It is right that statements by scientists that touch on ethical issues be scrutinised for departures from scientific objectivity. It is also sensible to be wary when scientific methods such as quantification are used to lend validity to everything from league tables to form-filling. But to a great extent, the mistrust and misapprehension of science's activities are not really its fault.
Decisions taken in the public domain are not those of scientists, but of politicians and industrialists. But when these groups claim that they acted according to scientific advice, the public tend to believe that science is an accomplice to the crime.
Nevertheless, on science's wilder fringes, the existence of a quasi-religious element - seen in a tendency to reductionism or expansionism - does not help its image. Reductionism is strongest among some geneticists and socio-biologists who see the gene as the atom that determines our humanity and social culture. The expansionist tendency is to be found among physicists, particularly particle physicists, cosmologists and quantum theorists, and is either some sort of mathematical theology or some sort of hyper-physics, or maybe a rather esoteric art form. Whatever, it is so far from having that simplistic view of humanity of the reductionist theory as to have no view of humanity at all.
A common belief is that science will, given time, explain and describe everything, including us as individuals, rendering the humanities obsolete. But I believe it will yield results of only limited interest in fields such as psychology, economics, politics and sociology and of no interest to ethics, aesthetics and religion. In short, it is limited by the very method it has found so blindingly successful in its study of the material world, the essential features of which include reductionism, reproducibility, quantification and self-referentiality - the experimental limitation of having to use one bit of the universe to extract information about another.
Reductionism involves the system under investigation being separated as far as possible from the rest of the universe and then analysed into its component parts. Holistic features are often lost as a result. The need for reproducibility means science can have nothing to say about the unique. This makes a study of individual flair, talent, genius and idiocy, impossible. Quantification involves the mathematics of probability and statistics, thus introducing the ideas of chance and random behaviour - useful for explaining observable properties, but not for understanding intentional behaviour or religious belief.
In the 17th century, science grew out of a world seen as magical, a world with its own "theory of everything" in which everything was connected to everything else via occult sympathies and correspondences.
In the hands of people such as Gassendi and Galileo, science took over from magic. But the limits of science meant that not all of what was seen as magical disappeared. Much of what we now call natural magic - that is, the influence on us of music, art, poetry, personality, ambience, landscape, the night sky - retained its power and still informs the arts and humanities.
The phenomenon of consciousness, which we all directly experience, is perhaps the most mysterious and inexplicable thing of all. It accommodates mathematics and Mozart and sees science end where Schubert begins. It needs science to talk about public experiences; it needs literature and the humanities to talk about personal experiences; it needs science and the arts to increase public knowledge. The cultural world has science and art as complementary aspects. This fundamental complementarity should never be forgotten.
Brian Ridley is a fellow of the Royal Society. His book, On Science , published by Routledge, £7.99, explores in more depth the issues raised here.