Big changes in the way science is practised is leading to a new framework for the production of knowledge that could mean a decline in scientific objectivity, according to physicist and science policy observer John Ziman.

Giving the Royal Society's Medawar Lecture last week, Professor Ziman said that there have been so many changes in the way science is practised over the course of the past few decades that a new research culture - "postacademic science" - is now emerging.

Some changes reflect scientific and technological progress while others are caused by governments taking the view that science, having grown steadily for centuries, has now become too large and expensive to go its own way. The result, he argues, is that teams, electronic networks, interdisciplinary centres and technology foresight exercises are replacing individuals, learned journals, university departments and inspired hunches.

Professor Ziman argued that in postacademic science, research is directed towards specific ends, and relates to problems which will not have been set by individuals themselves or even by the teams. The choice of problems to be worked on operates at the level of the bodies funding science: "All policy talk about foresight, priorities, accountability etcetera is really focused on 'problem choice'."

³ÉÈËVRÊÓƵ

Professor Ziman, emeritus professor of physics at Bristol University, says that while this may mean that the science that gets done is "better" and more "relevant" than if left to the judgements of individual scientists, "it also means that a few wild conjectures never get the chance to show their hidden capabilities, which are just occasionally revolutionary".

Professor Ziman highlighted a scenario for the future of science compiled by six distinguished metascientists led by Michael Gibbons of the Science Policy Research Unit at the University of Sussex.

³ÉÈËVRÊÓƵ

In The New Production of Knowledge, they suggest that most research work will eventually relate to problems arising in the context of application that is determined by corporate and government bodies. Another member of the team was Peter Scott, professor of education at the University of Leeds and former editor of The THES.