Book Review
March 2007
by Carol Van Strum
Useless Arithmetic: Why Environmental Scientists Can't Predict the Future, by Orrin H. Pilkey & Linda Pilkey-Jarvis. Columbia University Press, 2007. $29.50
Bellwether, by Connie Willis. 1996. Bantam Books, $7.99, paper.
The Futurist, by James P. Othmer, 2006. Doubleday, $23.95. Paperback release June, 2007, $13.95.
"The future, wave or no wave, seems to me no unified dream but a mince pie, long in the baking, never quite done."
E. B. White, One Man's Meat, 1940, 1950
|
The oracles of Delphi or the benign forecasts of the Old Farmer's Almanac may seem quaint anachronisms in the Age of Science and Technology, but the fact is that from regulatory policies to political and military strategy, our society relies more than ever before on prognostications that are no more accurate than chicken entrails. The difference is that we rarely, if ever, see the guts of predictions that dictate our lives. Exposing them is a job for both science and satire.
* * * * * *
REVIEW
Useless Arithmetic: Why Environmental Scientists Can't Predict the Future, by Orrin H. Pilkey & Linda Pilkey-Jarvis. Columbia University Press, 2007. $29.50
A far more apt title for this invaluable little book would be Fraudulent Arithmetic. It is essential reading for anyone challenging, studying, or proposing environmental policy in any field, and should be required reading for media and commentators before they accept numerical predictions on any subject (e.g., Iraq will produce six million barrels of oil a day within two years of our invasion).
Without resorting to a single equation, the authors, both geologists, expose the shameless numerology that masquerades as science in mathematical models supporting hazardous, lethal, and sometimes insane environmental projects. In tales reminiscent of the "magic pencil" studies supporting pharmaceutical and pesticide registrations, Useless Arithmetic demonstrates how blind, unquestioned faith in mathematical models has supplanted hard data and field work, hoodwinking the courts, the public, and decision-makers into approving colossal blunders with deadly, long-term consequences.
The collapse of the North Atlantic cod fishery dramatically illustrates how "the mystique of models and the mystery and apparent sophistication of mathematics" seduce decision-makers and blind them to reality. The primary cause of the codfish collapse was reliance on mathematically-modeled maximum sustainable yield numbers in setting catch limits, on the assumption that fish always produce a "harvestable surplus" that can be taken perpetually without affecting total fish stocks. Whether such a surplus existed and if so what size it was, no one troubled to find out. Year after year, decade after decade, while biologists and fishermen themselves warned of declining fish populations, Canadian fishery managers - under pressure from industry and government - ignored reality in favor of prediction models built on best-case assumptions, invented numbers, and absurd oversimplifications. Suddenly, in 1992, the North Atlantic cod catch, which had totaled nearly two million tons in the late 1960s, collapsed; there were, quite simply, no more cod. More than a decade after this monumental disaster, there are still no cod -- yet according to the models used for setting "sustainable" catches, the North Atlantic is still full of cod!
From toxic mining sites to beach erosion, similarly flawed models have persuaded managers and decision-makers to proceed with projects that defy common sense, physical law, and direct observation. Driving such models is regulatory demand for quantitative predictions - i.e., hard numbers - that can be plugged into risk-benefit equations. Because natural processes involve so many variables, most of them unknown and many immeasurable, quantitative models of natural systems are necessarily constructed from estimates (read guesses) that are routinely fudged, tweaked, discarded, or invented to serve the purpose of the modeler.
"Today's scientists have substituted mathematics for experiments," wrote Nicola Tesla in 1934, "and they wander off through equation after equation and eventually build a structure which has no relation to reality."
With this prescient quote from Tesla, the authors introduce the flawed models supporting construction of a vast nuclear waste repository inside of Yucca Mountain in Nevada. Equally flawed, they point out, were the assumptions behind the development of nuclear power and weaponry in the first place: chiefly the assumption - blithely repeated to this day -- that technology would solve the problem of how to dispose of thousands of tons of deadly radioactive wastes whose long-term health hazards "far exceed the known duration of civilization." A 1987 environmental assessment of suitable dump sites for this waste, conducted in a highly charged political battle, settled upon Yucca Mountain as the only suitable location. Since then, $4 billion have been spent devising models that meet EPA's court-ordered requirement that the repository be leak-proof for up to a million years (a preposterous standard, considering that humanity has yet to design a septic tank that is leak-proof for even fifty years).
The $4 billion Yucca Mountain model, the authors note, is actually a cluster of models. "There are 13 comprehensive mega models of model clusters based on 286 individual models. Clusters include models predicting future climate, infiltration, percolation, and water behavior in drifts. There are thousands of input parameters, hundreds of thousands of lines of equations in hundreds of computer codes, and hundreds of linked mathematical models in the system: complexity built upon complexity, assumption built upon assumption."
The whole elegant, artificial structure rested not on real data but on guesses, assumptions, and made-up numbers, most notably the absurd assumption that groundwater would percolate through the rock no faster than half a millimeter per year. This model house of cards collapsed when construction of the tunnel began, allowing real-world, physical sampling of water percolation rates. Instead of the 0.5 millimeter per year assumed in the models, it turned out that water actually moved through the rock at 3,000 millimeters per year, completely undermining the entire hydrologic model for Yucca Mountain.
Not to worry, though! In the wonderful world of numbers, the Yucca Mountain modelers simply voted in a new percolation rate number to preserve the model - and added titanium drip-shields to protect the already-begun construction.
"Applied models," the authors conclude, "out there in the midst of politically sensitive societal issues, are easily moldable to favor a cause, be it a cost-benefit ratio, an environmental impact statement, the cause of a disease, the size of a fish population, or a prediction of future hazard potential. Add a fudge factor here and tweak the model there, and you have the 'correct' answer. And the alterations are invisible to the managers who use the models."
Instead of the virtual, impossible "reality" of quantitative models, Useless Arithmetic encourages the use of qualitative models, which track trends and directions, rather than impossible certainties. "Accurate predictions of future climates, sea level changes, shoreline erosion rates, and fish populations should be recognized as impossibilities," the authors urge. "We will simply have to move into a more qualitative world."
* * * * *
Bellwether, by Connie Willis. 1996. Bantam Books, $7.99, paper.
The Futurist, by James P. Othmer, 2006. Doubleday, $23.95. Paperback release June, 2007, $13.95.
The "more qualitative world," however, is no less vulnerable to commercial exploitation, political pressure, and corruption than its mathematical counterpart. Qualitative modeling, or trend forecasting - today's manifestation of the Delphic Oracle - has even acquired its own professional status, known in recent years as Futurism, which serves as well-paid soothsayer to corporate marketers and media. A serious danger, beyond their obvious satire potential, is that trend-forecasting models can - and often do -- exert considerable influence on the future they predict, to the point of becoming self-fulfilling prophesies.
The absurdity of selling a pre-ordained future can best be expressed in fiction, because no one would believe it otherwise.
Connie Willis's Bellwether is a guided tour through the bottomless depths of such absurdity in HiTek, a well-funded corporate think tank, where sociologist Sandra Foster studies the origins and causes of fads. "HiTek would like nothing more than to know what causes fads so they could invent the next one," Foster says.
HiTek itself owes its very existence to the futurism fad, however, as well as embodying every corporate fad to come along, from sensitivity exercises to sixty-eight-page Improved Funding Allocation Application forms for paper clips, software or macaque monkeys. With whimsical inevitability, as corporate structure succumbs to its own complexity, Foster's fad research - hula hoops, bobbed hair, corn rows, angels, fairies, latté, aromatherapy, backpacks, tattoos, coonskin caps, mesmerism, duct tape accessories -- meshes with the information dispersion study of a chaos theorist whose request for monkeys has resulted in a shipment of sheep.
While Foster tracks fads and battles for paper clips, the sheep perversely develop a collective mind of their own. Daily routines at HiTek increasingly become a madcap adventure in entropy, instabilities, feedback, iteration, and other factors that make natural systems so wildly unpredictable - and human attempts to model them so laughable. Had North Atlantic fishery managers or the Yucca Mountain designers read The Bellwether before creating their absurd models, disastrous boondoggles might well have been avoided.
The Futurist further confounds the field of trend-spotting by adding the cult of personality to the mix. Its protagonist, Yates, is a hugely successful, world-renowned futurist, a self-described "bona fide A-list player in the culture of expectation."
Yates owes his phenomenal success not to the accuracy of his predictions, however, but to how well he tailors the future to his clients' needs and creeds1. Flying first-class to deliver a keynote speech at the Futureworld Conference in Johannesburg, South Africa, the fact that his high-priced predictions are "utter and complete bullshit" confronts him with a double whammy: he learns that his long-term girlfriend has left him for a sixth-grade history teacher, and along with the entire world he watches the slow demise of the celebrity space tourists whose malfunctioning space hotel Yates had shamelessly promoted.
In a fit of inebriated, maudlin remorse, Yates rewrites his Futureworld speech, denouncing the fraud of futurism and his role in it. "We are not innovators," he proclaims. "We're fucking abominations. … I am the founding father of the Coalition of the Clueless." This attempt at career suicide backfires spectacularly, however. Yates is hailed as a hero for turning truth into a new marketing gimmick. Worse, an unnamed government agency blackmails him into a world tour to manipulate global attitudes toward the United States, his ultimate destination a starring role in the PR tragedy of Iraq.
Between the scientific world of Useless Arithmetic and the absurdities of satire, the idea of predicting the future remains, as ever, a joke, albeit an expensive and sometimes deadly one. "The hot dessert this fall will be pineapple upside-down cake," Sandra Foster pronounces, wrapping up Bellwether. "There will be a sharp upswing in significant scientific breakthroughs, and chaos, as usual, will reign."
Amen.
1"Given Othmer's background in the advertising industry," says futurist Peter von Stackelberg, "perhaps it should come as no surprise that he draws from the unfortunate intersection of marketing and trend-watching, a combination many professional futurists consider the bastard child of futures research." For a more nuts and bolts description of futurism, see von Stackelberg's web site: www.futureswatch.org.
* * * * *
To Top of Page
BOOK REVIEWS
