R. A. Fisher and the last fifty years of statistical methodology

Firstly may I express my appreciation of this invitation to address this conference. I feel particularly honoured at being the first to give this R. A. Fisher Memorial Lecture, perhaps partly because I have always tried to combine my profound admiration of his scientific achievements with a reluctance to be blinded by their brilliance; and it could be a matter of opinion how far such an attempt at detachment qualified me in particular to discuss his work. I understand that there was no obligation in this lecture to consider Fisher’s work, but it seemed rather odd if in this first memorial lecture one did not take the opportunity to do so; and a convenient way of doing this, in view of Fisher’s great influence on the development of statistics during the last fifty years, was to try to survey this period, with particular reference to the position Fisher occupied