False Presumptions


Pandora’s Poison: Chlorine, Health and a New Environmental Strategy

It all started with electricity. Just before the turn of the century, a German scientist discovered a way to split the salt molecule into its constituent components, sodium and chlorine, by passing an electrical current through ordinary saltwater. He was after the sodium, used to make alkali (more commonly called caustic soda), a key building block in the manufacture of glass, paper, textiles, soap, and many other products. Electrolysis proved to be a much more efficient means of manufacturing alkali than any previously known method, but it also had one significant drawback. For every measure of alkali produced, the new process generated a roughly equal amount of something new under the sun: chlorine gas. Unlike naturally occurring chlorine, which is found in a variety of ubiquitous and harmless salt compounds, this unwanted gaseous byproduct was deadly, difficult to dispose of, and had no known uses. Thus was born a foreboding conundrum of Europe’s new industrial economy: science had stumbled upon a seemingly promising technology, but one which it could not completely control. What to do? The dubious wisdom of the market prevailed. The potential profits from the worldwide demand for caustic soda were too great; if this uniquely deadly byproduct, this destroyer of life, had no uses, then uses would have to be found for it.

From this inauspicious beginning, the modern chlorine industry was born. Today there are forty-two plants in the United States producing chlorine gas and caustic soda, using essentially the same method developed by the Germans one hundred years ago. (Twelve of the largest plants, accounting for over 70 percent of domestic production, are along the Gulf of Mexico in Texas and Louisiana.) And the Faustian dynamic that drives the industry remains the same: the production of the much more valuable caustic soda is still limited by the availability of markets, or “sinks,” for dumping chlorine gas. Thus the first boom in production of chlorine (and, therefore, caustic soda) came with the onset of World War I, when vast quantities of the heavy, green gas were ordered by the military, to be made into horribly effective munitions. There followed in the interwar years a litany of inventive uses, many of which would prove to be no less deadly: from Monsanto’s marketing of PCBs (now banned) as insulating fluids in 1929, to Du Pont’s ill-fated production of CFCs as a new type of refrigerant in the early thirties, to the use of DDT as an insecticide just prior to World War II.

After decades of use in this country, these three chlorinated compounds were all demonstrated to be human carcinogens, and these are just the tip of the iceberg. Thousands of chlorine-containing chemicals, referred to collectively as organochlorines, are now used in a plethora of applications, including PVC plastic (by far the largest single use of chlorine in the U.S.), paper processing, solvents, pesticides, and water treatment. Unlike the pharmaceutical industry, where rigorous toxicity testing prior to approval is the rule, in chemical production the chemicals are treated like human beings on trial: innocent until proven guilty. Of the roughly 11,000 organochlorines used in industry, only a very small fraction have ever been tested for their toxic effects on humans and the environment, yet virtually every one that has been tested has been proven to cause severe health effects. In effect, the chemical industry, which exploded after World War II, has for the last sixty years conducted the largest and least controlled experiment in toxicology in human history, with all of humanity (and the rest of the animal and plant kingdoms) as the subjects. There is no uncontaminated control group in this experiment. Organochlorines have been borne by wind and water to the far reaches of the globe. Because they are slow to degrade and prone to bioaccumulate, a toxic mixture of organochlorines has now been detected in the fat of polar bears at the North Pole, in the tissue of unborn infants, and at the bottom of the ocean.

This is the morbidly fascinating history of the industry, illustrated with unsparing thoroughness by biologist Joe Thornton in Pandora’s Poison. But Thornton’s book is much more than an explanation of the problem. Thornton is a former research director for Greenpeace, and his lifetime of research and expertise has produced what will surely be considered the definitive text on this subject for a generation. This is a staggeringly well-researched volume, impressive not just for its depth but for its breadth. Beginning with a detailed account of how chlorine is made, Thornton provides thoroughly documented chapters on what makes it so chemically dangerous (delving into a brief lesson on subatomic chemistry), the growing evidence of the link between organochlorines in the environment and this nation’s cancer epidemic (cancer now kills one in four in the U.S., up from one in five in the 1950s), and a close look at the way the chlorine industry operates – including the increasing reliance on production for third-world consumption as U.S. markets for chlorinated products stagnate. For every banned application, two new ones spring up, as the industry strives to maintain the all-important “chlor-alkali” balance. Along the way, Thornton takes on the limitations of toxicology and epidemiology, the epistemology of science, the role of scientists in a democracy, and much more. It is both a far-reaching compendium of what is known about the dangers of organochlorines, and a compelling call for a reassessment of the very way we think about protecting public health.

Throughout, the book provides a devastating critique of the way we currently assess an industry’s impact on our world, based upon a model Thornton calls the Risk Paradigm. The Risk Paradigm begins with the assumption that some exposure to toxic chemicals is inevitable in industrial society; thus the proper task for regulators is to determine acceptable levels of exposure, or “thresholds,” for each single pollutant. Working backward from this determined level, regulators then calculate the maximum allowable release rate for a given facility that will still fall within the threshold of contamination for its neighbors. Recent evidence on a variety of chemicals including dioxin (the most toxic class of organochlorines), however, has suggested that there are no acceptable levels of exposure for some substances; that in fact, adverse health effects, including endocrine disruption, development problems, and even cancer, can follow exposure to the smallest traces of these toxins. The history of lead regulation is a telling example of how thresholds have failed regulators and public safety. The original consensus among toxicologists as recently as the 1960s was that low or “background” levels of lead found in the human bloodstream were benign. As new evidence of the effect of lead on childhood development accumulated over the next twenty years, the acceptable threshold was lowered again and again, until it finally reached zero.

Yet the most egregious shortcoming of the risk control method is that it views each chemical and its source in isolation, failing to account for the cumulative pollution burden in the environment. “Thousands of individual facilities, each discharging the ‘acceptable amount’ of thousands of different substances, together produce a cumulative global impact; the current system, focused only on the local parts, is and always will be blind to this problem of the whole.” Thus, Thornton notes, a facility with a somewhat taller smokestack, distributing its waste over a much broader area, appears under this paradigm to be more protective of public health than one with a shorter stack, which has a more demonstrable contaminating effect of people in the immediate vicinity. That pollution must fall somewhere, of course, increasing somebody’s exposure; for the purposes of the regulatory permit in question, however, it doesn’t “count.” Neither do the hundreds of organochlorines, introduced into the air and water as byproducts in various chlorine processes, that have never even been identified, much less assessed for toxicity. Nor does the Risk Paradigm, with its chemical-by-chemical testing in the laboratory, account for interactions between chemicals we are exposed to in the real world.

Thornton advocates a new model for environmental policy, termed the Ecological Paradigm, which focuses on preventing pollution rather than managing its harms. “The new framework is founded on the view that ecosystems and organisms – and society, too – are extraordinarily complex and dynamic systems in which innumerable parts are connected in webs of interdependency, multiple causality, and feedback loops, all of which change over time.” Science is limited in its ability to comprehend how individual chemicals act on natural systems, thus “we should avoid practices that have the potential to cause severe damage, even in the absence of scientific proof of harm.” This is the precautionary principle: “We should err on the side of caution when the potential impacts of a mistake are serious, widespread, irreversible, and incompletely understood, as they are with the hazards of global toxic contamination.”

It sounds modest enough, but the implications are enormous. Because the addition of chlorine to an organic compound virtually always increases toxicity, persistence, and tendency to bioaccumulate, Thornton advocates assessing chlorinated compounds as a class, rather than chemical by chemical, which would take hundreds of years and may yield inconclusive results in any case. (Who can ever conclusively say which of the many poisons in our environment has made someone sick?) Taken together, in the view of Thornton and a growing number of biologists and public health experts, the accumulated evidence against chlorine is damning enough to reverse the onus: any chemical in this class should have to be proven safe before it can be released into the environment. Over the next generation, existing uses of chlorine should be phased out through the use of substitutes, in a process known as a chemical sunset. As Thornton carefully outlines, safer and in many cases cheaper alternatives exist for everything from PVC plastic to water treatment, and many companies and municipalities in the industrialized world have already begun to slowly wean themselves from chlorine. Improved efficiency and recycling, meanwhile, have the potential to greatly reduce the demand for caustic soda, and alternative sources, as well as other alkylating agents (such as lime), are available.

As the anti-chlorine movement gained steam in the early 1990s, the industry, acting through its various flack groups (including the Vinyl Institute and the Chlorine Chemistry Council), launched a counterattack, couched in the authoritative language of science. (And some in the language of tabloids: “The end of chlorine would spell the end of modern civilization itself,” the industry-funded Competitive Enterprise Institute announced in 1996.) The industry rejects calls for chemicals to be judged as classes, calling instead for judgment to be made on the basis of “sound science” (not coincidentally, a term Governor Bush has begun using more and more in recent months), and predicts dire circumstances for the economy in the event of a phaseout. As Thornton demonstrates, in industry public relations literature the buzzword “sound science” has become shorthand for chemical-by-chemical, quantitative-risk assessment. But after reading Thornton’s devastating critique, we might reasonably ask what, if anything, is “sound” about this approach. As Thornton writes:

This framework assumes, contrary to science’s actual findings, that ecosystems have assimilative capacities for substances that persist or bioaccumulate, that toxicity thresholds exist and are discoverable for all toxic effects, that the hazard posed by the total burden of chemical exposure is the sum of the effects of each individual compound in isolation, that risk assessments focused on human effect predict the potential impacts on all the diverse species and ecosystems in nature, and that pollution control technologies are an effective way of preventing chemicals from getting into the environment. The Risk Paradigm steadfastly ignores what we have learned about these assumptions from experience, observation, and experiment: that they are false.