Peter Medawar’s “The Limits of Science”

W.J. Astore

Note to reader: I wrote this back in 1988 when I was a first-year graduate student in the history of science at Johns Hopkins University.  I took my first graduate seminar with Owen Hannaway, a distinguished professor of early modern science and alchemy.  He asked us to do a book review, and I chose Peter Medawar’s The Limits of Science.  I dedicate this article to the memory of Owen Hannaway (1939-2006), a distinguished scholar and a gallant man.

The Limits of Science is an intentionally short book dealing with topics in the history and philosophy of science. It consists of three different essays written in three different styles, yet it yields a general outlook on science which can be nicely summarized.  Sir Peter sees science as the most successful of man’s enterprises, but he is quick to observe that science has limits, although the growth of science itself is not self-limited.

Medawar first defines science.  Science, he says, is not a mere collection of facts but organized knowledge, knowledge that can be used to predict the behavior of the sensible world.  Medawar is careful to emphasize the difficulty of obtaining scientific knowledge, and the need for confidence based on trust within the scientific community.

Medawar then discusses whether there is such a thing as the scientific method and traces the development of different approaches.  Before the Renaissance, deduction in the form of the Aristotelian syllogism was used to advance science, while intuition and revelation were used to support science.  For philosophers in the Middle Ages, divine revelation guaranteed absolute certainty.  Francis Bacon lit a new path for enlightenment in the late sixteenth and early seventeenth centuries through the use of induction.  Bacon’s new method was the development of general premises through the use of experimentation and the collection of observations.  The frontispiece of Bacon’s Novum Organum summed up the new ideal of Plus Ultra (more beyond): it depicted the pillars of Hercules with a biblical inscription (Daniel 12:4) prophesying the advancement of knowledge.

Medawar next examines deduction and induction and finds them lacking.  The chief difficulty with deduction is that it begs the question; it can only discover something already contained in the major premise, therefore it is not a way to new knowledge.  By comparison, a major premise arrived at through induction cannot contain more information than the sum of its known instances.  A theory consisting of a legion of facts summarized by an iterative inductive process can thus be overthrown by a solitary contradictory instance.  In sum, a deductive premise merely makes explicit information that is already present in the premise, while an inductive premise is no better than the sum of its parts.  Neither method leads to new knowledge.

Considering these arguments, Medawar sides with the conclusion of Bertrand Russell and Karl Popper that there is no scientific method.  The myth of induction as the method for scientific advancement, developed by John Stuart Mill and Karl Pearson in the nineteenth century, persists today mainly because it agrees best with the public’s conception of science and the scientist’s desire for a positive self-image.

What then is the catalyst for advances in science? Medawar adopts Shelley’s idea of poesis in poetry: creation through the act of imagination.  The source of scientific hypotheses is these flashes of vision, and it is these hypotheses which guide and limit further science.  Medawar clearly rejects the idea that scientific discovery can be premeditated and cites the role of luck in scientific discovery.  He carefully qualifies the role of luck by showing how the scientist places himself in a certain mindset amenable to luck through his studies and associations with other scientists.

Medawar’s last essay discusses the limits of science. His fundamental assertion is that science does not yield absolute knowledge, and he quotes Kant as support: “Hypotheses always remain hypotheses, i.e., suppositions to the complete certainty of which we can never attain.” Science’s goal then is not the absolute but the nearest approximation possible; the nearer the approximation, the better its predictive capability.

Continuing the discussion, Medawar observes that there could be either a cognitive inadequacy or a restriction arising out of the nature of the human reasoning process that limits the growth of science, but since any such limitations would be present from conception we would never know of them (just as we could never perceive the Pythagorean celestial music due to its continuous presence in our lives). Are there then limits of science?  Not if science is understood as the art of the soluble.  If something is possible in principle, Medawar states, it can be done if the intention is sufficiently resolute and sustained.

The one limit to science as Medawar sees it is that it cannot answer ultimate questions, e.g. “Does God exist?” Medawar goes on to say he is not indicting science; rather he is recognizing that these questions require transcendent answers, which neither arise from nor require validation by empirical evidence.  He actually takes this argument one step further and asserts these questions have no possible answers. (Medawar recognizes that Immanuel Kant felt the opposite; since somehow man’s nature drives him to ask these questions, Kant felt that answers necessarily exist.)

According to Medawar, the question of whether God exists is outside the realm of science; the leap of faith required for a belief in God is one he himself is unwilling to make.  Although Medawar did not personally believe in transcendent answers, he did feel that these answers had a usefulness measured by the peace of mind they bring people.

I bought this book because as a Roman Catholic I was interested in what a scientist had to say about the limits of science in answering ultimate questions.  Medawar confirmed my suspicions that science can play at best only a subsidiary role with regards to these ultimate questions and the religious beliefs they help spawn.

For anyone looking for an introduction into what science is, how it advances, and what questions it can and cannot answer, Medawar’s book is excellent.  Perhaps the one idea I am always left with after reading this book is although science has limits, as long as man retains his ability to create imaginative hypotheses and his inclination to ascertain whether his guesses correspond to reality, there will always be more beyond for intrepid explorers in the realm of science.

Professor Hannaway appended the following note at the end of my review:

“What do you think your reaction would have been if you had read a book by a scientist less sympathetic to the claims of religion?  Perhaps you can find one, read it, and then critically assess the arguments of Medawar.”

“Why do you think a famous scientist like Medawar was so concerned by such questions to write about them in this way?  Could you find out something about his life that might explain this?  Try sources like the Times obituary columns, Nature, Notes and Records of the Royal Society.”

That was Owen: always generous with advice, and always trying to spur you to dig deeper, to learn more.

Bonus Anecdote: I’ll never forget this saying of Owen’s: “Scotch is for after dinner.” The last time I saw him in Denver at a conference, I was really pleased to track down a glass of single malt whisky for him.  He was a wonderful man.