Failure

This is the sequel to the author's book Ignorance, and builds on some of its insights. The main thrust of the book will be familiar to historians of science (Firestein spent a year at the history and philosophy of science department in Cambridge): that the public perception of science as a rule-based and fact-accumulating system inexorably advancing its frontiers of knowledge is incorrect. The real picture, as Gerald Holton showed with his distinction between S1 and S2 (science as pursued and science as written up) is much messier: failure is a stepping stone to eventual success, although one should remember that scientific knowledge is always incomplete and that some of our current theories will prove mistaken – many brilliant scientists of the past held views that we now know to be wrong. Apparently Sir Alan Hodgkin always spent more time with his students who came up against problems, as this might enable him to come up with better questions and experiments. The fifteen chapters range across defining failure, the scientific basis and method of failure, the integrity of failure (negative results are an opportunity for honesty), medical aspects and pluralism.

One of the author's messages, echoing Samuel Beckett, is that it is good to learn how to fail better. Failures give us useful feedback – think of Edison's lightbulb. He uses some intriguing historical examples such as Harvey's discovery of the circulation of the blood in 1612 (even Leonardo – not mentioned here – missed it). He withheld publication until 1628 for fear of publicly disputing the ideas of Galen. At this point the author could have discussed more explicitly the role of authority in science, embedded as it is in science as a social and hierarchical institution. His contention that science is about what we don't know highlights the contribution of Sir Karl Popper and Firestein relates a telling anecdote when 17-year old Popper meets Alfred Adler and witnesses him rationalising an incongruous case to fit it in with his preconceived theory: Adler – and Freud – were not open to falsifiability. The author is not keen on hypotheses and models, preferring instead to place his emphasis on the search for explanation. Towards the end of the book he draws on Sir Isaiah Berlin's essay on the hedgehog and the fox (the hedgehog knows one big thing while the fox knows many) to argue against monism in science and for pluralism. This is a very useful discussion as it allows more flexibility into scientific thinking. The book is very engaging and witty at times; if your experiment is going badly, the author recommends you go for a coffee if it is still too early for a beer.