Some Points in a Time

This paper offers a personal perspective on the development of language and infor-mation processing over the last half century, focusing on the use of statistical methods. Introduced, with computers, in the 1950s, these have not always been highly regarded, but were revived in the 1990s. They have proved effective in more ways than might have been expected, and encourage new thinking about what language and information processing involve. First, to say how much I appreciate the completely unexpected honour of this award, and to thank the ACL for it. I want to look at one line, or thread, in natural language processing (NLP) research: how it began, what happened to it, what it suggests we need to investigate now. I shall take some papers of my own as pegs to hang the story on, but without making any claims for particular merit in these papers. My first paper, ‘The analogy beteen MT and IR’, with Margaret Masterman and Roger Needham, was for a conference in 1958. The analogy it proclaimed was in the need for