It is generally believed that something called Science is an influence that has transformed human life in the last two centuries, but what is the nature of this agency that has such great power? Practicing scientists are often unwilling even to try to define it. If they make the attempt they come up with very different answers according to the type of work that they do. Those who are concerned with fundamental research think of science as discovery and may give a definition quite different from that of technologists. Does science consist in the discovery of new facts and ideas or the application of them in human affairs? And if it is both of these, which comes first? Did navigators discover the compass or was it vice versa?

A further difficulty is the tremendous differences in the theoretical and practical methods that are used in the various sciences. There would probably not be much agreement on a definition between a psychologist and a nuclear physicist. The technology of satellites is quite different from that of agriculture—yet we call both scientific.

Clearly all these scientists use activities of the brain, and it would be very satisfactory if we could specify the particular nervous processes that are at work. Unfortunately neuroscience is still in a very primitive state. The physiologists who study nerve fibers and reflex actions can hardly even imagine that they could ever tell us what goes on in the brain when a person solves a mathematical problem. Nevertheless it is useful to consider what would be involved if our successors ever come to be able to discuss the question of the nature of science in terms that are themselves scientific. The very mention of this possibility itself raises puzzling questions about the nature of knowledge. The problems involved seem to be those that might be disentangled by the brains of philosophers rather than of scientists, which leaves us in a fine muddle.

Perhaps if we want to know what science is it may be better to look at it in operation, rather than analytically. This is in effect what Horace Freeland Judson has done by recounting the history of a large number of scientific discoveries. These are grouped to illustrate a set of themes that are prominent in scientific work, such as pattern, change, chance, and prediction. He is therefore very much interested in ideas and theories as well as facts. Indeed a general theme of the book is that science consists in discovering the relations between things. There is much emphasis on questions of mapping and modeling. But the layman, for whom the book is intended, will probably enjoy chiefly the accounts of the phenomena and processes that are described. Moreover there are many beautiful colored pictures. The general effect of the book is thus to give something of the atmosphere or “feel” of science, without having to go through the painful process of detailed technical study. It is a method that will appeal to many people even if scientists themselves feel that it is cheating to leave out so much detail.

In order to give the subject some personal flavor the text is interspersed with interviews with famous scientists. I find these interludes to be confusing, rather than helpful. Does it help to understand science to be told that one of Glenn Seaborg’s colleagues is “plump, open-faced, and relaxed” while the amateur astronomer George Alcock “is of medium height, stocky, in his late sixties, gruff, and matter-of-fact” and Manfred Eigen “is short and dapper, with a tanned, youthful face, white hair, and a compulsion to explain science”?

The interviewees no doubt do their best to describe how they think that science proceeds. But their various contributions add up to a pretty confused view. Murray Gell-Mann, the physicist who is author of quarks and their “charm,” tells us that “one way to describe what’s going on is to say that nature apparently resembles itself at different levels.” Can anyone make sense of this? Perhaps easier to understand is the contribution of the physicist Paul Dirac, when he says it is “most important to have a beautiful theory,” and that new ideas come from trying “to imagine what the universe is like.” This is appealing but again can hardly be said to tell us very much about the scientific process. Still the personal touches may make it easier for some people to follow the complicated network of ideas that the author holds to be involved in scientific inquiry.

Judson begins by suggesting that the laws of the universe are somehow related to patterns and rhythms. Curiously the first case he discusses is the disturbance of our own bodily rhythms that is produced by jet-lag when we travel. This example suggests the thought that perhaps the search for the rhythmicity and regularity and pattern of phenomena is somehow an attribute of ourselves, not necessarily of the environment. Yet rhythm certainly appears in many forms in nonliving as well as living matter. We see it in nature and imitate it in art. The book gives many beautiful examples from the meanderings of snakes and of rivers to the spirals of DNA.


So Judson moves on to the question of pattern and then to models and maps. There are some interesting examples from early atlases, but the author’s object is to show that constructing models is an essential feature of scientific discovery. He uses the example of how John Snow investigated the epidemic of cholera in London in the last century by making maps of the distribution of the patients and of the wells or pipes from which they got their water. So he was able to show that infection came from one but not the other of two private water suppliers.

This may not seem to be very great science but Judson makes his point by putting it on the same page as one of the most famous of all scientific patterns, the periodic table of chemical elements. I can remember well the first time that I read about this at school and thought there was something interesting in chemistry after all. What the Russian Mendeleev showed in 1869 was that all the elements of which the Earth is composed fall into a series of families. As he wrote, “The elements, if arranged according to their atomic weights, exhibit an evident periodicity of properties.” For example fluorine, chlorine, bromine, and iodine all have some things in common and differ from other families such as sodium, potassium, and calcium. Indeed the whole universe is composed of these same elements and Mendeleev’s table of families allowed him to predict the properties of elements unknown at that time and these have duly been discovered since. Here is a principle that makes sense of all the many strange things that happen when different substances are mixed: chemistry becomes logical. So, rhythmical repetition is certainly a very important property of nature and not only of our own lives.

Prediction, then, is an obvious feature of science but Judson does not proceed to it at once. There are many other matters to be dealt with. The first is change, and for this he has the curious idea of telling the history of armor—used in the defense of men in battle or by animals. How was it that Richard III was such a famous warrior although he was so small? Perhaps for that very reason his armor was lighter and so left him more mobile. This principle of the importance of size and scale “explains an astonishing variety of puzzles in science and technology: why flies can walk on the ceiling, why the tallest tree cannot be more than 300 feet high,” and so on. Judson then explains the answer to these riddles if you don’t know it. He goes on to show how the evolution of the thickness of defensive armor can be compared with something quite different, the development of an airplane sufficiently light to be powered by a man. The connection lies in finding the solution to problems of weight. The book is full of such examples of how the same method of approach can be used to solve very different problems.

However as examples of the process of change the evolutions of armor and the flying machine are rather forced. One can think of so many more dramatic elements of change in the universe. Similarly the following discussion of chance seemed to me unhappy. Here Judson deals at length with those accidental discoveries often called serendipitous. Galvani discovered electrical currents in 1791 when the legs of dead frogs hung on copper wire twitched when they swung in the breeze against an iron railing. Certainly this was lucky and he followed it up very cleverly, but he had hung the frog’s legs up in order to study their actions by pinching the nerves with a forceps. It was an accident that they swung in the breeze, but his discovery was not “pure chance,” it was a characteristic use of human brain power. Judson cites many other famous examples of chance and they all raise the same thought.

But of course the greatest discovery in this field is not made by chance but about it—the fact that chance seems to be a fundamental feature of the behavior of atoms. Judson describes this uncertainty principle of physics only briefly and perhaps wisely does not discuss the difficult logical problems about determinism that it raises. Feedback, which he next considers, is conventionally defined as the process by which the factors that produce a certain result are themselves modified or corrected by that result; but it is often used to mean correction of error of one sort or another. It is obviously fundamental in scientific investigation and here we are given the story of the study of “governers,” from Watt’s steam engine to Wiener’s cybernetics. The word feedback is used in several different ways, often rather vaguely. Yet the concept is very powerful.


The patterns of feedback have subtle explanatory power. They bring together matters startlingly divergent—the biochemistry of the cell, for example, the engineering of computers, the price of beef, the way children learn to talk. Indeed, the idea of feedback interlocks with other ideas, about the nature of information and messages, to form the set of theories from which has sprung the second industrial revolution. For all that, the idea of feedback is simple.

And the author illustrates it with a very simple example showing that some feedback is necessary for the continuation of any process. Even the most persistent telephone bore requires that his victim say “yes” or “no” or “um-hum” from time to time. If you are plagued by such a conversationalist, try leaving the responses out; the flow will dry up in a few seconds. This may seem to be a very unscientific example, but the whole problem of “response” and so of the nature of communication and of information is increasingly seen to be the central theme of any theory of knowledge. Judson has little further to say about language, but, as a biologist, I feel that he could have given more attention to science as a product of the special capacity of the human brain for communication.

Any system of control requires some reference standard that sets the level at which the operations shall proceed. This book has something to say about such reference points in the brain. They tell us how much to eat and drink and regulate our sex lives. It is still not generally realized how greatly these parts of the brain determine our appetites and actions and so make us the sort of people that we are. Ultimately these set-points are in turn determined by the information received by hereditary mechanisms in the DNA. This is the elaborate system that sets the aims of life—ensuring that it shall go on. For the biologist it is absurd to ask whether there is any “point” in life. If there were no aims there would be no life. The point of life is life.

An organism, like a traveler, therefore needs a map or model inside itself by which to decide which course of action it would be best to take at each point of life. It is no surprise then that the use of models is at the center of many human enterprises, including science. Models in general function by showing the interrelations of things. They serve an immense range of purposes, from forecasting the weather to regulating the economy, or attempting to do so. In science they often embody hypotheses, making predictions that stimulate experiments. Computers help model makers enormously, indeed some people say that they encourage an excess of speculation about problems too difficult to solve. When a system has many variables there are so many ways that they can interact that any given hypothesis is largely a guess. This is in a sense a limitation on the scientific method. It is felt acutely by biologists, who have to produce models of thousands of chemical actions going on in every cell and between the cells. The same difficulty may lie behind the sad fact that no economist seems to have the sure basis of prediction that was the inspiration of Maynard Keynes.

So we come at last to what Judson calls “strong predictions”; perhaps by using the adjective, he wants to acknowledge the difficulty we have just mentioned. His discussion emphasizes the interconnectedness of science, that new discoveries must always take account of earlier knowledge. This is often not appreciated by the layman who thinks of the discoverer as starting from some entirely new viewpoint. The special theory of relativity is taken as the example here, with an attempt to show how Einstein’s view developed from the defects of the classical Newtonian Theory. The author does not sufficiently emphasize the widespread effects of scientific prediction. The examples he gives deal with the forecasting of earthquakes and the prediction and discovery of new elements, including substances that may last only for one ten-billionth of a second. These are no doubt very serious matters, but in one sense the really important predictions of science are about industrial and agricultural processes that affect us all every day and not only when there is a quake or a spectacular new discovery. This is of course a theme that the book aims to include, yet inevitably it uses the famous examples of discoveries and cannot dwell upon the humdrum daily work in hospitals, in industry, or on farms, which is where science is practiced.

So we come back to the question of whether the essence of science lies in discovery or its application to human welfare. The author does not offer a final summary, and a quotation he gives near the end by the biologist Joshua Lederberg shows the continuing complexity of the problem: “I don’t think there is one logic for science and another logic for the commonsense world…. I think there is a somewhat more systematic use of formal reasoning…. The ability to discover analogies, the ability to generalize…to imagine oneself inside of a biological or other situation—these are some of the pretty obvious talents.” This is a brave attempt, but surely it should be possible to produce something better than such vague statements. Perhaps we shall be able to define science only when we understand what is now unknown: how the brains of scientists operate. If this is ever achieved I think it will show that they do rather more than imagine themselves inside atoms.

This Issue

May 14, 1981