American Scientist

Plows, Plagues, and Petroleum: How Humans Took Control of Climate. William F. Ruddiman. xiv + 202 pp. Princeton University Press, 2005. $24.95.

In Franconia in southern Germany (where I grew up), there are plenty of towns with names ending in reuth, the best known of these being Bayreuth, famous for performances of operas inspired by Teutonic mythology. Reuth denotes a clearing—that is, a place ridden of trees. To the south we find town names with the endings ried and reid, whereas going north we note rode, roda and rade, until in Denmark we encounter roed. Typically, the settlements so named are more than 1,000 years old. In each instance the meaning of the suffix is the same. Settlers would have had to clear forest not just for the towns but for farmlands to support them.

The necessary amount of clearing can be figured from the caloric requirements for food, if one can estimate the number of settlers. Considering that we know that Roman generals complained about the hazards associated with the impenetrable forests north of the Alps, it is easily deduced that the clearings took place within the past 2,000 years or so. Presumably, beginning with Roman colonization and accelerating during the centuries after the collapse of the Roman Empire, the rate of forest removal in Europe has followed the shape of a logistic curve, exponentially increasing until one half of the forest was gone, in the late Middle Ages, and decreasing since then toward the present situation, in which we find that patchy forests remain only on soil that is poorly suited for farming.

This much is general knowledge. What William Ruddiman has done in Plows, Plagues and Petroleum, an attractive, well-written new book aimed at a popular audience, is to explore the geochemical and climatological implications of worldwide deforestation over the past several thousand years. He concludes that the clearing of forests on a global scale resulted in the input of significant amounts of heat-trapping carbon dioxide to the ocean-atmosphere system. In addition, he points out as possible greenhouse drivers the warming effects of the methane supplied by expanding rice fields and increasingly numerous cattle herds. In short, his thesis is that anthropogenic climate change began long before the Industrial Revolution. What's more, Ruddiman argues that this meddling with climate was probably a good thing, because it prevented the general cooling he says one should otherwise expect according to the predictions of Milankovitch theory.

Milankovitch theory posits that Earth's climate changes as a result of cyclic variations in the way our planet orbits the Sun (which affect the way sunlight gets distributed over the globe). Although some aspects of the theory remain controversial, it has proved largely successful in accounting for the dramatic swings in climate that have taken place in the past on the scale of thousands of years.

Whether or not one accepts Ruddiman's Milankovitch-based arguments for what should have happened, the basic proposition that human activities have affected climate during much of the Holocene seems entirely reasonable. After all, there are strong arguments that humankind had a crucial role in the extinction of mammoths, woolly rhinoceroses, cave bears and dozens of other types of Pleistocene mammals. It would be surprising indeed if our activities had not influenced patterns of vegetation, erosion, albedo (the fraction of sunlight reflected back into space) and the chemistry of the atmosphere. The question is not whether we have had an effect, but whether one can quantify it, and whether this quantification warrants the conclusion that global climate was substantially altered.

Ruddiman, to his credit, makes the effort. With great ingenuity he digs up clues to the growing numbers of humans on the planet, their impact in terms of forest removal, the addition of trace gas to the atmosphere resulting from that removal, and the positive feedbacks within the climate system that amplify the effect. His main rationale for searching for human impact on atmospheric chemistry over the past several thousand years (rather than the past two centuries) is that there is a perceived discrepancy between the carbon dioxide content seen in ice cores and what might be reasonably expected by comparison with earlier interglacial periods. According to Ruddiman, the discrepancy can be quantified, and the results support the claim that were it not for human emissions of carbon dioxide (and methane), northern ice masses would have begun to build up beginning some 5,000 years ago. He further maintains that the apparent stability of the climate for most of the Holocene is the result of a natural trend moving the system toward cooling being coincidentally canceled out by a human-made trend moving it into warming.

Ruddiman's line of argumentation seems compelling in its major aspects. However, there is room for disagreement and discussion. For example, invoking "coincidence" is rarely a good strategy. It is better, I think, to look for feedback keeping the Holocene climate system in its preferred state. In my opinion, Ruddiman has identified an important factor that has, against expectations, kept this feedback operating.

There are other places where one might argue as well. The focus on plows trivializes the impact of burning before farming spread. I suspect that pyromania may be equal in importance to plows.

Ruddiman's contention concerning plagues is that they depopulated many previously deforested regions, allowing forests to grow and bind carbon, thus reducing the greenhouse effect—an interesting proposal, but one that presumably implies a sensitivity of climate to carbon dioxide that is toward the high end of possibilities. The result, according to Ruddiman, was a short-term cooling, manifested as the well-known "Little Ice Age." A more traditional view would be that the rise of this period featuring high numbers of nasty winters and wet summers was caused by a dimming of the Sun. This era produced poor harvests and widespread hunger, making people more susceptible to disease. The evidence supporting a role for the Sun in climate change has grown stronger over the past several decades.

The petroleum section seems more solid. Experience shows that reserves expand with increased efforts in exploration as the price of oil rises in response to supply falling behind demand. As the cost of producing oil rises, we shall likely turn increasingly to other energy sources, including coal. Ruddiman's projections of the resulting increase in atmospheric carbon dioxide seem conservative in a "business as usual" context.

What might be the general impact of Ruddiman's propositions on the global-warming debate? Most likely, as Ruddiman realizes, his thesis will please no one. Those alarmed by present trends will object to the notion that anthropogenic warming saved us from glacial advances and associated climate deterioration. And those who, when it comes to energy use, place comfort over concern or who want to protect special interests will resist the implication that climate is indeed quite sensitive to human additions of carbon dioxide and methane to the atmosphere. Above all, Ruddiman's argument makes it clear that there is no "natural" baseline of climate in the late Holocene from which to reckon the human impact of the past two centuries. Ecologists refer to this sort of conundrum as a "shifting baseline." The application of this concept to climate change is a major contribution.