A clear demonstration of the egocentric nature of English speech emerges from a novel experiment at Lancaster University which recently recorded ordinary adult conversations for a week. The research found that "I" followed by "you" were the most commonly spoken words. In at number three came "it".
Geoffrey Leech, director of the unit for computer research and the English language, has measured the frequency with which words are used in everyday speech. Using a corpus of more than four million words from ordinary conversations at home and at work, Professor Leech discovered that "he" is used 50 per cent more frequently than "she" - a gender imbalance which has improved considerably since the 1960s, when a written corpus collected at Lancaster showed more than twice as many "he"s as "she"s.
A sample of adults were lent a personal tape recorder for a week to record whatever conversations they had and the results show that female words exceed their male counterparts when referring to the older generations with mum, mummy and mother winning over dad, daddy and father.
"Familiar terms like bloke and guy are also common," Professor Leech said. "The British bloke is holding out against the American guy although guy is making inroads in this country and is already crowding out the native chap."
The lists are derived using a computer programme developed by Paul Rayson in the university's computing department from the British National Corpus, a 100-million-word English language database. This is the result of a collaboration between Lancaster, Oxford University and the British Library.
Professor Leech found that the frequency of swear words was perhaps less than expected. "Those who feared that they are the most frequent words in English speech will be somewhat relieved to find that "bloody" - the most common one - is 167th on the list."
More than half the people surveyed were women but the six most common names mentioned in the corpus are male (John, Paul, David, Richard, Michael and Mark). The most common female names (Ann, Jean, Charlotte, Mary, Emma, Sarah) occur less frequently. Why?
"The frequency list is only a first step," said Professor Leech. "The corpus is ready to be analysed and many fascinating research projects are waiting to be born."