If you're seeing this message, it means we're having trouble loading external resources on our website.

If you're behind a web filter, please make sure that the domains *.kastatic.org and *.kasandbox.org are unblocked.

Main content

Theories of the early stages of language acquisition

What is language?

Language is the primary method of human communication, but there are also other ways to communicate without the use of language. When asked to define language we tend to think of a verbal and written system in which certain sounds and symbols come together in a specific way to convey meaning. Language in its most complex form is unique to humans, although some animals have been found to have basic communication patterns. Languages often have verbal and written components, but how we classify something like American Sign Language? Animals manage to communicate — do they have language? How did language evolve? How do we learn enough language ourselves to begin to answer this question?

Why is it so surprising that we can learn language?

If you’ve ever tried to learn a new language, you know it’s not easy. There are new rules of grammar which come with many exceptions, new sounds that are hard to make, endless lists of vocabulary to commit to memory and so on. And yet, you managed to learn the basics of your very first language around the time you were two years old; no textbooks in sight.
Not only are children able to absorb the complicated rules of grammar without formal teaching, they do so from a limited vocabulary. Regardless of how much a child is spoken to, they will not hear every possible word and sentence by the time they begin speaking. Yet when they do start to talk, children begin to follow grammatical rules and apply them to form new, innovative phrases. This level of information processing is incredibly impressive in anyone, much less someone still figuring out counting and skipping!

What do we know?

As is often the case in psychology and sociology, it’s hard to get what we normally think of as data about language acquisition. It’s not a chemical we can test for or a distance we can measure (imagine asking a 2 year old how many words they know — not a particularly useful or productive task, right? ). However, there are some facts that are generally agreed upon by the scientific community. The first couple years of life are the critical period for language learning, which becomes a much harder task as people age. Children usually say their first words around 10-18 months of age, and graduate to phrases sometime before they are two years old. In fact, studies have shown that 18 month olds can tell the difference between correctly formed verb pairs (is jumping) and incorrect ones (will jumping). Somewhere between four and seven years old children begin to be able to tell stories that more or less make sense.
We also know that learning a language is not like walking up the steady increase of a ramp, but more like walking the hills and valleys of a country road. Usually when we learn a new skill, the more we practice the better we get. However, this isn’t always true in the early stages of language development. When children are first learning to talk, the verbs they use are usually the most common such as go, eat, talk, give, run, etc. These are often irregular in the past tense. Although at first they use the past tense properly (“I ran”, “he went”, etc), kids typically regress for a while. They often over-follow rules, saying phrases like “I runned” instead of “I ran”. As their vocabularies expand rapidly (known as vocabulary burst), some researchers believe children notice patterns in language, and that leads to over-correction as described in the example above. Eventually, children begin to understand where the rules apply and where they don’t, and then properly form the past tense once more. This is known as a U shaped learning curve, because the language mastery started high, dropped for a period of time, and then improved again. Thus, there appears to be a mimicking (copying) phase first and then a time of broad generalizations before children settle into language.

What are the major theories about language acquisition?

The most well-known theory about language acquisition is the nativist theory, which suggests that we are born with something in our genes that allows us to learn language. It proposes that there is a theoretical language acquisition device (LAD) somewhere in our brains that is responsible for learning a language the same way the hypothalamus is responsible for maintaining your body temperature. If language was partly biological, it could explain why humans seem to have far more complicated communication patterns than any other species .
Although no physical “language organ” exists in the brain, language acquisition can be hampered if certain parts of the brain are damaged during critical periods of language development. Damage to the left hemisphere, for example can lead to aphasia - a disorder which causes problems with language, while leaving intelligence untouched. For example, in Wernicke’s aphasia, patients with damage in a certain region of the brain can no longer understand language. Although they can still form normal sentences, neither what they say nor the words of others make any sense to them. Patients with Broca’s aphasia on the other hand, have problems forming language but no trouble understanding what is said to them. Studies have shown that young children with damage in similar regions of the brain can actually grow up with only slightly impaired language ability - implying that the brain can develop new language pathways that are good, but not quite as good as the original (Reilly, 1998).
Nativist theory also suggests that there is a universal grammar that is shared across differing languages, because this grammar is part of our genetic make-up. The majority of world languages have verbs and nouns, although this is not true in every instance, as well as similar ways to structure thoughts. Language is thought of as having a finite amount of rules from which we can build an infinite amount of phrases, and the core of these rules is somehow programmed into our brains. This is an ideal theory for explaining how young children can learn such complicated ideas so quickly, or why there are so many similarities in language around the world. This theory is comparable to how we think of numbers; regardless of cultural background, math always works the same way.
Another way to look at language learning is to treat it like learning a new skill. The learning theory of language acquisition suggests that children learn a language much like they learn to tie their shoes or how to count; through repetition and reinforcement. When babies first learn to babble, parents and guardians smile, coo, and hug them for this behavior. As they grow older, children are praised for speaking properly and corrected when they misspeak. Thus, language arises from stimuli and stimuli response. While this is logical, it fails to explain how new words or phrases come about, since children are only parroting the things they have heard from others.
The interactionist approach (sociocultural theory) combines ideas from sociology and biology to explain how language is developed. According to this theory, children learn language out of a desire to communicate with the world around them. Language emerges from, and is dependent upon, social interaction. The Interactionist approach claims that if our language ability develops out of a desire to communicate, then language is dependent upon whom we want to communicate with. This means the environment you grow up in will heavily affect how well and how quickly you learn to talk. For example, infants being raised by only their mother are more likely to learn the word “mama”, and less likely to develop “dada”. Among the first words we learn are ways to demand attention or food. If you’ve ever tried to learn a new language, you may recognize this theory’s influence. Language classes often teach commonly used vocabulary and phrases first, and then focus on building conversations rather than simple rote memorization. Even when we expand our vocabularies in our native language, we remember the words we use the most.
It’s important to keep in mind that theories of language acquisition are just ideas created by researchers to explain their observations. How accurate these theories are to the real world is debatable. Language acquisition is a complicated process influenced by the genetics of an individual as well as the environment they live in.

How do scientists today study language learning?

Many of these theories initially came about as a result of what is called “armchair psychology”; that is, sitting and thinking about a problem. It is extremely difficult to collect objective and accurate data on what’s going on in the brain in terms of its direct relationship to a behavior such as language. That said, some computational models of language acquisition have been gaining traction in the past several decades. A computational model is a mathematical way to recreate complicated systems we see everyday; from how water flows in a river, to how children learn languages. The model is built to represent the way we think something happens. For example, in the model of the learning theory approach, a word would be learned faster if it came up a lot or the subject received a lot of input about it. Then, linguists change how different variables work to see what affect that would have on the system. If the model behaves and “learns” the same way that we do, it’s a good sign that the model is on the right track. These models have helped to identify and measure linguistical features such as the critical period for language learning, the vocabulary burst, and the U-shaped learning mentioned earlier.
New brain imaging technology, such as MRIs and fMRIs have also allowed scientists to look at the brains of children and patients with language-acquisition disorders to understand this complicated event. An fMRI can track where and when our brains use energy. If a certain part of your brain lights up while you’re learning a language, that part of your brain is using energy, and in this context might be related to language-acquisition. Of course we learn over time and not all at once, so there is a limit to what we can learn via imaging which represents the brain in a single moment.
While we still have a ways to go before we completely understand how we learn a language, we definitely know enough to know that it’s a pretty incredible feat. So give yourself a well-deserved pat on the back and just remember that the phrase “it’s so easy, a child could do it!” doesn’t always apply.

Want to join the conversation?