How do we learn vocabulary?

It’s something that linguists, neurologists, speech therapists, programmers and language learners have been researching for ages. Just how do we learn a language and, in particular, vocabulary?

We learn our mother tongue, naturally, don’t we? Babies somehow just pick up voices around them and over the course of their development form them into language. In fact, most of what we say to babies is gibberish: “goo goo gah gah, baby, look at how poochipocchi you are, little coochichoo!” And even as toddlers, no one ever really explains the grammar and meaning of the words we hear, yet somehow we have the ability to understand and speak within a couple of years, and then move swiftly on towards using advanced vocabulary and developing our own personal lexicon.

So, just how does this ‘magic’ happen, and how can it be recreated when we learn a new language?

Is grammar really a skeleton we hang words on?

For years, in education, there’s been this idea that grammar is a kind of skeleton that holds up a sentence, and we put English vocabulary - or lexis - into appropriate spaces to form sentences. This has been the traditional tacit view of English language learning and as a result many courses and books teach grammar and vocabulary separately. You are given long lists of words, synonyms and verbs to learn, maybe in their infinitive form, and perhaps with translations to help you. Maybe your teacher has set quizzes for you: translate these words into English using an English dictionary? Or you’ve very studiously made long, long, pages-long lists of words to learn yourself, and only managed to learn the first ten!

But, let’s stop and think about how efficient this really is. If we actually stored language in our brain as lists, and you had to find the word you wanted to use, your brain would have to read through the entire list until you found it. You might be lucky and find it in the middle of the first page but if it was on page 63 of 100, conversation would be very slow indeed. And native speakers would be even slower at speaking, because they would have knowledge of even more words than a non-native, and so would have to search through an even longer list.

Conversations would grind to a snail’s pace while we searched for the word we wanted: communication would be painfully slow indeed, and emojis would be far more suitable!

How native speakers actually store words

The thing is, native speakers don’t learn and store words individually in their heads like that. They don’t pick a word from list a), then a word from list b), then add it to a word from list c) to create a phrase. That would be so inefficient. Instead, they begin by storing groups of words in context, and then using this knowledge to predict where they may use them again.

What do I mean? Well, let’s look at a (yummy) word like cake. Here are a few uses:

• That shop sells beautiful cakes.
• Hold on, I’m in the middle of baking a cake.
• What kind of cake do you want to eat?
• He scoffed the whole cake in two minutes flat.
• Would you like a piece of cake?
• How was the test? Yeah, a piece of cake.
• Can I order a birthday cake, please?
• To make ten potato cakes, you need…
• You can’t have your cake and eat it.
• That was just the icing on the cake.

All of these uses of the cake are different, but a native speaker will have access to all of these, and more. That’s because it’s now assumed that when native speakers come across a new use of the word cake, they’ll store it in context in their heads, along with all the words that went with it, and make a subconscious note of how it was used.

Was it said jokingly in a conversation with friends, or written in a formal report? Was it said in a happy or angry way? Bit by bit, native speakers start to build up collections of phrases around the word cake, and not only that, their brains then start to predict in which context they may hear the word again. This process of subconscious noting, is known as lexical priming, and was first discussed by Professor Michael Hoey in a ground-breaking work that has changed the way linguists think.

How can language learners do the same?

So how can language learners come anywhere close to achieving this native-like ability and improve their vocbaulary?

The first thing is: ditch the single word lists. Yes, that’s right, you heard me. Get rid of the long lists of individual words with their translations.

Instead, keep a record of chunks of language, phrases, and expressions around the word cake in context. If you’ve heard the word cake used with a verb, make a note of not only which one, but in which form. For example, it’s much more common to hear “he loves making cakes” (gerund + cake) than “he makes cakes every week” (present simple + cake). The second one is not grammatically wrong, it’s just not something we would say that often! So, don’t just write down to make a cake, but making a cake. Think about what other words were used in the same phrase as well. Then, make a note of where you found it, what the context was, and what the whole phrase meant.

In summary think about:

• the words that come either side of your target word,
• the grammatical patterns it occurs in,
• the meanings it’s associated with,
• what the context was - polite, hostile, formal, colloquial, used by someone older or younger, etc.

Lexical priming in action

Once you build up these chunks of language, your brain will then start predicting when you can use a word in context.

If you ask a native English speaker to finish this sentence, “a light Madeira sponge”, they will say “cake”, because they know the words “Madeira sponge” are more likely to be followed by the word “cake” than, for example, the word “biscuit”.

And that’s what tech is progressing towards doing right now. Just think about predictive text on your phones and emails which finishes sentences for you after you’ve typed in just a couple of words.

So, get efficient, get yourself thinking about groups of words, and get yourself lexically priming!