The book I want to discuss here isn't terribly old, having been published in 1982. But it's a book that might not come to your attention in the array of more recent work on the philosophy of science. The book I'm referring to is Peter Medawar's Pluto's Republic (Oxford Univ. Pr., 1982). No, that's not a typographical error in the title; Medawar is referring here to the "intellectual underworld." He states that the most prominent citizens of this Republic include IQ psychologists, psychoanalysts, practitioners of scientism, and mystical theologians, but this book is much more that just Medawar airing his pet peeves. By carefully explaining why none of these individuals are really scientists, he comes up with a rather clear picture of what science is. The book also contains a number of rigorous yet readable essays on the philosophy of science, including the whole of Medawar's earlier work The Art of the Soluble and three lectures on Induction and Intuition in Scientific Thought.

For me, these lectures, along with the essay on "Hypothesis and Imagination," were the most rewarding parts of the book. Though many philosophers of science would probably find fault with some of Medawar's discussion, I think his description still gives a nonscientist a clearer picture of what it is to do science than is to be found in most works on the subject. He argues that there are two conceptions of science. The first view holds that science is an imaginative and exploratory activity, "a great intellectual adventure with intuition as the mainspring of the advancement of learning," while the other view sees science as a critical and analytical activity with evidence required.

Medawar sees both these views as partial and therefore defective. Science involves both perspectives. In most cases, a scientist will go back and forth between imaginative and critical work many times in the course of an investigation: "Scientific reasoning is a dialogue between the possible and the actual, an interplay between hypotheses and the logical expectations they give rise to: there is a restless to-and-fro motion of thought, the formulation and rectification of hypotheses."

Medawar is a Popperian, so he stresses falsification as the key process in scientific inquiry and downplays the role of induction. He argues that inductivism doesn't give an adequate account of scientific fallibility and fails to explain how the very same processes of thought which lead towards the truth also lead us so very much more often into error. In other words, inductivism doesn't explain why mistakes are so common in scientific inquiry. If science is a matter of collecting information and developing an hypothesis on the basis of that information, then most hypotheses should be correct. Yet that is hardly the case. Medawar notes that "nearly all scientific research leads nowhere." He admits that four-fifths of his time in science was wasted--and this from someone who managed to win a Nobel Prize.

This odd situtation, which inductivism cannot adequately explain, the hypothetico-deductive method can. Medawar describes this method as the development of ideas, of hypotheses, followed by the testing of those hypotheses. In this view of scientific inquiry, as opposed to the inductivist view, the idea, not the data, comes first. Data cannot be gathered until there is an hypothesis to guide data collection. This makes sense in terms of recent research on how science is done, though some observers would say that Medawar's view is as dated and inaccurate as the inductivist one because it doesn't go far enough toward constructivism [the view that we impose, perhaps, our views, including our prejudices, on the world].

But whether or not you agree with Medawar, his writing is worth reading. He writes well, and presents a wealth of interesting ideas. Even the essays which are not directly involved with the philosophy of science are valuable. Included in Pluto's Repubic are a number of book reviews. It's interesting to read from our more detached perspective what Medawar had to say about The Double Helix and The Phenomenon of Man [a best-seller by biologist-mystic, Pierre Teilhard de Chardin]. And his postscript to the biography of D'Arcy Thompson written by Thompson's daughter is a beautiful mix of homage and criticism. By the way, D'Arcy Wentworth Thompson, the scholar-naturalist, 1860-1948, by Ruth D'Arcy Thompson (Oxford Univ. Pr., 1958) is a rather uncritical but nonetheless interesting account of Thompson's life. It reveals the depth of his interest in and knowledge of not only biology, but mathematics and the classics. And as medawar argues, it was probably only because of this three-sided background that Thompson was able to produce such a classic as On Growth and Form (Cambridge Univ. Pr., 1942).

Medawar is always a joy to read, and Pluto's Republic is probably his best book in that it is not his most substantial one, and much more rewarding that his rather thin autobiography Memoir of a Thinking Radish (Oxford Univ. Pr., 1986). Medawar obviously had a penchant for snappy titles, and I highly recommend Pluto's Republic despite his description of how he came upon the title: "A good many years ago a neighbor whose sex chivalry forbids me to disclose exclaimed upon learning of my interest in philosophy: 'Don't you just adore Pluto's Republic?' "

--Maura Flannery