Some evidence we have for this interpretation is the fact that moms often use more positively valued forms (like “talking” versus “talkin’”) with girls than they do with boys, which tells us that they are trying to use the pronunciations that will prove most socially beneficial to their kids. In this way, moms are attending to the generalized cultural belief that girls need to be ladies, while boys can get a bump for being tougher or coming across as a bit more slangy. When we look at studies of adult speech, this pattern of women tending to use “better” pronunciations holds—women often use more standard forms of speech compared to men. Dads, on the other hand, don’t seem to change up their speech when directed to children nearly as much—something that seems to reinforce the associations of more local or less standard speech as a boy thing.

All this brings us back to how we get regional, class, or ethnic accents within the same language. While moms and dads provide some of our initial training, their influence becomes less important once we discover something way more exciting: cool peers. What we find in sociolinguistic research on kids exposed to more than one dialect, say when moving to a new area, is that success in getting a new accent down depends greatly on two factors: the age when they move and their integration with speakers of the new dialect.

A number of studies have traced kids from the same families that have moved from one place to another where a different dialect is spoken and discovered that, as with acquiring a completely new language, the youngest kids are more successful at picking up native-like abilities in the new dialect compared to their older siblings. That success is in large part driven by their early exposure to a peer group in preschool or kindergarten, something that also explains why kids who are born into families with a different home language don’t end up sounding like their parents.

One well-known study looked at kids who had moved to Philadelphia from elsewhere in the country at different ages and found that kids who moved before age four ended up sounding very similar to Philly natives. Those older than four picked up many accent features, i.e., they might drink “wooder,” the iconic Philly pronunciation of water, but their accent was not completely native-like.

Interestingly, where one’s parents were from made a difference even for locally born kids. Those with parents also from Philly were the only ones who were able to replicate all aspects of the local accent perfectly, while those with out-of-state parents were not. This doesn’t mean that they couldn’t pass for locals, but simply that, from a linguistic perspective, they were not completely conforming to the local dialect patterns. What this tells us is that our local accents start forming even before we are born, probably when listening to our mother’s speech in utero, and are darn hard to get rid of even in the most perfect circumstances.

When we consider that most of us grow up and go to school with people from backgrounds, neighborhoods, and ethnicities similar to ours, the idea that it is somehow simple to just “pick up” a new or socially preferred accent is pretty ludicrous. Second dialect acquisition works similarly to second language acquisition—once your system is set, picking up another variety gets much more difficult because your mouth and mind have already zeroed in on specific sounds and patterns. Being surrounded by speakers of that new variety certainly helps, especially if you are strongly motivated because of your peer group, but that means most speakers will shift closer to the local accents of their friend group, not toward some other accent that teachers or principals use. And, since a lot of us don’t move until we are adults, our ability to do the accent hokey-pokey is often close to nil. The gist is that much of the way we will sound—whether we are talking about a foreign accent or an English one—is related to our experience with language and its speakers in early childhood.

• Amazon
• Bookshop

If you’d like to read more about the science of baby talk, check out my own series on the topic:

🔢 The science of baby talk

• Part 1: Why you should be talking to your infant
• Part 2: What’s the point of baby talk?
• Part 3: Is baby talk good for your child?
• Part 4: Do all cultures use baby talk?
• Part 5: Baby talk in the languages of the world
• Part 6: How much should you talk to your child? [forthcoming]
• Part 7: What really matters when talking to your child