Poll results – How do children learn language?

A month ago, as part of a discussion about linguistics and language acquisition, I asked my readers what they think about how children learn language. You can read that post and view the poll by clicking on the link here: We Need to Talk About Language.

Language Acquisition wordle. Created by Grace @ Cultural Life using wordle.net

Language Acquisition jargon. Created by Grace @ Cultural Life using wordle.net

In the poll I asked the following question: “How do children learn language?” It seems simple, doesn’t it? But there are no simple, straightforward answers. In order to give you some background information before I discuss the answers of the poll, I’ll outline three main approaches to language acquisition, with reference to another linguistics post I wrote: The Language Instinct. I wrote about the behaviorist and nativist theories at greater length in that post if you would like to read a more detailed explanation.

1. Behaviorist theory = based on Skinner’s experiments in the 1950s where rats learned to press a lever when they received positive reinforcement. Skinner said that native language acquisition is based on a system of imitation and reward.

2. Nativist theory = the ground-breaking linguist, Noam Chomsky, proposed that we are born with an innate ‘Language Acquisition Device’. A key part of the LAD is Universal Grammar: the concept that “children arrive in the world with grammatical principles wired into their brains” (quoted from my previous post which contains a more detailed summary of Universal Grammar).

3. Constructivist theory = as its name suggests, constructivist theory hypothesizes that children learn the grammar and syntax of their native language by acquiring a set of constructions, e.g. nouns, pronouns, verbs etc, based on the input around them (note that it is not the same as imitation). These components of language can then be formed into sentences. The constructivist theory does not agree with the concept of an innate language device.

It was very interesting to see the outcome of the poll.


41% of you chose the option that children begin by imitating the language they hear around them. While it is true that imitation plays some part in language acquisition, the exact nature of it is disputed. As I wrote in a previous post, “this argument for how children acquire language has many flaws. Firstly, if children learn how to produce their language solely as a result of [imitating others], their lexicon would be extremely limited”. The book, Language Acquisition, by Jill and Peter De Villiers explains that “The child…needs to extract the rules of the language in order to produce sentences appropriate to his changing situation” (De Villiers & De Villiers, 1972:199). Therefore, language acquisition is much more than simple imitation.

The second most popular option, with 35% of the vote, was the constructivist approach. The least popular option, at 25%, was the theory that we are born with innate linguistic principles. There is a lot of discussion and debate about these two theories. There aren’t any conclusive answers because each theory has advantages and disadvantages and it is very hard to disprove either theory for definite. I wonder if we will ever find definitive evidence on how we acquire what is arguably the most important component of our daily lives.

Thank you to everyone who voted in the poll. I hope you have enjoyed the linguistics posts I published here during the past couple of months. Let me know if you would be interested in more linguistics posts (but not about language acquisition – I think I’ve said enough on that topic for now) here on the blog.


De Villiers, J.G. and De Villiers, P.A. (1972). Language Acquisition. Harvard College: United States of America.

9 thoughts on “Poll results – How do children learn language?

  1. Isn’t it true that babies in different parts of the world make different baby ‘sounds’ depending on where they’re from? I.e. a German baby’s babyspeak sounds different from an American etc. and isn’t that because Americans are trying to teach the words ‘mama’ and ‘dada’ and German parents are trying to teach them ‘mutti’ und ‘vatti’? If there was a universal grammar or language for babies, wouldn’t they all sound the same?


    • Not quite…. The concept of Universal Grammar (UG) doesn’t state that lexical items (i.e. the words of each language) are innate. That would be impossible.

      UG does, however, theorize that we are born with a set of general with grammatical principles wired into our brains, e.g. categories such as N (noun) and V (verb) and phrases such as NP (noun phrase) and VP (verb phrase).

      Therefore, because children are born with these principles, they simply have to note which setting their native language uses. Then, “Huge chunks of grammar are then available to the child, all at once, as if the child were merely flipping a switch to one of two possible positions” (Pinker S. The Language Instinct. London: Penguin Books Ltd; 1995. p. 112)

      Does that clarify it? It took me a while to get my head around UG and what it actually meant because it is quite abstract!

      Early babbling is very similar across languages but as babies develop, it is possible to see some influence of the native language in terms of intonation and rhythm. Around the age of 6 months, the babbling stage starts. From the approx. ages of 6 – 12 months babies from every nationality start out by producing consonant-vowel (CVCV) sounds such as ‘ba ba’, ‘pa pa’ and ‘ma ma’.

      Many languages use words such as ‘mama’ (or similar) to refer to the mother figure but this came about due to children’s early babbling because consonant sounds which are formed using the lips, such as /p/ and /m/, are the easiest consonants for babies to produce.

      So, babies produce these babbling ‘mama’ and ‘papa’ sounds without any meaning attached to them. But “Parents are quick to help a child assign meaning to these early noises, decreeing that mama means ‘mother’ and papa or dada means ‘father'” (O’Grady, W. How Children Learn Language. UK: Cambridge University Press; 2005. p. 7). It is striking that the majority of languages use /m/ sounds to refer to ‘mother’ and /p/ or /d/ to refer to ‘father’. Sometimes this can even be the other way round: O’Grady’s book states that the Georgian language uses ‘mama’ to refer to ‘father’.

      Thanks for your comment. 🙂


      • Thank you for that very thorough response! I’ve given myself a week to finish it! No it really is interesting. So now I can tell my wife that my daughter’s first words ( which were mama) was actually for me…in Georgian times at least. Actually I trained her to say ‘mama’ so when she woke up during the night she would call for her mama, instead of me! Not very scientific but still a very clever idea I thought.


  2. So interesting to read this and find out how others think on the topic. When I took a college course about Neurolinguistics (it was not my field – I was just curious and it fit the schedule) many years ago, the class was very heavily weighted to nativists – I think it was the “popular” theory at the time. I wonder how much these beliefs shift with time/education/locale. Anyway, I like the linguistics posts. Keep ’em coming.


    • Thank you for reading and leaving a comment. I think one of the most interesting things about language acquisition is that there isn’t a concrete answer. Sometimes that is fascinating and at other times, frustrating! It seems as though the nativist and constructivist theories go in and out of “fashion”.

      I’m glad you’re enjoying the language-themed posts. 🙂


  3. This is absolutely uncharted terrain for me..and to be honest, I have never felt the need to think about how we learn languages! But, your post is quite interesting…and I tried to make sense of what you are trying to convey( it’s a little complicated for me), but I have couple of queries:

    1) Is it not remotely possible that maybe, we learn languages due to a combination of all the 3 factors you mentioned( imitation, constructivist and innate)…After all, we do learn the nuances of our mother tongue from the people around us. But, there definitely has to be something innate, which separates us from other species of living beings. Also, constructivist and Innate seem somewhat inter-related to me. Am I assuming too much?

    2) I once read or heard somewhere that across languages the word for mother is similar..and that apparently comes from the sound a newborn baby makes while suckling. Is this theory correct?


    • I apologize if this post is too in-depth. One of my previous posts which I linked to in this post explains things clearly in a lot more detail so I guess I assumed that readers would view that post before reading this one. It took me a while to fully understand the concepts and theories which linguists talk about in terms of language acquisition…they are certainly quite complex!

      It is certainly possible, indeed, probable, that language acquisition is based on a combination of factors. Out of the three main models, there isn’t a single one which encompasses and explains everything. It would be lovely to have a model which takes an integrative approach.

      Hmm, I wouldn’t really view constructivist and innate as inter-related. While they both view input (the language we hear around us) as one of the key factors in learning a native language, they differ in quite a few ways and are generally regarded as opposing viewpoints.

      For nativist linguists, they theorize that Universal Grammar (the idea that languages share universal grammatical properties which we are born with) is the underlying factor behind language acquisition. On the opposite side, constructivist linguists do not believe that grammatical principles are innate. Constructivists theorize that language develops solely from cognitive processes in the brain but nativists think that it is shaped by innate principles (i.e. ones which we are born with).

      In answer to 2), that theory is almost correct but the use of ‘mama’ to mean ‘mother’ in many languages is actually from the babbling sounds which babies make when they are around the ages of 6 – 12 months. All babies, of every nationality, start out by making sounds like ‘ma ma’ and ‘pa pa’. The simple reason for that is because consonant sounds which use the lips are the easiest consonants for babies to produce.

      As I wrote in a reply to an earlier comment, ““Parents are quick to help a child assign meaning to these early noises, decreeing that mama means ‘mother’ and papa or dada means ‘father’” (O’Grady, W. How Children Learn Language. UK: Cambridge University Press; 2005. p. 7). It is striking that the majority of languages use /m/ sounds to refer to ‘mother’ and /p/ or /d/ to refer to ‘father’. Sometimes this can even be the other way round: O’Grady’s book states that the Georgian language uses ‘mama’ to refer to ‘father”!

      Thanks for reading and taking the time to leave a comment. 🙂


      • Hey! Thank you so much for taking the trouble to provide such a detailed explanation! And you have nothing to apologize about!! I really loved the post…but, because I read about these theories for the first time..it got jumbled up inside my peanut sized brain…sigh…it’s tough being a know-all 😀


Leave a comment and share your thoughts....

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s