Foreword
by Professor Michael Hoey, Chief Adviser

Corpora

Concentrating on the rare words was not my only childhood mistake when I attempted to create my own dictionary. A more serious one was that I arrived at my definitions by looking at the definitions of other dictionaries. I reworded them of course – even at the tender age of 12, I was intuitively aware of the dangers of plagiarism – but I saw my role as one of collating the wisdom of previous lexicographers. Of course that way there is room for little new wisdom.

Perhaps shockingly, until 20 years ago my practice would not have been out of place in many dictionary teams. Lexicographers would draw on a mixture of previous practice, intuitions, and half-remembered examples, supported by chance encounters with the word in print. With the advent of large corpora and the development of powerful computer software capable of exploring those corpora, dictionary-making has changed beyond all recognition. The lexicographers who worked on the Macmillan English Dictionary had the opportunity of examining hundreds and in some cases thousands of instances of a word in use. From these instances they could work out what a word really meant in contemporary English, rather than what it was supposed to mean.

Take the example I gave above of the use of conventional in the phrase conventional oven. It may seem obvious that a conventional oven is one that is not a microwave oven, but it only seems obvious once it has been pointed out. If the lexicographers who worked on this dictionary had relied on intuition, they might easily have forgotten this use of the word, and of course if they had relied on previous dictionaries they could easily have missed it because many of those dictionaries were prepared before the microwave oven came into popular use.