Watching children learn to talk has always been one of my greatest joys. It is a passion that has implicitly informed my professional path since I took a job in a toddler daycare fresh out of college. While there I could bear daily witness to the explosion in vocabulary that accompanies the second and third years of life. What better window into the mysterious magic of the human brain? At the time, I had no idea how much has already happened beneath the surface, long before that amazing moment when a new word is spoken. From birth, infants are silently hard at work: learning which sounds in the stream of ambient noise deserve the important designation of “word”, then learning the meaning associated with these special vocalizations, before finally learning to replicate what they have heard. It is no wonder that during the first 3 years of life, 700 new synaptic connections are formed each second (that’s over 60 million neuronal links created every single day!) No cognitive achievement in our adult lives matches the complexity and breadth of early language learning. Babies are clearly geniuses.
Here’s a glimpse into the amazing process that is quietly unfolding within the mind of my baby right now: Trytoreadasentencewithoutanyspacesbetweenthewords… That is, try to read a sentence without any spaces between the words – and you can imagine what the flow of routine speech must sound like to the ears of a young infant. Prior to comprehending words, they must first begin to segment sound into word units. Though we cannot see it, babies undergo a major perceptual shift between 6 and 8 months old. At 6-months, they are unable to segment words from ongoing speech. (Werker, p205) By 7-months, they begin to do so by listening for sequential probabilities: how often do 2 syllables show up together? They learn that statistically, “baby” is much more likely to be a word than “byba”. (Kuhl, ELA p834).
Just as quickly, they shift to an entirely different strategy. By 7.5 months they have begun to use “language-specific rhythmic biases” to guide segmentation. (Werker, p206) For example, 90% of English words stress the first syllable, as in ‘pencil’ and ‘stapler’. At 7.5 months-old, “English-learning infants can segment words from speech that reflect this strong–weak pattern, but not the weak–strong pattern. For example, when they hear ‘guitar is’ they perceive ‘taris’ as the word-like unit, because it begins with a stressed syllable” (Kuhl, ELA p835). How I love imagining the change in Dani’s awareness, over the last couple of months, as the steady stream of sounds she is immersed in within our household (it is steady and abundant: my wife and I are talkers!) begins to take shape into distinguishable words.
Just recognizing words must be challenging enough for Dani, especially since each speaker says them slightly differently. I cannot begin to fathom how she makes the leap into understanding meaning. This is such a profoundly complex process, and theories abound as to how it occurs. The general consensus is that babies don’t begin to grasp word meaning until between 9 and 13 months old (Kuhl, NS p520). The hypothesis is that they need to share in social referencing first, a skill that begins around 9-months, when infants learn to follow the pointing finger of a parent and place their joint attention on the referred object. However, a recent study demonstrates that some babies do know how to link a few words with the objects they represent by as young as 6-months old (Bergelson, p3253)
Certainly most babies understand a few key words, such as their names, well before the 9-month mark. “Mommy”, “daddy”, “no” and “milk” are also often recognized early. The words that babies learn first are the ones they and their family have deemed most important. As such, they share a number of common traits: they are stressed – the tone, facial expression and other non-verbal cues of the speaker combine to emphasize “pay attention to this”. They are said frequently, and they are often spoken in single word utterances, eliminating the need to distinguish them from the stream of fluent speech (Roy B., p1)
Learning to speak places even greater demands upon the infant brain. Along with the areas that control hearing and comprehension, those responsible for motor movement must simultaneously be engaged. Brain imaging studies have shown that the motor systems of young infants are already active as they listen to their parents talk. Well before they have the muscle control to speak the sounds, babies’ brains are planning what they will say (Kuhl).
Babies are approximately 6-months old when they “first display coordination of the gestures involved in closing the vocal tract for a consonant” (Werker, p198) This allows them to produce what is known as canonical babbling: repetitions of the same syllable (such as “bababa”). This is followed at about 8–9 months by variegated babbling: the ability to combine varying syllables in a single production (such as “babeedo”). Since the earliest appearing syllable forms are those that are the easiest to produce, the consensus is that “in initial stages of development, there do appear to be universals in vocal production.” (Werker, p198) In other words, early babbling appears to be identical across languages.
My own passion for language development can be traced back to this realization. When I was 16 years old, I went on a student exchange trip to Russia. My first night there, reeling with culture shock, I found solace in watching the infant cousin of my exchange host. She was probably about 9-months old: crawling, pulling herself up, and babbling. I remember being blown away by the familiarity she offered me. Here I was on the other side of the planet, encountering a culture that was very much seen as the “enemy” of my own – and the baby sounded exactly the same as all the infants I had ever known. Babies, I realized, all speak the same language.
Now that Dani is approaching the age where comprehension (receptive language) and then talking (expressive language) will soon burst forth, I am ever more excited. I am immersed in the constant, subtle and extraordinary changes in her communication skills. It is like watching someone come alive. How I wish I could record every moment of this precious process.
Amazingly, someone else has done just that. Deb Roy is clearly even more of a language geek than I am. A linguist at MIT, he had video cameras installed in every room of his house, and recorded over 90,000 hours of video, so he could study the nuances of his son’s language acquisition within the natural environment.
His team has been studying the subtle, subconscious social interactions that underlie language development. As I first explored here, Patricia Kuhl points out that social interaction is the essential ingredient for learning language. We evolved to have social development gate language learning, literally opening pathways in the brain for language acquisition. The social gating of language ensures that infants focus their resources on “speech that derives from humans in the child’s environment, rather than on signals from other sources” (Kuhl, ELA p838). Interestingly, “humans are not the only species in which social interaction plays a significant role in communication learning.” The same is true for songbirds (Kuhl, NS p 518).
Roy is discovering multiple, complex ways in which social reciprocity scaffolds language acquisition, beneath the level of our conscious awareness. For example, his team tracked the interactions between caregivers and child in the days leading up to the first utterance of a new word. They discovered that, without knowing it, all three caregivers increasingly simplified the sentences in which the word appeared. Once his son spoke it, they gradually added complexity back into its context, “subconsciously restructuring (their) language to meet him at the birth of a word and bring him gently into more complex language.” All of his primary caregivers demonstrated this pattern of interaction without discussing it with each other, without knowing that they were doing so, or even realizing that the word in question was about to be born. A remarkable demonstration of how “people are in these tight feedback loops and creating a kind of scaffolding that has not been noticed until now.” (Roy, D.)
“Milk” is the first word that Dani has mastered, and the process illustrates the complex feedback loops at work between us as she learns. When she was about 4-months old, we went through a period of disorganization with our communication in which it became difficult for me to tell when she was crying from hunger. She was tired to so much of the time, that she was often expressing both exhaustion and hunger at once, and I could not always tell which need was primary. At about 5-months, we overcame this barrier in an unexpected way, when she developed the ability to wait a bit before nursing – as long as she was in my arms. If she was hungry and I put her down, her fussing immediately escalated to wails, and I knew exactly what she needed.
At about 6-months, she expanded her communicative capacity in another way, as she began to stare expectantly at my breasts whenever she was hungry – a subtle, early cue that I could easily miss if I was not paying close enough attention. By 7-months, she solved this problem by adding movement – reaching out towards my chest. And now, at 8-months, she will shriek and dive-bomb my chest when she wants to nurse. This is not exactly my favorite way to communicate, so I have been teaching her to sign when she wants milk.
I first began showing Dani the sign for milk when she was about 5-months old. In addition to the qualities listed above – importance, stress, frequency, and simplicity – “milk” has another quality that makes it relatively easy to learn: it is “contextually focused” (Roy, B. p6). I would say the word and show her the sign while she was latched on and nursing, leaving no doubt about what it referred to.
Indeed, while Dani seems to be understanding the meaning of a handful of signs at this point, “milk” is the first word she has communicated expressively, and she began doing so at 7-months old. Not that this has eliminated her other forms of expression. Now, when she flings her shrieking self towards my chest, I enjoy beseeching her to “use your signs.” It is our first step towards using language to manage her social behavior.
I am aware of another way in which our interactions around language are creating a feedback loop that is often unconscious. I am sure it is quite obvious by now that language matters to me, a lot. Even before I became enamored of the developmental process, I was in love with words. I began dreaming of becoming a writer at six years old. I was an avid and early reader, and an early talker. All of which pleased my family immensely. To this day, my father’s favorite story is about how I endeared myself to my new first grade teacher by reading to her from Little House on the Prairie when we met.
The strive for intellectual achievement, with language development as the first milestone, has deep cultural roots in my Jewish heritage. And it continues to be the crowning accomplishment in my family of origin. Currently it is my niece who holds the badge of linguistic honor. She was a very precipitous talker, and is already sounding out written words at just three years old.
I don’t want any of these things to be coloring my behavior with Dani as we play, sing, laugh, sign, and move through the myriad other reciprocal, contingent interactions that prime her for language acquisition. But, like all ghosts in the nursery, of course they do. My excitement over her developmental process is tinged with expectancy. And perhaps for exactly that reason, she is not performing as planned.
In all other areas of development, my baby is the picture of typical. She is moving through the standard steps and meeting milestones on time, and for motor development, early. But her language development is a bit quirky. For example, the “universal” canonical babbling I mentioned above – Dani never did it. She skipped right to the variegated kind, and continues to do so almost exclusively with “g” sounds. I hardly ever hear her speak an “m” or a “b”. I have noticed other unusual things as well. She rarely ever imitates us, a game she should be well into playing at almost 9-months old. Instead, she invariably responds to our attempts to get her imitating with peels of laughter.
My wife thinks I am worrying unnecessarily about Dani’s progress with language because I know too much about what to expect. But I don’t feel worried – just aware, intrigued, and somewhat bemused. It is my first challenge in allowing her sovereignty: recognizing her for who she uniquely is, and honoring her by making room for her her unique being (Kabat-Zinn, p52). Letting her be the active, driven little person I love, who is far more interested in moving than in talking, and will forgo vocalizing all together in the days leading up to a major motor accomplishment. It is the first hint that she will have different priorities, different passions than I do. And it is the first opportunity to ensure that the shared meaning we create around her developmental process is one that allows her to live her truth and shine.
Bergelson, Elika & Daniel Swingley. “At 6-9 Months, Human Infants Know the Meanings of Many Common Nouns.” PNAS. February 2012, v109(9) pp 3253-3258.
Kabat-Zinn Myla and Jon. Everyday Blessings: The Inner Work of Mindful Parenting. Hyperion: New York, NY. 1997.
Kuhl, P. (November 7, 2014). Early Learning and the Child’s Developing Brain. Lecture conducted from IPMHPCP, Napa, CA.
Kuhl, Patricia. “Early Language Acquisition: Cracking the Speech Code.” Nature Reviews / Neuroscience. November 2004, v5 pp 831-843.
Kuhl, Patricia & Maritza Rivera-Gaxiola. “Neural Substrates of Language Acquisition.” Annual Review of Neuroscience. 2008, v31 pp 511-534.
Roy, Brandon C et al. “Relating Activity Contexts to Early Word Learning in Dense Longitudinal Data.” Proceedings of the 34th Annual Meeting of the Cognitive Science Society. Sapporo, Japan, 2012.
Roy, Deb. (March 2011) “The Birth of a Word.” Retrieved from http://www.ted.com/talks/deb_roy_the_birth_of_a_word?language=en#t-348225
Werker, Janet F. & Suzanne Curtin. “PRIMIR: A Developmental Framework of Infant Speech Processing. Language Learning and Development. 2005, v1(2) pp 197-234.