In this talk, Leo Selivan challenged the conventional approach of choosing what vocabulary to teach using frequency lists. He suggested that frequency has become so ubiquitous that it’s now included in dictionary entries and that we use these frequency lists for a number of reasons:
- To grade and select vocabulary
- 80% of English text consists of high frequency words
- Frequency lists are used for defining vocabularies in learner dictionaries (core)
- To counter the teacher’s intuition which may be wrong. For example, we overestimate the frequency of ‘blond’ and underestimate ‘arise’
The most frequently used words tend to be function or grammar words and it’s commonly believed that we can “learn 2000 words and then it’s all plain sailing.”
Leo argues that frequency is not the same as usefulness and analyses this through several areas of lexis.
Polysemy is a word with multiple meanings such as ‘aid’ (come to my aid, foreign aid, hearing aid). Words that have polysemous relationships can be close or distant in meaning. In the following examples, rough is closer relative to run which is more distant.
(1a) He runs 10 miles every day
(1b) She runs a restaurant
(2a) His hands were rough from hard work
(2b) I’ve had a rough day today
Leo asked the attendees to translate several words into another language that they are familiar with: accident, (to) join, condition, to gain, business (I realised that I don’t know how to say condition in Hindi and I had to look it up in a dictionary). The actual learning from this activity is that the translation depends on the meaning we want: have an accident vs. by accident, join the club vs. join the army etc. This might in fact be an interesting activity for the multilingual classroom.
He then contrasted two schools of thought – Charles Rhul’s On Monosemy establishes the first one where there is a core primary meaning and all other words are seen as derived. The other school of thought which he subscribes to is embodied by John Sinclair’s Corpus Concordance Collocation which holds the view that meaning is established by collocation and that to truly understand what a word means, you have to look at co-text (e.g., by accident). Therefore he argues that advice focusing on meaning before looking at collocation is ‘dubious’.
With some words, it’s easy to find other members of the word family because they follow specific patterns for example the suffix -ment in development and improvement signal nouns. However, this is not as obvious in other words such as the following:
name – namely
price – priceless
fish – fishy
parent – parenthood
It’s entirely possible that learners encounter the derivative form before the base form for instance, suddenly before sudden, crazy before craze, reveal before revelation, computer before compute and conventional before convene.
Some words which are low frequency have a greater likelihood of being taught because of the context in which they may be used.
- Classroom language: verb, vocabulary
- Cultural terms: candle, mosque
- Conceptual difficulty: admit, issue
- Ease of learning: guitar, basketball
- Elementary school needs: porridge, knight, wand
- Personal use: brunch, sociable
I’m not sure what he said with respect to conceptual difficulty – perhaps he was contrasting it with ease of learning – that we are more likely to teach words with a lower frequency if they are easier to learn. Apropos personal use, these two examples are from Leo’s own personal repertoire. He was trying to suggest that learners might find low frequency words useful because of their personalities and preferences.
With multi-part verbs such as ‘the plane took off’ and ‘stand by your friend’ and non-compositional chunks such as ‘take place’, meaning can’t be understood from individual elements. Some items need to be learnt as chunks for example ‘at your disposal’ and ‘to some extent’. The PHRASE and PHaVE lists are apparently two new lists designed to address this.
Called disponibilité in French, this refers to words that easily spring to mind when presented with a prompt. So if you’re asked to come up with words related to travelling, many of the words that you may generate are lexically available but probably low frequency such as luggage.
The bottom line is that frequency doesn’t translate to usefulness and vocabulary acquisition is incremental and non-linear. You can download his handout with the meta-language he uses from this link.