Cheaper than printing it out: buy the paperback book.

Out of Control

In the fabric of knowledge we call science, there was a rent here, a hole. It was filled by young enthusiasts not burdened by wise old men. This gap made me wonder about the space of science.

Scientific knowledge is a parallel distributed system. It has no center, no one in control. A million heads and dispersed books hold parts of it. It too is a web, a coevolutionary system of fact and theory interacting and influencing other facts and theories. But the study of science as a network of agents searching in parallel over a rugged landscape of mysteries is a field larger than any I've tackled here. To deal fairly with the mechanics of science alone would require a larger book than I've written so far. I can only hint at such a system in these closing pages.

Knowledge, truth, and information flow in networks and swarm systems. I have always been interested in the texture of scientific knowledge because it appears to be lumpy and uneven. Much of what we collectively know derives from a few small areas, yet between them lie vast deserts of ignorance. I can interpret that observation now as the effect of positive feedback and attractors. A little bit of knowledge illuminates much around it, and that new illumination feeds on itself, so one corner explodes. The reverse also holds true: ignorance breeds ignorance. Areas where nothing is known, everyone avoids, so nothing is discovered. The result is an uneven landscape of empty know-nothing interrupted by hills of self-organized knowledge.

Of this culturally produced space, I am most fascinated by the deserts -- by the holes. What can we know about what we don't know? The greatest promise looming in evolution theory is unraveling the mystery of why organisms don't change, because stasis is more common than change yet harder to explain. What can we know about no-change in a system of change? What do the holes of change tell us about the whole of change? And so, it is the holes in the space of wholes that I'd like to explore here.

This very book is full of holes as well as wholes. What I don't know far exceeds what I know, but unfortunately, it is far easier to write about what I know than about what I don't know. By the nature of ignorance, I am, of course, not aware of all the places and gaps where my own knowledge fails. Recognizing one's own ignorance is quite a trick. That goes for science, too. Mapping the holes of ignorance is perhaps science's next advance.

Scientists today believe science is revolutionary. They explain how science works via a model of ongoing minirevolutions. According to this perspective, researchers build a theory to explain facts (for example, rainbows occur because light is a wave). The theory itself will suggest places to look for new facts (can you bend a wave?). It's the law of increasing returns again. As new facts are uncovered they are incorporated into the theory, buttressing its strength and reliability. Occasionally, scientists uncover new facts that aren't readily explained by the theory (light sometimes acts like a particle). These are called anomalies. Anomalies are set aside at first, while new facts that concur with the reigning theory continue to stream in. At some point, the accumulating anomalies prove too great, too troublesome, or too numerous to ignore. Inevitably then, some young turk proposes a revolutionary different model that explains the anomalies (such as, light is both wave and particle). The old is gone; the new quickly reigns.

In the terminology of science historian Thomas Kuhn, the reigning theory forms a self-reinforcing mindset called a paradigm that dictates what is fact and what is mere noise. From within the paradigm, anomalies are trivia, curiosities, illusions, or bad data. Research proposals endorsing the paradigm win grants, lab space, and degrees. Proposals operating outside the paradigm -- those dabbling in distracting trivia -- get nothing. The famous scientist who made his great revolutionary discovery while denied funds or credibility is so common it's become cliché; I've trotted out several of those cliché stories in this book. One example is the ignored work of scientists dabbling in ideas that contradict neodarwinian dogma.

Real discovery in science, according to Kuhn in his seminal The Structure of Scientific Revolutions, only "commences with the awareness of anomaly." Progress is an acknowledgment of the opposition. A series of established paradigms are overthrown by downtrodden and oppressed anomalies (and their finders) as they rebel and usurp the throne by their countertruth. The new ideas reign, at least for a while, until they too become ossified and insensitive to the squawks of new anomalies, and are eventually overthrown themselves.

Kuhn's model of paradigm shift in science is so convincing that it has become a paradigm itself -- the paradigms of paradigms. We now see paradigms and paradigm overthrows everywhere, inside of science and out. Paradigm shifts are our paradigm. The fact that things don't really work that way is, well, an anomaly.

Alan Lightman and Owen Gingerich, writing in a 1991 Science article, "When Do Anomalies Begin?," claim that contrary to the reigning Kuhnian model of science, "certain scientific anomalies are recognized only after they are given compelling explanations within a new conceptual framework. Before this recognition, the peculiar facts are taken as givens or are ignored in the old framework." In other words, the real anomalies that eventually overthrow a reigning paradigm are at first not even perceived as anomalies. They are invisible.

A few brief examples of "retrorecognition," based on Lightman's and Gingerich's article:

  • The fact that the shape of South America and Africa fit together like a lock and key did not bother any pre-1960s geologists. There was nothing troubling to them or their theories of continent formation in this observation, or in the observed ridges down the center of the oceans. Although the remarkable fit had been noticed since the Atlantic Ocean was first mapped, it was a fact that did not even need an explanation. Only later was the fit retrorecognized as something to explain.

  • Newton precisely measured the inertial mass of a great many objects (what it took to get them moving, as in getting a pendulum started) and their gravitational mass (how fast they fell to the Earth), to determine that the two forces were equal, if not equivalent, and could be canceled out when doing physics. For hundreds of years this relationship was not questioned. Einstein, however, was struck that "the law has not found any place in the foundations of our edifice of the physical universe." Unlike others, he was perplexed by this observation which he successfully explained in his revolutionary general theory of relativity.

  • For decades, the almost exact balance between the universe's kinetic and gravitational energies -- a pair of forces that kept the expanding universe balanced between blowing up or collapsing -- was noted in passing by astronomers. But it was never a "problem" until the revolutionary "inflationary universe" model came along in 1981 and made this fact a troubling paradox. The observation of the balance did not begin to be an anomaly until after the paradigm shift, when in retrospect, it was seen as a troublemaker.

    The common theme in each example is that anomalies begin as observed facts that don't require any explanation at all. They are not troublesome facts; they just are. Rather than the cause of a paradigm shift, anomalies are the result of the shift.

    In a letter to Science, David P. Barash tells of his own experience with nonanomalies. He wrote a textbook of sociobiology in 1982, where he stated that "evolutionary biologists, beginning with Darwin, have been troubled by the fact that animals often do things that appear to benefit others, often at great cost to themselves." Sociobiology was launched by the 1964 publication of William Hamilton's inclusive fitness theory, which provided a workable, though controversial, way to interpret animal altruism. Barash writes, "However, stimulated by the Lightman-Gingerich thesis, I have reviewed numerous pre-1964 textbooks of animal behavior and evolutionary biology and have discovered that, in fact -- and contrary to my own above-cited assertion -- before Hamilton's insight, evolutionary biologists were not very much troubled by the occurrence of apparently altruistic behavior among animals (at least they did not devote much theoretical or empirical attention to the phenomenon)." He ends his letter by suggesting, half in jest, that biologists "teach a course in what we don't know about, say, animal behavior."