The igloo paradox

In the previous post, it was mentioned that a vocabulary of 3500 words is considered sufficient to pass the Cambridge FCE. This irresistibly raises the question, which 3500 words are they?

Unfortunately, the answer isn't so straightforward. While it is claimed that the Cambridge Vocabulary for First Certificate "covers all the vocabulary First Certificate candidates need", clearly it doesn't cover all the basic vocabulary that candidates will be assumed to have acquired in their first years of learning, and it doesn't actually contain 3500 vocabulary items anyway.

One sensible approach to identifying the words that learners need is that taken by Michael Lewis and other advocates of the Lexical Approach. Building on the research of early corpus linguists which showed that 70% of our everyday conversation is made up of just 700 words, they designed a course that set out to teach those words in the first year. 

There are some practical problems with this approach, particularly because most of these very frequent words have several different meanings, so it isn't a simple matter to teach or to learn words like "get" or "take".

Nonetheless, teaching materials, particularly reference works such as dictionaries and grammar books, are informed by the principle that the most frequently used vocabulary items and grammatical structures are the most useful ones for learners.

And accessing the tools of corpus linguistics through a simple website searchbox, we can very easily discover which the most frequent items are. Conversely, for any given item, we can find how frequent it is. For example we find that the word "igloo" is ranked as the 26,896th most frequent word in English, compared to:

ice - 1234th

island - 1517th
iron (n) - 2294th

That would seem to indicate that if we're looking for a useful word to represent the letter "I" when teaching the alphabet, "igloo" wouldn't be a good candidate.

Of course, the relative frequency of a particular word depends on whose English is included in the corpora from which the frequency counts are derived. Most corpora are based on adult language. While analysing such corpora can accurately predict which words will be useful for adults, these aren't necessarily the same words that children need.

Looking at lists designed specifically for teachers of children, such as the Fry word lists, we find that there are several function words beginning with "I" in the 100 most frequent, including "in", "is", "it", "I", "if" and "into". The most frequent concrete noun beginning with "I" is again "ice", which occurs within the most frequent 700 words for children, whereas "igloo" doesn't occur in the lists covering the 1000 most frequent words.

This is hardly surprising. Imagine a Greek learner of English working as a waiter in a cafeteria, or using English while travelling abroad. Offering or requesting ice cream, or ice for drinks, would be a  pretty useful thing to be able to do, more useful than being able to ask for an igloo.

This is a word which most people are unlikely ever to use, and yet in many schools of English, it's one of the first words taught. Most authors and publishers of coursebooks are doing a great job, but we still owe it to ourselves and our students to reflect on the materials that we use and adapt them in cases where the contents aren't useful. A small but significant first step would be to consider the words that are used to teach the alphabet. And if you're using one of the many courses that include the word "igloo", replace it with "ice" or "ice cream", which are twenty times more frequent and twenty times more useful.