World-renowned climatologist and ICTP Scientific Council member S. George Philander explores the impact and limitations of science in our modern society.

Science in an Uncertain World

Most people first heard of El Niño in 1997 when newspapers and television gave extensive coverage to devastating floods in California, severe droughts in Indonesia, and strange weather everywhere---all attributed to El Niño.
Today most people are familiar with the term, but few realise that the phenomenon has been with us for millennia, and that originally it was welcomed as a blessing.
The name was initially given to a modest, warm, seasonal current that appears along the shores of Ecuador and northern Peru around December when the accompanying rains transform the barren coastal desert of that region into a garden. El Niño is Spanish for 'the boy' and refers to Child Jesus.
Over the past few decades, even though the phenomenon has remained essentially constant, our perceptions of it have undergone a remarkable transformation. We now regard El Niño as a global hazard that we anticipate with trepidation.
El Niño is an example of the paradox that, as we grow in wealth and in population, so does our vulnerability to natural hazards. Insurance companies find that claims related to damages inflicted by severe storms, hurricanes, floods and earthquakes are rising steeply even though there is no evidence of an increase in the number and intensity of those hazards. For example, today El Niño still turns the desert of Ecuador and northern Peru into a garden, but few people have time to behold that miracle; they are preoccupied with the roads, bridges and houses that are washed away by the rains.

3D view of El Nino phenomenon

To cope with environmental hazards we turn to scientists for predictions. Meteorologists have responded by transforming weather prediction from an augury into a source of reliable information. Attention has recently shifted to the prediction of longer-term climate fluctuations. Progress has been so rapid that, although El Niño caught everyone by surprise in 1982, by 1997 scientists anticipated its arrival in the Pacific several months in advance. This was an impressive achievement, but it carried a most unfortunate blemish.
During the summer of 1997 scientists alerted Californians, on television and in newspapers, of a high probability for exceptionally heavy rains during the winter of 1997-1998 because of a very intense El Niño. Scientists also advised the people of Zimbabwe that rainfall there would probably be below normal. Californians did indeed experience floods, and were prepared, but Zimbabweans had normal rainfall and were unprepared. Because of the expectation that crops would be poor, and that farmers would be unable to pay back loans, financial institutions in Zimbabwe declined loans to farmers. The consequences were dire: Crop production fell 20 percent below normal. The prediction of a drought in effect caused a drought.
The tragedy in Zimbabwe underlines the urgent need to improve communications between producers and users of scientific information. The latter group faces by far the bigger challenge because the decisions it has to make are extremely difficult, even when the scientific information is perfectly accurate. Available resources to cope with a multitude of problems that range from public health to agriculture and education are limited. Any setting of priorities reflects values that depend on race, gender, religion and culture, making the formulation of policies acceptable to everyone difficult. Given the enormous complexity of social issues, we must guard against exaggerating the value of scientific information. Consider the case of the Indian monsoons.
In the second half of the 19th century, poor rains over India often contributed to poor harvests, famines and the death of millions of people. At the time, it was assumed that the solution to the problem required scientific information that would allow accurate predictions of the monsoons.
Today scientists have a much better understanding of monsoons than they did 100 years ago, but they are still unable to predict accurately when monsoons will arrive or, for that matter, will fail to arrive. Nonetheless for several decades now, there have been no disasters comparable to the famines of the 19th and early 20th century. Of prime importance to this impressive achievement are political changes in India. Development and implementation of effective policies to cope with famines started once India became an independent country and acquired a democratic government. Science has played a secondary role.
Although much progress can be made on certain environmental problems even in the absence of scientific information, we need to acknowledge that such information can be valuable, especially if it is accurate. For example, government officials now act swiftly in response to predictions that a hurricane will make landfall. As recently as World War II the admiral of a fleet in the Pacific chose to ignore the prediction of a typhoon because of a lack of confidence in the prediction. His fleet suffered enormous losses. Because of advances in weather prediction such a decision on the part of an admiral would be unthinkable today.
Progress in weather forecasting comes from improved measurement networks, better understanding of such weather phenomena as fronts, tornadoes, hurricanes and cyclones, and frequent tests of the theories and models of meteorologists. Of central importance is a balance between science for the sake of science and science for practical purposes.
Finding a similar balance in the case of El Niño studies is proving difficult. During the cold war, scientists studying El Niño enjoyed considerable freedom to explore a beguiling phenomenon mainly for the sake of understanding it. They made rapid progress and, after the exceptionally intense El Niño of 1982, could explain what had happened. However, they failed to sound a public alert beforehand, in part because there was no pressure to do so. Sponsors of scientific research subsequently insisted on useful results. This emphasis on science for practical purposes contributed to a significant increase in the number of scientists involved in social and policy issues related to environmental problems. In principle, this is a laudable development, but in practice it is fraught with dangers, as the tragedy in Zimbabwe illustrates.
For science to flourish, scientists must have a sceptical attitude towards their own results, constantly questioning and testing apparent solutions to problems. When trying to persuade potential clients that those results are useful and merit attention, they have to adopt a very different attitude, one of confidence in their results. This is not a problem in the case of weather forecasting for tomorrow because the accuracy of the forecasts can be checked immediately. However, it is a serious dilemma in the case of predictions related to infrequent environmental phenomena that develop over months and longer.
El Niño forecasts have huge uncertainties at present. Is such information of any value at all? Consider the daily weather forecasts that the British government started making available to the public in the 1860s. Scientists objected and forced the government to halt those forecasts on the grounds that the predictions had huge uncertainties and were not based on any understanding of weather phenomena. After a public outcry, the forecasts were resumed because they were proving useful, especially to those involved in coastal shipping. The forecasts significantly reduced the number of shipwrecks, and therefore saved many lives. Scientific information with large uncertainties can be of considerable value, but such an evaluation is best made by the public, not by scientists.

S. George H. Philander

The worlds of science and of human affairs are radically different. Each scientific problem, in principle, has a well-defined solution that can be found by means of universal methods independent of race, gender or religion. A social problem, by contrast, has a multitude of solutions, each with advantages and disadvantages that are weighed differently in different cultures.
The methods and skills required to solve the problems encountered in the cold, uncompromising world of science have very limited applicability in the heated, nuanced world of human affairs where compassion is a virtue and compromise a requisite.
Scientists live in both worlds and thus face a dilemma. The most effective way to minimise this quandary is to reduce the uncertainties in scientific results. To achieve that goal the producers and also the sponsors of scientific research should strive for a balance between science for the sake of science and science for practical benefits.

S. George Philander
Princeton University,
ICTP Scientific Council

This article includes excerpts from the author's recent book, Our Affair with El Niño; How we transformed an enchanting Peruvian current into a global climate hazard (Princeton University Press, 2004.)

Back to Contents Forward to Features

Home