The Human Side of Science: Edison and Tesla, Watson and Crick, and Other Personal Stories behind Science's Big Ideas (2016)

Introduction: Science's Evaluation System

Science is best defined as a careful, disciplined, logical search for knowledge about any and all aspects of the universe, obtained by examination of the best available evidence and always subject to correction and improvement upon discovery of better evidence. What's left is magic. And it doesn't work.

—James Randi, The Mask of Nostradamus, 1993

This book needs an additional subtitle: “Contention, Cooperation, and Connection between a Wide Variety of Scientists throughout the World over the Last 2,500 Years That Resulted in Good, Bad, and Even Ugly Interactions among Scientists.” That is a mouthful, but it gives you a far more complete picture of what you're about to read.

Our chronology begins in ancient Greece with conflict between the ideas of Democritus and Aristotle regarding the existence (or nonexistence) of atoms. Although Democritus and Aristotle never met in person, their ideas clearly influenced each other. It ends in the 1990s with contention between J. Craig Venter's Institute for Genomic Research and the Human Genome Consortium, headed by Francis Collins. This was good, in that both groups probably worked faster and more efficiently since there was a competition. But it was also bad in the sense that cooperation or collegial interactions between researchers became strained, and valuable insights were not shared in a timely fashion.

Isaac Newton's rivalry with Robert Hooke was bad in that the feud consumed valuable time that could have been spent more productively. But it was good in that the contention may have spurred both of them to accomplish more than they might have otherwise. Good and bad are actually in the eye of the beholder and may vary as time progresses. The real question here is the determination of BIG scientific ideas: they must be judged by objective standards rather than by subjective evaluations.

So before we start in on the juicy details of the lives and times of a wide variety of interesting people (almost four hundred in all) from around the world, let's take a look at the process that is intended to govern the evaluation of scientific ideas and determines which ideas are actually BIG. Ahh, now we're down to the real nitty-gritty: Science's evaluation process is the ultimate arbiter of scientific ideas.

Hang on; there's a good bit of detail here, but grasping the backbone of the scientific process is key to understanding what makes some ideas big and some not so big.

Let's start off by dealing with a giant misconception: To some people, science may seem like a huge, rambling house with many suites of rooms corresponding to the branches of science, all nice and neat and tidy.


Science Building. Used with permission from Sidney Harris.


Science's process or evaluation system can be viewed in retrospect as a sequence of specific steps, often referred to as the scientific method:

Observation: Certain specific happenings in physical reality are sensed.

Hypothesis (H): A general idea of the nature of all such events is created.

Prediction: Presuming the H to be true, some similar happening is forecast.

Experiment: A test is conducted to see if the predicted outcome will occur. A new specific happening in physical reality is sensed as a result of the test.

match => H is supported but not proved
Compare Experiment with Prediction
doesn't match => H must be modified


Within the observation step, some specific, real occurrence is perceived by our senses with or without the aid of instrumentation. Viewing the sky with sharp eyes is a good start, but the telescope opened whole new vistas. While the natural sciences (e.g., physics, chemistry, biology, etc.) have a large number of identical subjects to observe (think carbon atoms), the human sciences (e.g., political science, sociology, economics, etc.) have a smaller number of distinctly different subjects (think human beings, even identical twins).


Human nature being what it is, information will be collected for just so long before the mind, in its search for order, begins to sift through this information and to construct patterns or develop explanations of these data. Viewing the stars is good, but noting that some seem to be fixed starts a whole new process. This is called the hypothesis step.

The reasoning that considers specific observations and constructs a general hypothesis is known as inductive logic, which involves making a generalization, and is the most precarious type of reasoning. Some people make an art form of jumping to conclusions, but within the context of the scientific method, that activity is restricted because succeeding steps help bring the hypothesis back to reality. Developing a hypothesis or a theory means about the same thing, but hypothesis is a more fancy-sounding word.


Often, the hypothesis is framed as a whole or in part in a language different from normal linguistic fare. That language is mathematics. Because mathematical skills require a great deal of effort to acquire, explaining scientific hypotheses to people not trained in mathematics requires translation into conversational language. Unfortunately, the meaning of the hypothesis often suffers in the process.

Mathematics is like vitamins—every science needs at least its minimum daily requirement. The type of mathematics used in the sciences varies according to need. Virtually all sciences make use of arithmetic to express their results in tangible form. In addition to arithmetic, algebra and geometry are used by all natural sciences. Physics and chemistry go further to include calculus as well as more esoteric forms of mathematics. Biology and the human sciences often employ statistics to characterize and evaluate their nonidentical populations in statistical terms, like averages (means) and variations from the norm (standard deviations). As the figure shows, some math is very advanced.


Square Root of Chicken. Used with permission from Sidney Harris.


Once a particular hypothesis is formulated, it is used to forecast some future event that will occur in a particular way if the hypothesis is true. This prediction is derived from the hypothesis using deductive logic. This form of logic starts from a true general statement and derives a true specific example from it. For example, Isaac Newton's second law of motion is expressed as F = ma, where F is force, m is mass, and a is acceleration. If m = 3 units and a = 5 units, then F should be 15 units. Carrying out this step is an ideal task for computers, which have deductive logic built into their programs. Newton's hypothesis was applied to a specific case, and a prediction was made that could be tested in reality.


Once the prediction is made, the next step is to perform an experiment to see if the prediction is supported by evidence. While this sequence is easy to state, in many cases it is extremely hard to accomplish. Intricate, expensive, labor-intensive scientific apparatuses have been constructed and operated by dedicated experimenters, and much valuable data has been collected. The natural sciences have the great advantage of being able to isolate the object of study (think test tubes), while the human sciences often have to contend with several variables being simultaneously filtered through the minds of people having their own agendas (think surveys in which participants complete questionnaires).


Once the experiment phase is complete, the result is compared with the prediction.

If the experiment matches the prediction, then the hypothesis is supported. Since the hypothesis is a general idea and the experimental results are specific results from reality, a specific favorable result can't necessarily prove a general hypothesis, it merely supports it. On the other hand, if the experimental result doesn't match the prediction, some aspect of the hypothesis is false. This feature of the scientific method is called falsifiability, and it places a stringent requirement on hypotheses. As Einstein said, “No amount of experimentation can prove me right, one experiment can prove me wrong.”1 Positive results only lend more support to the proposed hypothesis, but negative results can undermine it completely.


A hypothesis that is shown to be false in some way must be recycled. That is, it must be modified slightly, changed radically, or abandoned altogether. The judgment about how much to change can be an extremely difficult call. The recycled hypothesis will then have to work its way through the sequence again and hopefully survive the next prediction/experiment comparison.


Another facet of the scientific method that keeps the process on target is replication or repeatability. Any observer suitably trained and equipped should be able to repeat prior experiments and obtain comparable results. In other words, there's constant rechecking going on in science.

For example, a team of scientists at Berkeley Laboratory in California attempted to synthesize or produce a new element by bombarding lead targets with an intense beam of krypton ions (ions are atoms that have become charged by removing an electron) and then analyzing the resulting products. The Berkeley scientists announced their finding in 1999: one of the products was the synthesis of element 118 (118 is the number of protons in the atom's nucleus).

Synthesis of a new element is interesting news because of the new element's novelty. In this case, a favorable result would also support previous ideas about the stability of heavy elements like lead. Scientists at other laboratories in Germany, France, and Japan, however, were unable to duplicate the reported synthesis of element 118. An augmented Berkeley Laboratory team repeated the experiment, but they, too, failed to reproduce the earlier reported synthesis. After the Berkeley team reanalyzed the original experimental data using revised software codes and were unable to confirm the existence of element 118, they retracted their claim. This refinement process shows that science's quest to understand reality is, and must be, a never-ending story.


Scientific Method. Used with permission from Sidney Harris.

Another example of rechecking in science involves repeating the testing of a prediction. In February 2001, Brookhaven National Laboratory in Long Island, New York, reported an experimental result for a property known as the magnetic moment of the muon (a negatively charged particle similar to the electron but considerably more massive) that was slightly larger than the prediction from the Standard Model of particle physics. Because the Standard Model's predictions had been matched by experimental results to an extremely close tolerance for many other particle properties, there was a strong implication that this discrepancy in the magnetic moment of the muon indicated that the Standard Model was flawed.

The prediction was the result of a complex and lengthy calculation that had been done independently by groups in Japan and New York in 1995. In November 2001, the calculation of a muon's magnetic moment was repeated by physicists in France. The French physicists discovered an erroneous minus sign on one of the terms. They posted their results on the Internet. As a result, the Brookhaven group rechecked their own calculations, acknowledged the mistake, and published their corrected results. The net effect of this correction was to reduce the disagreement between the prediction and experimental results to an amount within the accuracy of the experiment. Although the Standard Model survived this challenge intact, it awaits and must withstand future challenges as science's never-ending search continues by testing to see if the hypothesis yields predictions that are matched by experimental evidence.


Let's follow an example of the scientific method at work, step by step:

OBSERVATION: J. J. Thomson, the director of the Cavendish Laboratories in England just before the turn of the twentieth century, observed what happened to a beam of light produced in a cathode-ray tube (forerunner of the modern TV picture tube). Since the beam (1) deflected toward positively charged electrical plates and (2) hit its target, producing individual flashes of light, it had to consist of negatively charged material since opposite charges attract, and the beam was attracted by positively charged plates. Since individual flashes of light were seen, this beam likely consisted of individual particles. These particles were named electrons by Irish professor George FitzGerald in his comments on Thomson's experiment.2

HYPOTHESIS: Since atoms are uncharged (neutral), and Thomson had found negatively charged particles within them, he deduced there must be some positive charge in atoms as well. In 1903, Thomson theorized that the positive charge was smeared throughout the whole atom, with the negatively charged electrons embedded inside the positive material. This depiction resembled a traditional British dessert and was therefore referred to as the Thomson Plum Pudding Model of the atom.

PREDICTION: Ernest Rutherford was an expert on positively charged particles known as alpha particles. He predicted that if these particles were shot at atoms consisting of the sparse and smeared-out positive charge of the Thomson Plum Pudding atom, it would be like shooting pool balls at a fog cloud. All particles ought to rip right through the smeared-out positive charge.

EXPERIMENT: In 1909, Hans Geiger and Ernest Marsden set up an apparatus to shoot alpha particles at an extremely thin sheet of gold atoms (gold was used because it could be made so thin). The results were somewhat different from what they expected. Although most of the alpha particles did go straight through, some alpha particles were deflected at large angles, and some even bounced back. Rutherford said, “It was almost as incredible as if you fired a fifteen-inch shell at a piece of tissue paper and it came back and hit you.”3

RECYCLE: The Thomson Plum Pudding Model was replaced by the Rutherford Solar System Model, in which the positive charge was concentrated in a relatively tiny nucleus at the center of the atom (and thus could deflect a small number of alpha particles) and electrons (analogous to planets) that moved in circular orbits around the nucleus (analogous to the sun). (Recycling is often partial, in that some aspects of an earlier model are maintained, but it can be total, as we will see in later chapters.) Later in the twentieth century, as a result of subsequent predictions and experiments, the Rutherford Solar System Model of the atom was replaced by a series of other, more sophisticated models.

Whenever experimental evidence doesn't match the prediction of an existing hypothesis, it's time to recycle the hypothesis. Despite the popularity of an earlier idea, the celebrity status of a theory's proponents, the unattractiveness of a competing new theory, or the difficulty in understanding it, the bottom line is: experimental evidence rules.


Perhaps as a result of seeing scientific methodology presented in this way, some people believe science operates in a cut-and-dried fashion, with rational logic always prevailing. Those people are mistaken.

There are elephants in science's rooms. All the steps in the scientific method involve people. And you know what that means: The seemingly well-defined procedural steps of the scientific method are, when put into practice, actually fuzzy—and subject to the full range of human foibles. We may be an imperfect lot, but we are very curious about our surroundings and can have several opinions about any one matter. Individually and collectively, all sorts of contentioncooperationand connection—and even serendipity—take place. Often, we resist change and become contentious or even obnoxiously ugly. On the other hand, we might see immense value in the ideas of our fellow humans and cooperate nicely. Additionally, our seemingly vast world is actually much smaller than it appears. There are connections among people that appear in unexpected places and contexts. This book aims to explore this complex territory, with an emphasis on the people involved.


Elephant House. Used with permission from Sidney Harris.