Read an Excerpt

The Politics of Earthquake Prediction

PRINCETON UNIVERSITY PRESS

Introduction: Politics and Science

In his seminal 1962 book The Structure of Scientific Revolutions, Thomas Kuhn argued against a simple linear interpretation of scientific progress and gave great attention to "paradigms" and "paradigm shifts." An early reviewer of Kuhn, however, noted that he had used the term "paradigm" in twenty-two apparently different ways. In a subsequent (1970) edition, Kuhn apologized and offered a two-tiered redefinition: A paradigm comprises the basic theoretical tenets, shared beliefs about appropriate models, and common values of a scientific community. A paradigm, however, is also a set of "exemplars." These are model problems with preexisting solutions used to illustrate general laws and to train inductees in the field. The newcomers are then expected to extend and deepen the field and "see [a new] problem as like a problem ... already encountered."

Kuhn's image of a basically conservative and narrow-minded science disturbed scientists and nonscientists alike. As Nathan Reingold noted in 1980: "Kuhn's description of normal science as the norm outraged philosophers and others to whom science was an enterprise forever at the edge of knowledge, undertaken by individuals continually challenging existing concepts. A tradition-bound community was anathema to those believing in scientists as a band of adventurous explorers of the unknown."

Kuhn noted that paradigm shifts did not occur easily. Generations of scientists were often committed to a particular world view, with its attendant set of models and problem-solving techniques. They could be counted upon to resist a new approach which might suddenly render their training and experience obsolete. How a scientific community confronted a potential innovation fascinated Kuhn. Arguing that "there is no area in which more work is so badly needed," he asked a series of key questions: "How does one elect and how is one elected to membership in a particular community, scientific or not? What is the process and what are the stages of socialization to the group? What does the group collectively see as its goals; what deviations, individual or collective, will it tolerate; and how does it control the impermissible aberration?" Basically, these are political questions, but scientists are reluctant to admit that politics plays any part in their professional lives. Of course, the appearance of being "apolitical" or "above politics" is itself a very astute political tactic.

Among the relatively few major studies which focus on "the politics of science," the works of Don Price, Robert Gilpin and Christopher Wright, Stuart Blume, W. Henry Lambright, and David Dickson are noteworthy. Nonetheless, the dominant focus in all of these works is some combination of the following: (1) the evolution of science as a public policy issue, (2) the impact of science or the "scientific way of thinking" on government, (3) the administration of science, especially at the federal level, and (4) the organization of scientific communities to interact with government.

While traces of concern about the internal politics of science can be found in most of these studies, only two works really probe the question that interested Kuhn most: how scientists control challenges to established orthodoxy. A partial attempt was made by science writer Daniel Greenberg in The Politics of Pure Science, but the major effort was by Marian Blissett in Politics in Science.

The core of Blissett's book is three sections on (1) the structural characteristics of research, (2) the social parameters of science, and (3) the political process of science, especially "the emergence and resolution of scientific disputes." Blissett argues that a fundamental tension exists within all scientific communities between conformity to a prevailing consensus and "satisfying one's own creative, and possibly deviant, curiosity." Scientific communities are thus miniature social orders designed to control conflict; Blissett defines the politics of science as "a collective method for selecting and perpetuating ('enforcing') consensual patterns of perception and ultimately ordering them into a convincing, intelligible picture (model) of some aspect of the natural world." While the degree of control varies considerably from discipline to discipline, little doubt exists that each scientific community has "an establishment," what Michael Polanyi called "chief Influentials," who determine research priorities, career escalation, publication possibilities, and often funding flows within programs. As Blissett notes, the function of this "restricted leadership of science" is clear. Their job is "to preserve high standards of inquiry by screening out suspect or divergent judgments."

Echoing Kuhn, Blissett makes a special plea for research on how scientific communities handle challenges to orthodoxy:

Even though the resolution of paradigm conflicts provides an unusual opportunity to see the politics of science from exaggerated dimensions, few, if any, studies have been directed to this end. The techniques of interpersonal adjustment and manipulation surrounding these events must therefore be inferred from sketchy historical accounts. Interestingly, at this level of conflict one can sometimes learn a great deal from unsuccessful challenges to existing paradigms or from the reaction of scientists to theories they regard as distinctly unscientific.

The purpose of this book is to accomplish what Kuhn and Blissett wanted most: a detailed dissection of a major scientific controversy of the modern era. This particular controversy revolves around earthquake prediction, very much on the frontier of current research in geology, geophysics, and seismology. Specifically, the case focuses on a prediction by two U.S. government-employed scientists, Dr. Brian Brady and Dr. William Spence, for a series of great earthquakes to strike Lima, Peru, in the early 1980s. The fact that the prediction, and the underlying theory, were ultimately deemed failures does not diminish the importance of the case. Indeed, it actually matches the Kuhn/Blissett plea for detailed research on defeated challenges and on theories regarded as "distinctly unscientific."

A Prediction Contained, 1976–1979

Background

Great earthquakes tend to be widely separated in space and time. As a result, their cumulative worldwide effects are underestimated. In fact, however, more than a million people have died as a direct result of earthquakes in the past century. Experts agree that the ability to predict major damaging earthquakes even a few days in advance would reduce casualties by at least fifty percent. Such a capability is on the horizon, although it is not as close as was once thought.

Because of rapidly accumulating knowledge of plate tectonics and indications that major earthquakes offered premonitory signs (such as uplift, foreshocks, radon gas emissions), coupled with some spectacular and well-publicized earthquake prediction successes, especially in the People's Republic of China, the 1970s geoscience community was optimistic that earthquake prediction would soon be a regular scientific achievement. In an attempt to identify and explore the various political, economic, legal, and social ramifications of earthquake prediction, the National Research Council of the National Academy of Sciences convened a Panel on the Public Policy Implications of Earthquake Prediction in April 1974. A year later, this group produced a report entitled Earthquake Prediction and Public Policy. The panel reflected the optimism of the time:

Within the past 5 years, many seismologists have become convinced that a new development is imminent, namely, the prediction of earthquakes. By prediction seismologists mean that the place, time, and magnitude of the quake are specified within fairly close limits, with the consequence that accelerated planning to save life and property is possible. Established methods for identifying high-risk areas depend largely on the past incidence of quakes and the mapping of fault structures. The new methods rely primarily on premonitory signs, such as changing physical properties of rocks under stress and surface tilting, that occur in advance of a quake. Prediction capability does not lessen the importance of other approaches to earthquake mitigation, but it adds one potentially telling weapon to the arsenal.

Anticipation of an "imminent" earthquake prediction capability was premature, however. Unfortunately, different fault systems have different earthquake mechanics, and the precursory phenomena are hard to detect and even harder to interpret. Optimism faded in the late 1970s, and a 1978 National Research Council panel reflected more caution, even as it emphasized the societal implications of the (eventual) capability to predict earthquakes: "Geological technology will probably reach a point within the foreseeable future at which scientifically credible earthquake predictions can be made. Constructive use of this new prediction technology will depend to a considerable extent on the accuracy and reliability of our knowledge about how people and organizations will respond to these predictions and warnings. Inadequate attention to the social consequences of using a particular technology may have counterproductive results."

Within the U.S., the United States Geological Survey (USGS) conducts or supports most of the basic geoscience research on earthquakes and earthquake prediction. There was an early (1960s) battle over who would take the lead in earthquake prediction research between USGS and the Environmental Sciences Services Administration (ESSA), which subsequently became the National Oceanic and Atmospheric Administration (NOAA), but USGS emerged the clear victor. The USGS budget for earthquake research grew from a few million dollars in the early 1970s to almost twenty million dollars later in the decade.

The Brady Approach

Between 1974 and 1976, Dr. Brian T. Brady, a research physicist in the Denver center of the United States Bureau of Mines (USBM), which is traditionally competitive with USGS within the Department of the Interior, published a series of four articles in the European scientific journal Pure and Applied Geophysics. In these "Theory of Earthquakes" articles, Brady argued that he had observed a structure in rock failure which was equally applicable to laboratory rock breakage, mine failures, and earthquakes. The Brady approach represented an unprecedented combination of geophysics, microphysics, and mathematics. Brady argued that his theory yielded a "clock" which provided the precise time, place, and magnitude of an earthquake, but only if he had the requisite historical and current seismicity data. More specifically, Brady argued that his "scale invariant inclusion theory of failure" isolated three time-dependent classes of earthquake precursors. Properly identified and interpreted, these precursors provided warning of an earthquake ranging initially from a few years, then to a few hours, and ultimately to a few seconds. The Brady model was deterministic in the sense that as the area under scrutiny moved through the precursor classes, the system "locked in" for failure. In effect, the Brady approach eliminated the need for probability estimates, the centerpiece of the conventional approach.

In the culminating article in his series, "Theory of Earthquakes IV: General Implications for Earthquake Prediction," Brady described the application of his theory to five cases: (1) a successful 1975 rockburst prediction in an Idaho mine; (2) a 1969 earthquake in the Soviet Union; (3) the 1971 San Fernando, California, earthquake; (4) a 1973 earthquake in upstate New York; and (5) two 1974 earthquakes near Lima, Peru. With the exception of the Idaho mine rockburst (Brady's specialty), all of the examples were retroactive applications of his theory. Within his analysis of the 1974 Peruvian events, however, Brady planted the seeds of an earthquake prediction. He noted that the 1974 earthquakes took place in a "well documented seismic gap that is of a size to suggest that it could have supported a much larger magnitude earthquake sequence than what did actually occur." Moreover, the aftershock sequence was unusually short and sparse, and Brady hypothesized that the November 9 event signaled the creation of the necessary and sufficient conditions for "an impending great earthquake ... 75 km off the coast of Central Peru." This statement by Brady was based in part on the work of USGS geophysicist Dr. William Spence.

A personal friend of Brady since the early 1970s and an NOAA geophysicist before his unit was absorbed into USGS, William Spence had been a member of a 1974 USGS aftershock study team in Lima. Spence was familiar with Brady's earlier theoretical work and shared with him some of the seismicity data from Peru. In September 1976 Spence convened a meeting on "Global Aspects of Earthquake-Hazard Reduction" for USGS. Held in Denver, the conference attracted the leaders in earthquake studies, and Brady made two presentations which generated "animated discussion." The combination of Brady's theory and Spence's data was the genesis of the 1976 earthquake forecast for Lima that would become known as the "Brady-Spence prediction."

Initial Contacts

The Peruvian government and scientific community were unaware of the 1976 Brady article. No scientist or administrator in Peru remembers any prior contact with, or notification by, Brady. In late 1976, however, Spence sent a reprint of the Brady article to colleagues at the Instituto Geofísico del Peru (IGP) in Lima. The reaction was swift. Understanding the reaction, however, requires an explication of the milieu for scientific inquiry in Peru.

Peru has such a long mining tradition that the discipline of geology is relatively well developed for a Third World country. In 1922, the Carnegie Institute in Washington established the internationally recognized Geophysical Station (Observatory) at Huancayo. In 1947, Carnegie transferred the station to Peruvian hands, and the Peruvian government created the Instituto Geofisico de Huancayo. The first director was Alberto Giesecke, a former Carnegie staff member and of a family long distinguished in the history of Peruvian education.

In Peruvian hands, IGP developed slowly and maintained the high reputation of the Huancayo operation. In the International Geophysical Year of 1957, however, IGP entered a period of explosive growth, because the equatorial placement and high prestige of the Huancayo station made IGP the logical recipient of more than fifty international research agreements. With the outside funding, staff grew from four to one hundred professionals.

In October 1968, the government of President Fernando Belaúnde Terry was overthrown by the military under General Juan Velasco Alvarado. This was not a "typical" military coup, however, because the leftist (but still military) Velasco government instituted a series of fundamental societal reforms which (1) broke the economic back of the traditional Peruvian oligarchy, (2) brought large sectors of the economy under state control, (3) mobilized whole classes previously excluded from Peruvian politics, and (4) eventually put the media under government direction. Moreover, the military settled in for extended rule, lasting (with a 1975 bloodless coup by General Francisco Morales Bermúdez) until mid-1980. In truth, the period deserves the term "revolution." That many of the reforms would contribute to economic disaster was not understood at the time.

With the advent of military rule, the halcyon days of an autonomous IGP came to a close. IGP found itself slowly strangled by an expanded state bureaucracy which curtailed its autonomy and by losses in real salaries caused by inflation and a collapsing Peruvian economy.

---

Excerpted from The Politics of Earthquake Prediction by Richard Stuart Olson, Bruno Podesta, Joanne M. Nigg. Copyright © 1989 Princeton University Press. Excerpted by permission of PRINCETON UNIVERSITY PRESS.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

---