Looking Back to Look Ahead: Using New Techniques to Assess Old Hypotheses

Recently, I picked up an unlikely book (for me) to read entitled Literary Feuds (Arthur, 2002). It contains separate chapters about intellectual or personal feuds between well-known writers. One of the feuds chosen by the author to highlight was one that developed between British technologist C.P. Snow, author of The Two Cultures and the Scientific Revolution and F.R. Leavis, a British literary critic. Aside from my general interest in their personal feud, I was intrigued by a statement by Snow that compared physical scientists and literary figures.

As the author noted,

Snow had observed that one of the chief differences between the accomplishments of scientists and artists is that no scientist "need ever read an original [scientific] work of the past," even of such giants as Earnest Rutherford, one of the greatest experimental physicists; their substance has all been infused into what is now known because science is cumulative, and embodies its past.

Literature, on the other hand, might change, but unlike science it does not progress: Shakespeare's and Tolstoy's works must read anew by each generation. The modern reader, or the one a hundred years from now, will not understand the "Shakespearean experience better than Shakespeare." But, he argued, the typical teenager will "know more physics than Newton" (p. 126).

Was Snow correct when he wrote that scientists need not ever read an original scientific work of the past? Personally, I think he is dead wrong. As researchers test a hypothesis and write about it, another researcher may try to reproduce the results. Yet another might challenge the original scientific findings and methods. So goes the scientific process. Eventually, researchers produce results that become generally accepted. Hence, there is no reason for students to go back to the original works of famous, as well as not-so-famous, scientists. Scientists today do indeed stand on the proverbial shoulders of their predecessors.

My argument with Snow's statement -- that students of science do not have to read the early works in their fields -- is based on the following belief: I think students would benefit greatly if they could be encouraged to read the earlier works in science. I believe this because using today's research methods and technological innovations may help new students to identify overlooked ideas and reevaluate hypothese of scientists from, say, the early decades of the 1900s, which can now be tested using contemporary knowledge. However, what most science students are taught today are summaries of previous summaries, which are, in fact, summaries of even earlier summaries of ideas identified and written by scientists decades ago. By getting young scientists to read the original materials of the older ones might help them to identify new ideas.

An example of the point I want to make centers on the work of a British scientist by the name of Gilbert Walker. Walker spent a good part of his professional life trying to identify linkages between weather conditions in one part of the world with weather events in other parts. Scientists call these linkages "teleconnections." Walker's work was at first designed to predict the onset and behavior of the monsoons in India, which in his day was a central part of the British empire. Later, he hired many statisticians (he called them computers!) to search for correlations among locations around the globe. Eventually, he identified a southern oscillation across the Pacific basin. The Southern Oscillation is an oscillation of sea level pressure between certain islands in the South Pacific and Darwin, Australia. When sea level pressure is high in Darwin, it tends to be low in, say, Tahiti. When it's low in Darwin, it tends to be high in Tahiti.

British mathematician Sir Gilbert Walker was the first to connect the Southern Oscillation (or SO, a see-saw in atmospheric pressure between the western and eastern tropical Pacific) to cyclic weather and climate patterns far field. Walker became interested in the Indian monsoon during a stint overseeing India's colonial weather observatories at the start of the 20th century. Through painstaking statistical work, Walker linked the SO to seasonal rainfall patterns in Asia, Africa, and South America. His findings were neglected, though, since no physical mechanism for the far-flung effects was apparent. Only in the 1960s did Jacob Bjerknes relate the SO to sea surface temperatures, which paved the way for understanding El Niño and other ocean-atmospheric interactions.

www.ucar.edu/communications/quarterly/spring03/winter.html

When Walker passed away, an obituary appeared in the Royal Academy of Sciences in 1959 that suggested he had essentially wasted his life pursuing correlations between weather events at different distant locations around the globe, and that the writer of the obituary believed that such correlations were spurious, misleading, or nonexistent.

However, had Walker lived another decade or so, he would have had the last word as well as the last laugh. His work on correlations has become the basis for present-day knowledge of El Niño occurrences in the tropical Pacific and several of the weather and climate anomalies that El Niño tends to spawn, not only around the Pacific basin, but around the globe as well.

This example supports my main point: it is valuable for those students new to climate and climate-related research to read the writings of the old guard. Perhaps some of these students of the atmosphere will have the potential to either identify new climate-related hypotheses or to reassess some old thoughts that can now be investigated, thanks to the development of modern research techniques. Using the analogy of TV stories in which newly developed forensic detection methods are used to solve decades-old crime cases, unsolvable then because of inappropriate detection capabilities. Well, new research methods of detection become available each decade, if not each year, and they can be used to re-test hypotheses and thoughts of researchers whose careers have preceded ours.

Arthur, A., 2002: Literary Feuds: A Century of Celebrated Quarrels. New York: MJF Books.

--Michael H. Glantz