This book contains a series of three lectures given at the University of Washington on the role of science in modern society. Feynman himself was against the publication of these lectures feeling that they lacked polish, focus and ultimately wouldn't be the same on paper as they were in person. Anyone who's ever seen the man in action can readily attest to his electrifying personality; it is really quite conceivable that you could listen to him talk about anything forever. I have to say that Feynman's instinct was right. While he does make some interesting points and observations about the role of science and common misperceptions related to scientific endeavor, these lectures lack focus. There are tons of tangents and anecdotes, brilliantly funny and charming, to be sure, but of the sort that lend themselves to the type of barroom storytelling Feynman was famous for - not in a published work.
The outline of the lecture series is roughly as follows:
Lecture 1: What the heck do people mean when they talk about science? Feynman presents a three-part definition. Science is a method for finding things out about the universe, it is the corpus of things that we do find out (the information and knowledge), and it is the application of this information for practical purposes (technology). What value does it have? Why should we fund it? Pursue it? For the very fact that science is exciting. The things we learn themselves are valuable in and of themselves in that they give us a sense of wonder about the universe. He leads out with a point that seems counterintuitive and thus always misunderstood. Science is the pursuit of answers, but really, all good science rests on uncertainty. Science is never, ever certain of things. We sift through explanations for observable phenomena and as our technology and understanding progresses the holes in the sieve become smaller and more discerning. Any theory we have to explain things, the laws of the universe themselves, are merely our best guesses within a certain degree of uncertainty. The prime example of this is the relationship between motion and mass. It was thought, in the past that an object maintained its mass no matter how fast it traveled. This hypothesis was tested, using a spinning top. When scientists weighed the top in motion and at rest they found the mass to be "the same" - validating the theory and elevating it to a law of nature, and they were right, to within the degree of accuracy they were capable of measuring mass and to the extent they were able to speed it up. Relativity smashed this idea. It turns out, if you accelerate an object to somewhere near the speed of light, its mass does indeed change and becomes heavier. Even a top traveling at nonrelativistic speeds changes mass, but only to a few parts in a billion, and thus, it is almost imperceptible. Anyway, the point is we must be skeptical of all our ideas if science is to thrive and survive, even with our most cherished and commonsense laws. I think this notion especially pertinent in our age of undetectable dark matter. Seems like almost everyone I read is universally convinced of the existence of matter we have not seen or detected directly merely by the effect it has on matter we can see. It may well be true, that the presence of dark matter is the reason for faster than expected galactic rotation and the like - it is certainly the most plausible answer in a field of many, but the ridicule which MOND activists and people who challenge the idea of dark matter with alternatives receive seems unhelpful and indeed counterproductive to the advancement of science. Challenging ideas should be entertained and tested as hurdles which probable explanations need to fly over if they wish to survive. They should be welcome as tests, not dismissed as quackery outright because they don't presume to follow the generally accepted norm. These arguments very closely mirror those presented by Lee Smolin in The Trouble with Physics: The Rise of String Theory, the Fall of a Science and What Comes Next and are worth taking seriously in physics departments and grant-giving institutions worldwide.
Lecture 2: What are the limits of science? In this lecture Feynman focuses pretty much on what science cannot be used for and particularly focuses on the realm of value judgments. He does so by pointing out frequent clashes between science and religion from a historical point of view and tries to ameliorate the tension between the two. His arguments are lucid, without condescension and full of the usual self-deprecating humor that I think many atheists today lack when dealing with religious people they disagree with chiefly by admitting his own ignorance on certain issues and relegating his criticisms to only pertinent parts of religions with which the scientific mind might conflict (these he terms metaphysics or the explanations given in religions for what things are, e.g. what is the universe, what kind of thing God is). In this realm, there have been great conflicts between religion and science that predate, but are most famously illustrated by the conflict between Galileo's observations and insistence in a heliocentric solar system and the Church's insistence on the ancient Ptolemaic model of geocentrism. Feynman admits in this respect the world's religions have been forced to give ground to science, but what has remained untouched are the values and ethics which religions harbor and inspire in humankind. Science can give no aid to deciding which ethical, or political systems for that matter, are superior to any other because it's not something science can directly measure through its chief analytical tool: observation. In the end, judgment must stem from the individual and because of our mutual uncertainty we should have tolerance and promote an acknowledgement of our own ignorance in matters - a viewpoint that naturally leads to humility and all of us getting along a lot better.
Lecture 3: In what areas can science or scientific thinking be put to broader use in society? This is probably the most jumbled and convoluted lecture in the series, but also the most charming and witty. The basic argument: if we were all a little more scientific and rational with our thinking perhaps we wouldn't be taken advantage of by advertisers so much and believe in dumb things like psychic powers and ufos. Again, the key to everything is uncertainty. There's no way to disprove all of these dumb things. So what makes them dumb? Their exceeding unprobability, not their lack of possibility.
If you're a big Feynman fan and know him very well, this might still contain some value for you, but if you're looking for some insight into one of the 20th century's greatest thinkers and personalities, you're better off sticking with Surely You're Joking, Mr. Feynman!, which is a shame because I think the core theme of acknowledging that we don't really know anything as certainly as we think we do is worth reminding a people and civilization in love with our own capability and strength. In an era when breakthrough follows breakthrough, it's important to remember that no matter how stunningly smart we think we are, we'll never know everything there is to know with absolute certainty - and why would we?! It would take the mystery and motivation out of the pursuit of science and we'd be a poorer culture for it. I'd love to get my hands on a recording of the lecture! I'm sure it would be much, much better!