| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
Plant Physiology 133:3-9 (2003) © 2003 American Society of Plant Biologists How Should Society Approach the Real and Potential Risks Posed by New Technologies?1University of California Riverside, Riverside, CA 92521
As the 21st century opens, two new technologies raise the promise of benefits for humankind—biotechnology and nanotechnology. Crop biotechnology promises new and improved foods, e.g. vitamin A-enhanced rice (Oryza sativa), more food at lower cost from highly productive plants, plants that will grow under adverse—highly saline or drought—conditions, plants that will reduce reliance on potentially poisonous pesticides, or even plants that will produce pharmaceuticals. Human biotechnology promises cures for genetic diseases in somatic cells, diagnoses of diseases before they manifest themselves into deadly or irreversible conditions, e.g. breast cancer, perhaps even the prevention of inheritable genetic conditions that could be corrected by genetic engineering of the germ cells. Biotechnology is well begun; nanotechnology has yet to penetrate public awareness or even university awareness beyond the major research universities. Nanotechnology, the science of the almost vanishingly small, is concerned with chemical products one-billionth of a meter in diameter. Yet, it, too, promises treatments for disease, innovative products to clean water supplies and protect the environment, and ultrasmall electronic switching mechanisms and components.
The early stages of a science or technology are replete with promises of benefits to humankind or the environment. Promised benefits, even soaring visions, motivate researchers to develop technologies, agencies to fund research, firms or venture capitalists to invest in, and ultimately the public to begin to accept or even to embrace, the technologies. Without promises of benefits, the science and technology will not develop. Yet, promises can be merely speculative, overblown, mistaken, or nonexistent. For example, the U.S. Food and Drug Administration withdrew approval for Parlodel (chemical name: bromocriptine), a breast milk lactation suppression drug, because it appeared to cause heart attacks and strokes, and its effects were "highly questionable" (Department of Health and Human Services, Food and Drug Administration, 1994 New technologies inevitably will pose some risks as well—perhaps to human health, e.g. breathing nanoparticles may be as dangerous as breathing tiny air pollutants; to other human endeavors, e.g. crops infected by unwanted genes; or to the environment that sustains human life. How should we think about risks posed by promising new technologies? One place to begin is with a brief review of the chemical revolution that began during and immediately after World War II; what can one reasonably conclude from this (ongoing) technological experiment? Beyond that, which risks are more and which less acceptable, and what considerations bear on these issues? Where should we locate the acceptability of risks posed by new technologies compared with other risks? What is the context into which the new technologies are introduced? Is it reasonable to suppose that the technology will exacerbate or ameliorate existing problems? What institutional strategies should we adopt to address risks that are less acceptable? The essay that follows focuses on genetically modified (GM) crop technologies. Addressing other technologies may need various modifications because they might raise different issues.
Without question, the post-World War II chemical revolution has produced substantial benefits for humankind. A significant portion of the gross domestic product in many countries results from the chemical industry and its products. It also has improved the quality of life and produced lifesaving products. It enhanced food production worldwide and helped make the United States a leading exporter of foodstuffs to the rest of the world.
The revolution has not been without costs, however. High-profile human health and environmental harms (and risks of others) are on the other side of the ledger. DDT (1,1,1,-Trichloro-2,2-bis(p-chlorophenyl)ethane), a "miracle" pesticide that killed malaria-carrying mosquitoes, was found to have substantial adverse ecological and health effects. Freon replaced highly toxic compounds that caused accidental deaths from use in home refrigerators, but this nontoxic and nonflammable substance has been found to destroy ozone in the upper atmosphere. Dioxins, at first contaminants of herbicide and pesticide products (now eliminated) and currently the contaminants of many industrial processes, are among the most potent mammalian carcinogens known. Polychlorinated biphenyls (PCBs), because of their thermal stability, their resistance to many chemical reactions, and dielectric properties, were used as hydraulic fluids, lubricants, plasticizers, insulators, and fillers in a variety of other products. PCBs became of concern because of their lipophilic and bioaccumulating properties, which transported them up the food chain, increasing their concentrations and toxicity as they moved to higher organisms. More recently, scientists have discovered PCBs can be transported far from their original sources by means of evaporation and condensation into colder areas of the Earth, contaminating fish, mammals, and humans to near-toxic levels at these remote locations (Travis and Hester, 1991 Apart from toxicity properties resulting from direct exposures, many chemical substances have been disposed of in ways that have contaminated soils, groundwater, drinking water, the air, and natural ecosystems. In addition to toxic waste dumps, air and water emissions of potentially toxic substances have created experiments with the environment and public health without scientific understanding of long-term consequences of such exposures.
Of serious concern is our profound ignorance about the effects of most of the substances that are registered for use in commerce. Even more serious is that society at large doesn't even know that we are ignorant! There are about 70,000 substances registered for use in commerce. Add to that another 30,000 metabolites and derivatives of these substances, and the total comes to about 100,000 chemical substances. About 23% of these are polymers and another one-third present little or no exposure, in both cases presenting at worst quite minimal risks. Another 800 to 1,000 substances are added to commerce each year with no or only minimal testing (U.S. Congress, Office of Technology Assessment, 1987
There are various estimates of what is known about the toxicity properties of the compounds, but none is reassuring. As recently as 1998, there was little or no basic toxicity information in the public record for 75% of the 3,000 chemical substances produced in the highest volume in the United States (Environmental Health Letter, 1998
A more detailed breakdown of this problem is as follows. In 1984, the U.S. National Research Council (NRC) found that there were 12,860 substances produced in volumes exceeding 1 million pounds per year, and for 78% of these, there was no toxicity information available, 13,911 chemicals produced in volumes of less than 1 million pounds (76% with no toxicity data), 8,627 food additives (46% with no toxicity data), 1,815 drugs (25% with no toxicity data), 3,410 cosmetics (56% with no toxicity data), and 3,350 pesticides (36% with no toxicity data) (NRC, 1984
What can we infer from these risks and harms resulting from the post-World War II chemical revolution? If nothing else, this picture should occasion considerable humility for human ingenuity, scientists, technologists and institutions of social control. Entrepreneurs have been quite innovative in developing products but much less successful in understanding their risks and preventing them. Even now, it is not clear how well the downside of this technology is understood because it appears we have not yet run out of surprises. For example, polybrominated diphenyl ethers (PBDEs) used as fire retardants in furniture, plastics, and many electronic products appear to possess most of the same bioaccumulative and toxicity properties of PCBs (Hooper and McDonald, 2000
As scientists develop more subtle understandings of adverse effects and their mechanisms, the substances of concern tend to increase—consider, for example, endocrine disrupters and recent evidence of carcinogenicity mechanisms (Colborn et al., 1996
What institutional controls and incentives have led to such extreme ignorance of the substances in question? Broadly speaking, there are two kinds of laws to regulate risks: premarket screening strategies try to provide some assessment of the risks humans and the environment from products before they enter commerce, whereas post-market strategies provide for regulation of products after they have been introduced into commerce (and, thus, into the environment) and the public exposed (U.S. Congress, Office of Technology Assessment, 1987  ). The best postmarket statutes would aim to use surrogates of human health or environmental harm to identify risks before they cause harm. However, use of these surrogates has been eroded over time by requirements for evidence of actual harm, thus removing the possibility of warnings before damage occurs (Cranor, 2003a). Premarket screening strategies have the potential advantage of greater safety than post-market statutes but with the disadvantage of burdening product development with higher costs and slowing the time from discovery of useful products to commercial production. Post-market strategies place lesser burdens on innovation but pose greater risks to safety. The history of the regulation of manufactured chemicals, their by-products, and pollutants suggests that most were subject to post-market regulation.
Virtually all post-market strategies create incentives for firms to develop products while ignoring or inadequately testing for potential adverse effects (or ignoring the byproducts or pollutants of production). Firms typically are not legally required to test substances extensively before they enter commerce (minimal data reporting is required under the Toxic Substances Control Act, a premarket notification law). Whether they do so depends upon a calculation of what is in their best self-interest. Would testing help them avoid regulatory action or personal injury suits? If potential adverse effects arise, ordinarily it is a governmental agency that must identify the problem, require needed toxicity data, and initiate regulatory proceedings to reduce exposure or remove the product from the market. While adverse health data and regulations are being developed, the firm can merely play "defense" and argue that there is insufficient scientific information (or too much ignorance and uncertainty) to justify taking regulatory action. Moreover, a firm's failure to do adequate testing in the first place means that it has even more time to benefit from the product while toxicity data are being developed. Similar problems attend personal injury (tort law) suits, with the burden of proof falling on the injured party (Cranor, 2003a
Scientific procedures and burdens of proof reinforce regulatory burdens of proof. Substances are assumed to have no properties whatever until established by data and theories. Scientific standards of proof and burdens of proof can be interpreted so stringently that it becomes very difficult to satisfy them. Moreover, some scientists assume that substances have no adverse effects until proven otherwise by overwhelming evidence. Even if standards of proof are not so stringently interpreted, conducting the tests, interpreting them, and coming to sufficiently firm conclusions to satisfy scientific advisory panels, regulatory bodies, and appellate courts (that review the regulatory actions) are costly and time consuming (Cranor, 1993
Substantial ignorance about the toxicity properties of substances is of course the expected result of postmarket regulatory schemes. This problem has now reached such proportions in Europe that the European Union has moved to require testing of 30,000 substances in commerce and to severely restrict 1,500 of the most hazardous substances (Loewenberg, 2003 Premarket screening statutes that require substantial testing ensure that there is no or very little health and environmental exposure to substances until an agency is satisfied that there is no legally specified level of risk from them and permits them into production and commerce. With sufficient agency review and approval authority, there is an independent body to assess the quality of health and environmental data and to help assure that the substance does not enter commerce if it presents unreasonable risks to health or the environment.
However, premarket screening laws, such as U.S. statutes for the approval for new drugs, do not always function well. Sometimes firms deliberately or negligently withhold information from the Food and Drug Administration (but at least this opens them to liability). In 1984, there were no toxicity data for one-fourth to one-third of all drugs and pesticides (both subject to premarket laws; NRC, 1984
Any regulatory structure needs to be appropriate to the risks in question and to their acceptability. Users and others exposed to risks from large machines, such as tractors, need much less protection than from risks posed by chemicals. To most people, the risks from tractors or other typical physical threats are more acceptable than the risks posed by many chemicals (Slovic, 1987 ). What accounts for these differences, and where do GM crops fit in? Are they more similar to chemicals or machines?
Risks from GM crops include, but are not limited to, risks from the movement of genes (e.g. increasing weediness, risk of extinction of local species, and increased pesticide resistance), from whole plants (e.g. threats to wild relatives), to nontarget organisms (e.g. adversely affecting other plants, beneficial insects, or soil organisms) and of causing resistance evolution (e.g. herbicide-tolerant weeds; NRC, 2002
Risks tend to be more acceptable the easier they are to identify, personally detect, appreciate and avoid (Cranor, 1995
Technologies that have direct personal benefits to those bearing the risks or that are a critical part of one's plan of life also tend to be more acceptable to persons from a generic point of view. We take airplanes although we know there are risks involved. Piloting airplanes is riskier than driving tractors, but pilots embrace such risks because they are important to their life plans. We drink chlorinated water, which may carry a low-level risk of contributing to bladder cancer, but that risk is counterbalanced by direct disease-preventing and lifesaving benefits provided by chlorination. Similarly, x-rays taken to try to detect disease and even life-threatening illnesses pose some risks of causing lung or bone cancer, yet such risks are acceptable precisely because of the important benefits accompanying them. Prescription drugs, created because they have beneficial effects, can have downside risks, yet if these are not too extreme and the benefits great enough, the risks associated with them may be acceptable simply because of the direct benefits to the users and absence of more benign alternatives (Cranor, 1995 Those exposed to risks from tractors have chosen to use them and have considerable continuing control over whether the risks posed materialize or not. Moreover, tractors for the most part do not pose risks far from their present location, with a few exceptions, e.g. their contributions to global warming and any health risks from engine exhaust. All of these considerations tend to make risks presented by tractors acceptable to those exposed.
By contrast, risks, for example, from chemical substances (Fig. 2) and GM crops, tend to be invisible, not easily detected, and might remain hidden for a long time. These tend to make the effects difficult to identify and detect, to protect against, and even to avoid (Cranor, 1995
The features of the risks from chemicals and GM crops tend to place such risks toward the more unacceptable end of risks to which persons will be exposed (Cranor, 1995
Another generic concern for evaluating emerging technologies would be the context into which they are introduced. Is the system or environment into which the technology enters robust and resilient or in poor condition and perhaps vulnerable to adverse perturbations? With regard to GM crops, which could potentially have considerable impact on the farming and natural environment, what is the condition of the world into which they will be introduced? At the most generic level, for example, is the natural environment into which GM plants are introduced more like an unlimited, bounteous frontier that is quite resilient, or more like a limited, confined fish bowl that is already substantially polluted and more susceptible to new insults?
Scientists, largely ecologists, are far from optimistic in their assessment of the current condition of the natural environment. They see the environment as being under considerable pressure, continual threat, degraded from previous healthier states, which will only worsen because of increasing human population pressures. Consider a few highlights (readers may have their own shorter or longer lists; Cranor, 2003b In the context into which new technologies will be introduced, what effect is it reasonable to suppose that technology will have on it? Might it ameliorate, exacerbate, or have more neutral effects on existing problems? Transgenic plants might successfully address some of the food shortages and shortage of adequate agricultural land in the future (e.g. transgenic plants might be created that would grow on degraded soils) and even the shortage of water (with the development of drought tolerate plants), although not the underlying problem of a population that may be too large for the world's resources.
With respect to GM crops, the U.S. NRC (2002
Thus, because scientists do not yet understand well either the impacts of introduced species on ecosystems or genetics, and because it will take considerable time to develop the understanding to fully assess such risks, it appears that regulatory decisions about the introduction of GM crops will be made under substantial uncertainty and in considerable ignorance (NRC, 2002
What do the combination of risks, the nature of the technologies, the context of introduction, and lessons from case study of chemical technologies suggest for how one should approach transgenic plants? First, the risks posed by transgenic plants will tend to be toward the unacceptable end of risks compared with many of the more familiar risks of life. Like chemical substances, genetic changes are invisible, undetectable features of plants and difficult to avoid, unless one is put on notice about their properties (and even that has limited value). Moreover, it is difficult to appreciate any risks they might pose because they are so far from our ordinary experiences and other common risks. For most of us, using or consuming transgenic plants are not central to our life plans.
Second, GM crops have risks that chemical substances tend to lack—they can replicate, propagate, migrate, mutate—and genes can "wander" from plant to plant within related species (Ellstrand, 2001
Third, the environment into which transgenic plants will be introduced has suffered from substantial human and technological impacts from previous technological advances. Consequently, it may be less resilient than it once was. Moreover, because according to the NRC both understandings of ecosystems and genetics are in their infancies, this creates additional reasons to be cautious in introducing transgenic plants with new and untested properties into the farming and natural ecosystems. Thus, it seems important to go slow at the beginning to understand as fully as possible the properties, risks, and possible problems from this new technology. Some risks will be more obvious, some much less so (van den Belt, 2003 Fourth, these first three features create a need for reliable trustees to provide protections that individuals cannot; governmental institutions to ensure that the risks individuals find it difficult to appreciate, detect, and protect against are identified and reduced before they materialize into harm to ecosystems and human health. What legal structure would such institutions take to address a new technology such as transgenic plants?
A first step would suggest a cautious approach, aspects of which the United States has tended to adopt. The U.S. Department of Agriculture in effect has a premarket approval law to guide review of transgenic plants; GM crops must be submitted to the agency, reviewed, and ultimately given permission to be planted for research, experimental, or commercial purposes (NRC, 2002
More important, it seems socially important to avoid being ignorant of risks from such products and losing social control of them, as occurred with chemical products. In addition, as the NRC (2002
A constituent feature of a premarket screening statute should be to make the approval, distribution, and manufacture of the product conditional upon quick removal when problems arise. That is, with the legally sanctioned distribution of a product that exposes the public and ecosystems, I suggest that the social permissibility of its distribution should remain conditional upon continued safety of the product. This is the case under some of the premarket approval statutes for drugs and pesticides (although it is not always as easy as perhaps it should be to withdraw a product when problems arise). As part of this legal structure, agencies should make the rights to public protection greater or of higher priority than private property rights of the manufacturer of the product so that property rights to the product do not make a product's removal so difficult that it is unlikely to occur. This more esoteric point goes to the particular design of the law in question, but it should be comparatively easy for an agency to show that a product no longer satisfies the condition of approval and withdraw it, if we seek to protect ecosystems and our health (Cranor, 2003a A further feature of such laws might require the manufacturers and distributors of the products to have a legal obligation to report adverse effects so that an agency, acting on behalf of the public, can be alerted to threats and risks before they become the next freon, 1,1,1,-trichloro-2,2-bis(p-chlorophenyl) ethane, PCB, or PBDE problem. Moreover, agencies will need to follow the guidance of the scientific community in identifying threats that constitute the basis of product "recall" once they are in commerce.
Finally, the collective effects of individual decisions can be lost in discussions about individual products or substances (Cranor, 2003b None of the above arguments suggests that new technologies should not be pursued and developed. Instead, they suggest that some of the technologies on the horizon pose risks that tend to less acceptable than many ordinary risks of life and that they will be introduced into an environment that may be less resilient than it once was. Finally, the chemical technology experience suggests that we should approach new technologies with considerable humility and try to ensure that they do not escape social understanding and control as a good many chemical products have. If such cautionary approaches are followed, perhaps we can have the benefits of new technologies without some of the risks and costs that have accompanied other technological revolutions.
I am grateful for comments on this paper from Norman Ellstrand. Received May 7, 2003; returned for revision May 20, 2003; accepted May 20, 2003.
1 This work was supported by the National Science Foundation (grant no. 99-10952) and by a grant from the University of California Toxic Substances Research and Teaching Program.  * E-mail carl.cranor{at}ucr.edu; fax 909-787-5298.
Brown LR, Flavin C, French H (2002) State of the World 2000. W.W. Norton Co., New York Colborn T, Dumanoski D, Myers JP (1996) Our Stolen Future: Are We Threatening Our Fertility, Intelligence, and Survival? A Scientific Detective Story. Penguin Books, New York Cone M (2003) Cause for alarm over chemicals. Los Angeles Times, Sunday, April 20, A1 Cranor CF (1993) Regulating toxic substances. Oxford University Press, New York Cranor CF (1995) The use of comparative risk judgments in risk management. In AM Fan, LW Chang, eds, Toxicology and Risk Assessment: Principles, Methods and Applications. Marcel Dekker, Inc., New York, pp 817-833 Cranor CF (2003a) Some legal implications of the precautionary principle: improving information generation and legal protections. Eur J Oncol (in press) Cranor CF (2003b) What could precautionary science be: research for early warnings and a better future. In J. Tickner, ed, Environmental Science and Preventive Public Policy. Island Press, Washington, DC, pp 305-319 Department of Health and Human Services, Food and Drug Administration (1994) Notice for a hearing on a proposal to withdraw approval of bromocriptine mesylate. Federal Register 59: 43347-43352
Ellstrand NC (2001) When transgenes wander, should we worry? Plant Physiol 125: 1543-1545 Ellstrand NC (2003) Going to great lengths to prevent the escape of genes that produce specialty chemicals. Plant Physiol (in press) Environmental Health Letter (1998) EPA, EDF, CMA agree on testing program targeting 2,800 chemicals. Business Publishers, Inc., Silver Spring, MD, p 193 Green MD (2000) Safety as an element of pharmaceutical quality: the respective roles of regulation and tort law. St. Louis Univ Law J 42: 163-190 Hooper K, McDonald TA (2000) The PBDEs: an emerging environmental challenge and another reason for breast-milk monitoring programs. Environ Health Persp 108: 387-392[ISI][Medline] Loewenberg S (2003) Public health. E. U. starts a chemical reaction. Science 300: 405
Lubchenco J (1998) Entering the century of the environment: a new social contract for science. Science 279: 491-497 Markowitz G, Rosner D (2002) Deceit and Denial: The Deadly Politics of Industrial Pollution. University of California Press, Berkeley McKibbon B (2001) Some like it hot. The New York Review of Books 48: 35-38 Melnick RL (2003) Carcinogenicity and Mechanistic Insights on the Behavior of Epoxides and Epoxide-Forming Chemicals. New York Academy of Sciences, New York (in press) NRC (1984) Toxicity Testing: Strategies to Determine Needs and Priorities. National Academy Press, Washington, DC NRC (2002) Environmental Effects of Transgenic Plants: the Scope and Adequacy of Regulation. National Academy Press, Washington, DC Rodgers WH Jr. (1994) Environmental Law, Ed 2. West Publishing Co., Minneapolis, MN
Slovic P (1987) Perception of risks. Science, 236: 280-285
Thompson PB (2003) Value judgments and risk comparisons: the case of genetically engineered crops. Plant Physiol 132: 10-16 Travis CC, Hester ST (1991) Global chemical pollution. Environ Sci Technol 25: 814-819 U.S. Congress, Office of Technology Assessment (1987) Identifying and Regulating Carcinogens. U.S. Government Printing Office, Washington, DC U.S. Congress, Office of Technology Assessment (1995) Screening and Testing Chemicals in Commerce. U.S. Government Printing Office, Washington, DC
van den Belt H (2003) Debating the precautionary principle: guilty until proven innocent" or "innocent until proven guilty? Plant Physiol 132: 1122-1126 This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ASPB Publications | PLANT PHYSIOLOGY | THE PLANT CELL | |
|---|---|---|---|
TOP
THE CHEMICAL REVOLUTION AS...
INSTITUTIONAL CONTROLS
