Review of Expert Panel IV minority position
Subsections
Scientific case for a bio-safety protocol
National legislation/sovereignty of the state
Administrative cost of separate treatment
Quantifying risks/risk assessment
Products to be covered by a protocol
Recent scientific evidence links the specific techniques of modern biotechnology to actual and potential adverse impacts on biological diversity. The term "living modified organism resulting from biotechnology" applies mainly to genetically modified organisms (GMOs; see page 5 about definitions). As we demonstrate below, regulation of the direct ecological risks of GMOs is scientifically justified. Ecological risks need to be considered in a global context, because local evaluation does not suffice to ensure global safety (Kareiva and Parker, n.d.). However, blanket assurances of safety for all GMOs are not scientifically sound. "Categorical exemptions" of certain classes of organisms, in particular, do not recognize the different expression of the same engineered gene or genes in different circumstances and different ecosystems. Unfortunately, promoters of biotechnology and even regulatory agencies tend to dismiss some of the real risks of GMOs, using arguments that neglect basic principles of ecology.
A summary of recent scientific findings follows.
Plants
The first known release of a genetically modified plant took place in 1986 in Belgium. By 1993 the total releases had exceeded 860. Of these only eighty-seven were not food related. The predominant trait tested in the field has been plants' resistance to chemical herbicides. The thirty largest US biotechnological and agrochemical companies spend hundreds of millions of dollars a year on research and development of this sort, which is meant to increase "productivity"--in market terms--and not necessarily "sustainability." (Anonymous, n.d.) Reflecting on the damaged done by the reputedly highly "productive" hybridized plant species of the "Green Revolution" (in terms of poor input/output ratios, soil poisoning and degradation, and water depletion), any allegations of the benefits conferred by new, specially designed crops should be treated skeptically.
The genetic engineering of plants usually involves "transgenes": genes from dissimilar organisms ro artificially constructed genes added to other organisms by techniques of molecular biology. It is wrong to conclude that "transgenic" crops are safe because past crops in general have not caused ecological disasters or because nothing unfortunate has happened as a result of thousands of GMO field trials to date. Past experiences do not apply to some of the new trait-crop combinations now possible through genetic engineering. As in all objective study, negative evidence of environmental risk should never be taken as evidence of no risk; it could simply be evidence that no one has examined a particular risk. (Kareiva and Parker, n.d.)
Although some GMOs may be safe, others undeniably pose substantial risks. Such is the case with transgenic crop plants. These plants are genetically engineered to contain traits of unrelated organisms. They pose two distinct types of risks because of their potential "invasiveness." The first kind of risk can be called "weediness risk." Despite the widespread belief that domesticated plants cannot proliferate widely without constant human care, there is ample evidence that crops and other cultivated plants have become weeds (i.e. plants unwanted for ecological or agronomic reasons) throughout history. In the United Kingdom, for example, the North American rhododendron (R. maximum and R. ponticum), imported for ornamental use, is now overrunning agricultural land, peat bogs and other ecosystems. The other type of invasiveness, known as "introgression" or "gene flow," is the movement of trangenes themselves from hybrids to traditional cultivars and wild relatives, and their spread through reproduction. The history of plant evolution is full of examples of introgression, genes moving between related species and even genera, especially by way of cross-pollination. Traditional cultivars and wild species receiving transgenes could in turn become weeds. (Kareiva and Parker, n.d.; Rissler and Mellon, 1993)
Among the most challenging risks of transgenic crops is the threat that they might pose to the populations of traditional cultivated plants ("cultivars") and their wild relatives that are centers of genetic diversity for crops, both as a result of the competition with the transgenic crops and by the transfer of the transgenes to the traditional cultivars or wild relatives. These centers of diversity, located primarily in the developing world, are regions that contain the greatest concentration of crop biodiversity. The likelihood of gene flow is difficult to assess in many places outside the United States because information is unavailable on the distribution of wild plants. This is often the case of countries comprising centers of diversity for crops important to world agriculture. (Rissler and Mellon, 1993)
Together, traditional cultivars and wild relatives are the richest repositories of crop genetic diversity. They are the natural reservoir for the traits needed to maintain the vitality of modern crops. The genes for important traits like disease resistance--few of which have been identified and isolated--are the natural capital on which both traditional crop breeders and genetic engineers depend. Crop genetic diversity is already diminishing at a stunning rate--a phenomenon known as "genetic erosion"--as farmers around the world are persuaded to abandon the numerous traditional cultivars of the past in favor of a relatively few modern crop varieties. Expensive transgenic plants, which will generally have to create large markets to recoup research costs, will exacerbate that trend. The US is the center of diversity for few crops. This means that transgenic crops sold and planted in the United States generally pose less of a direct threat to crop biodiversity than they do if they are planted elsewhere in the world. (ibid.)
Some plants are engineered with viral-coat protein genes as mechanisms of resistance to viruses. There is danger that viral genes inserted into plants could give rise to new viruses, through "transcapsidation" and other molecular biological processes already observed in transgenic plants. New viral pathogens could have a devastating impact on economically important crops, requiring considerable control costs. (ibid.)
Other plants are genetically engineered to produce potentially toxic or toxic substances like drugs and pesticides ("bio-pesticides"). Dow-Elanco, Ciba-Geigy, Monsanto, and Zeneca are among the corporations engaged in such research and development (Meister and Mayer, 1994). Not only can these specially expressed substances have "non-target effects"--such as poisoning and killing wildlife--they may also pose considerable health risks to humans. Furthermore, the targets of bio-pesticides can become resistant to the toxins. Currently, about ninety percent of bio-pesticide-producing crops contain transgenes from the bacterium Bacillus thuringiensis (Bt). Bt contains a gene that causes the secretion of a crystalline protein that dissolves the soft tissue of insects. Bt-resistance has already developed in populations of the diamondback moth and Colorado potato beetle. (Anonymous, n.d.)
Animals
Transfers of growth hormone genes into salmon has resulted in specimens that grow at much faster rates and may achieve huge changes in body size. Similar experiments have been conducted on carp and catfish for years in the US. If such genetically engineered fish were to escape, they might displace native salmon with irreversible economic and environmental consequences. Currently, there is no provision in the US regulatory framework for release of transgenic fish and other genetically engineered animals--not even the requirement to notify the public. (Devlin et al., 1994; Stabinsky, pers. comm.)
In 1987 researchers working for the US National Institutes of Health used micro-injection to insert the complete genetic material of the Human Immune-Deficiency Virus (HIV)--the virus linked to AIDS--into the chromosomes of laboratory mice. About ten percent of the offspring of the affected mice carried the HIV virus in their genetic make-up. An accident in the laboratory caused 127 mice so infected to die and go unnoticed for several days. Accidental or deliberate release or exposure of such genetically modified animals could result in the proliferation of untreatable diseases. (Kimbrell, 1993)
Bovine somatotropin (BST) is an example of a biotechnologically produced chemical, designed for administration to domesticated animals, which might have ill effects on human health, especially in the long term. BST is a hormone synthesized with components of genetically engineered Escheria coli bacteria. Cows to which it is administered produce fifteen percent more milk on the average the cows without BST treatment. This effect will lead to generally lower production costs for farmers using BST, eventually driving farmers who do not use it out of the market. Thus milked produced with BST is likely to take a greater share of the market gradually and hence be consumed by more people. Concern about the risks of BST-milk focus on, inter alia, its content of fatty acids with longer chain-length than those of natural milk and the possible incorporation of stray amino acids in the engineered hormone, with consequences for the human endocrine system. Among cows the use of BST is associated with reduced fertility, decreased endocrine function, disturbances in bone growth and increased risk of mastitis, among other symptoms. Long-term effects on humans and other animals are not yet known. (Lacey and Heritage, n.d.)
Soil and soil organisms
Biotechnological products have a range of impact much greater than the just individual organisms, their hosts or the vector of transmission. When any organism is added to an ecosystem, the entire ecosystem is affected. Use of a "principle of familiarity" is not sufficient for assessing risk. This holds true when we cannot clearly define ecosystem health; when we do not know the ecological function of most naturally occurring organisms, much less novel organisms; and when we do not know what bacteria, fungi, protozoa, nematodes or arthropods exist in most soils and ecosystems. As with plant GMOs, a case-by-case approach is necessary. Just because one species of a bacterium (for example, Pseudomonas paucimobilis) does not express its genetically engineered function in low-organic matter sandy soil, does not mean that we can extrapolate that a related species (P. fluorescens for example) will not express its engineered function in the same conditions. (Holmes and Ingham, 1994)
In every application of biotechnology soil and soil organisms will be affected to greater or lesser degrees. Soil organisms are crucial to plant growth. Without fully diverse and functioning soil organism communities, nutrients are lost to the groundwater, erosion increases, plants suffer greater nutrient deficiencies and greater disease. Without a complex soil food-web, ecosystem productivity will be negatively affected. Indeed, soil organisms and their aquatic cousins are what we rely on to produce clean water, clean soil, and clean air. (ibid.)
In particular comprehensive risk assessment of genetically engineered microorganisms (GEMs) has not been performed. One common attitude is that since those GEMs which have been tested have not shown significant ecological effects, there is no point in testing other GEMs for theirs. This is a misapplication of the principle of familiarity and a rationalization of "categorical exemption." Consider that perhaps only ten GEMs have actually been tested in any way for ecological effects. This is too small a sample from which to deduce that GEMs are in general safe. One of these tested GEMs, a pseudomonad engineered to degrade a chemical compound called 2,4-D, was not capable of carrying out its engineered function in the system into which it was placed. The conditions of the test, without plants or sources of food added to the system, prevented the bacterium's growth. Naturally, no ecological effect was observed. Further work, however, showed that this GEM had a negative impact on the fungal community in a slightly different soil, under slightly different conditions. In an agricultural soil, loss of the fungal component marks soil degradation. This loss might not be a problem for crop growth, if the loss is transient. In forest soils, however, loss of the fungal community can result in the loss of the existing trees. Thus, in a forest, release of this GEM might have serious implications, though the genetically engineered pseudomonad in question has never been tested in a forest soil. Yet by current risk assessment standards, based on the testing that has been done, there would be no restrictions of its use. (ibid.)
In the case of the bacterium Klebsiella planticola prior testing performed in laboratories indicated there was no problem with the GEM. When the GEM was tested in complex soil "microcosms" (self-contained units with field conditions, incubated in growth chambers), this particular GEM killed the wheat planted in the units. The parent bacterium and controls with no added bacteria had not effect on wheat, and the units were all incubated in the same growth chamber. The effect of the GEM on wheat growth was tested in several different environmental conditions as well, and the mechanisms by which the wheat was killed or negatively affected by the GEM appear to be different in different environmental conditions. Continuing investigation of the ecological effects of various GEMs has discovered that they are capable of significantly affecting ecosystem health. (ibid.)
"Bio-remediation," the ostensibly beneficial use of GMOs to attack hazardous waste, can also prove risky. Soil researchers tested K. planticola (mentioned above), which is genetically engineered to turn crop waste into ethanol, and found an unpredicted side effect: the bacterium cut the amount of ecologically important mycorrhizal fungus in the soil by more than half at critical early phases of plant development. Soil researcher Elaine Ingham noted that if K. planticola survived readily and spread widely, it would very likely not be possible to grow crops without external control measures for this organism. Chemical or technological control measures could easily have their own adverse effects on the environment, which would have to be strictly managed. (Hill, 1994)
Summary of scientific points
The blanket assurances of governments and industry about the safety of the products of biotechnology are insufficient and inadequate. CBD meetings on bio-safety will have to develop methodologies for the case-by-case assessment of the safety of biotechnological products, taking into consideration the major findings above:
Adaptability/modalities [p. 11]
Clearly a bio-safety protocol must be flexible, given the current states of ecological knowledge and institutional capacities. A bio-safety protocol should not impose restictions on the transfer, handling and use of the products of bio-technology based on vast generalities about their effects on the environment; nor should a bio-safety protocol attempt to predict every every conceivable impact that the transfer, handling and use of biotechnological products might have. Nevertheless, a comprehensive protocol should cover:
Public perception [p. 11]
The ability of people to understand basic ethical and political issues should never be underestimated. Calls for regulation of the biotechnology industry have not increased public concerns. People are concerned for a variety of reasons, be there regulation or not. Reasons include claims that new technologies, like biotechnology, are risk-free, which is not persuasive so long as people recollect the dangers posed and damage done by the unsupervised implementation of previous technologies. Industry may think that it is protecting its image by insisting there are no risks, but if people find out that they have been misled, they will almost certainly become hostile to biotechnology. This would create more problems for the image of biotechnology among informed people than regulation would. The motivation of sound industry-public relations should not be to sell the technology to the public first and then assess the possible dangers, but rather to establish scientifically sound standards of assessing the danger so that the technology can be safely and appropriately ultilized. As a first step toward honestly informing the public, Parties should consider the question of proper labelling of LMOs produced by biotechnology.
Certain members of governments have suggested that public perception should be molded to create "public acceptance" of biotechnology and that public acceptance is a prerequisite for transfer of this technology and its products. This suggestion is misguided. Public acceptance or rejection depends on careful, objective exposition of the actual and potential risks of living modified organisms. Article 19.3 concerns the safety of transfer, handling and use of products of biotechnology--not the facilitation of their transfer. The question of their safety, as well as the safety fo the methods by which they are produced, should not be begged by governments eager to promulgate biotechnology en bloc. A bio-safety protocol would help determine what biotechnology--including methods and techniques of risk assessment and management--is safe and appropriate to transfer. Thus it would help build adequate and appropriate national capacities.
There are claims that regulations have delayed benefits from biotechnology. However, the biotechnology industry has had other, much greater problems, including problems with the technology itself, its management culture and unrealistic expectations. The fact is that hundreds of projects have been approved for field testing without significant delay due to regulation. Regulations will force the industry to focus its developments and claims more carefully and in the end improve its scientific judgments and maximize the benefits that the technology may truly offer. Biotechnology industry officials have admitted that the prospect of facing regulations often tends to force companies to look more carefully at projects both in terms of potential benefits and technical feasibility and has had a salutary effect in many cases.
Lastly, scientifically sound regulation of biotechnology should not be dismissed merely as a "barrier to trade." Nations have the right to choose regulation based on science.
Overlap/duplication [p. 11]
There is no international safety agreement on biotechnology that deals with the specific global risks to biological diversity posed by products of biotechnology. As explanation of the scientific evidence shows, modification of an existing agreement would not adequately cover the potential risks that GMOs pose to biological diversity. International guidelines on safety have a reputation among scientific experts for having dangerous loopholes. The International Plant Protection Convention, for example, has no jurisdiction over animals or microorganisms (see p. 8 above). Parties to the UNIDO Voluntary Code of Conduct for the Release of Organisms into the Environment have failed to incorporate the proper risk assessment principles into practice before release of a GMO (Leonard and Schweiger, 1994). National and regional agreements, no matter how sound, cannot be relied upon, since they cannot regulate widespread trade in the products of biotechnology among diverse nations and regions.
If the nations of the world do not act in concert to establish high
standards for bio-safety, biotechnology industries will continue, as
they do now, to threaten to move their operations to countries that have
few or no safety standards. Bio-safety will degenerate nearly
everywhere into a façade, because poor and rich countries alike will
fear economic losses and retribution. Poor countries will be most
vulnerable to the temptation to take ecological and health risks in
pursuit of so called "economic growth," but even wealthy countries have demonstrated
that they will bend to such fears and adopt low or negligible standards
of safety.
Codes of conduct/"soft law" [p. 12]
Experience with non-binding arrangements for the handling and use of
chemicals and pesticides has shown that such "codes of conduct" and
"voluntary guidelines" do not alert governments sufficiently to the
importance of complying with their stipulations. Without a binding
treaty governments are not likely to pass local regulations or comply
with international norms or rules. Parties in developing countries
have complained that they cannot get their ministries to take voluntary systems seriously and thus often do not receive proper
information through
voluntary codes about the use of potentially hazardous substances. This is the case with the prior informed consent (PIC)
agreements, which have not been able to acieve their goals as voluntary
arrangements. Many governments have recently recognized this problem and
are considering alternative, comprehensive PIC agreements. Although many
manufacturers abide by voluntary arrangements, others will ship a banned
substance to a party in another country despite the ban. Finally,
developing countries do not have the resources to regulate certain
substances and therefore want manufacturers to take the responsibility
in the form of a mandatory agreement. (Fuller, pers. comm.)
National legislation/sovereignty of the state [p. 12]
One effect of a protocol would be to encourage and expedite the development and adoption of national legislation on bio-safety in conformity with the framework regulations of the international agreement. Article 6 of the CBD directs the Parties to "develop national strategies, plans or programmes for the conservation and sustainable use of biological diversity or adapt for this purpose existing strategies, plans or programmes which shall reflect, inter alia, the measures set out in this Convention relevant to the Contracting Party concerned...." Governments will only feel the need to enact bio-safety legislation if they ratify the Convention. This process ensures that there is no breach of national sovereignty in the development of a binding international bio-safety protocol.
While
regulation of releases of GMOs has been instituted in the United States
and other OECD countries, many developing countries do not have any
controls. In the meantime unregulated releases, where there is no supervision to ensure
environmental safety, already have taken place in thirteen countries: Belize, Burkina Faso, Côte D'Ivoire, Dominican Republic, Guatemala, Mali, Nigeria, Pakistan, Peru, Puerto Rico, South Africa, Syria and Zaïre. Illegal field tests have taken place in four: Argentina, Kenya, India and Ireland. Field
trials of genetically engineered plants are expected to take place
shortly in thirty-five developing countries. (Meister and Mayer, 1994)
Administrative costs of separate treatment [p. 12]
That consideration of a bio-safety protocol is mentioned in the CBD is recognition of the fact that the transfer, handling and use of biotechnological products is an international problem, due to international trade in genetically engineered organisms and the transboundary nature of the ecological risks posed by GMOs, and that it therefore necessitates a coordinated international response. Parties should not let themselves be distracted from this insight when negotiating a bio-safety protocol to the Convention.
A bio-safety protocol would be administered under the auspices of the COP, not by a separate governing body. One cannot contract to the protocol without being a Party to the CBD. Article 32.1 of the CBD (Relationship between this Convention and its Protocols) states: "A State or a regional economic integration organization may not become a Party to a protocol unless it is, or becomes at the same time, a Contracting Party to this Convention." This provision also precludes the possibility of states making decisions about international bio-safety without committing to the requirements of the CBD.
The scope of the protocol, to be decided by the Conference of the Parties, will determine the size and cost of a bio-safety protocol. Existing sector-specific procedures for transfers of products of biotechnology may be taken into account by the COP in determining the scope and administration of the bio-safety protocol. The costs of a bio-safety protocol are small when such an instrument will facilitate and expedite the adoption of biotechnology applications that have been proved sound. Furthermore, while the Party importing an LMO should be responsible for advance informed agreement, the CBD clearing-house mechanism (article 18.3) could provided for transfer of the relevant information, without significant added cost to the Parties.
In addition, article 6 of the Convention, directs Parties to "integrate,
as far as possible and as appropriate, the conservation and sustainable
use of biological diversity into relevant sectoral or cross-sectoral
plans, programmes and policies." By enhancing cross-sectoral and
interagency plans and programs, Parties are sure to save
institutional and financial resources, and become more efficient.
Implementing Agenda 21 [p. 12]
Agenda 21, another of the documents resulting from UNCED, sets forth areas for future government action to conserve biodiversity and use it sustainably. It constitutes a declaration of intent on the part of the world's governments. A number of chapters of Agenda 21 are relevant to the development of regulations for the safe transfer, handling and use of the products of biotechnology. Chapter 14, on sustainable agriculture and rural development, and Chapter 16, on "environmentally sound management of biotechnology," are especially important.
Chapter 14 recognizes that "increased use of external inputs and development of specialized production and farming systems tend to increase vulnerability to environmental stresses and market fluctuations." This statement is a direct reference to the environmental damage done by the recent single-crop agriculture relying on hybrid "high-yielding varieties" (HYVs) that require immense inputs of water, artificial fertilizer and mechanical energy. (See Shiva, 1991.) Contemporary biotechnology uncritically offers the same promises in the form of transgenic crops that the purveyors of HYVs did in the 1960s and 1970s. Companies like Calgene, Monsanto and Rhône-Poulenc are developing trasgenic crops specifically to resist the deleterious side-effects of chemical herbicides and pesticides, thus encouraging their continuing and increasing use (Rissler and Mellon, 1993; Meister and Mayer 1994).
Chapter 16 states that its
"programme areas ... seek to foster internationally agreed principles to be applied to ensure the environmentally sound management of biotechnology, to engender public trust and confidence, to promote the development of sustainable applications of biotechnology and to establish appropriate enabling mechanisms, especially with developing countries...,"including inter alia "[e]nhancing protection of the environment" and "enhancing safety and developing international mechanisms for cooperation." A bio-safety protocol to the CBD based on the recommendations of this paper would be a concrete step toward all these goals.
Finally, Agenda 21 (like the Rio Declaration) stresses the importance of
informed public participation in decision making. Thus the development
of a possible protocol on bio-safety should take into account the formal
and substantive contributions of NGOs, most of which are based on
membership and active participation of citizens. More importantly, by
developing a legally binding agreement that would inform mandatory
national regulation, through national legal instruments and
institutions, the public would have a greater chance of contributing to
and effecting measures for bio-safety.
Quantifying risks/risk assessment [pp. 27, 30]
The long-term, cumulative risks to ecosystems of releasing large numbers of GMOs are not sufficiently understood to allow their prediction except in broad outline. It is unlikely that ecosystem dynamics will be well enough understood in all its specific detail at any time in the near future to predict this aspect of environmental impact confidently. Nonetheless, what is known confidently about ecosystems and the effect of natural or genetically engineered species on them is enough to recommend that biotechnology be managed with extraordinary caution.
Criteria for assessing risk must be developed as part of a bio-safety protocol. According to the British Ecological Society and the Ecological Society of America (ESA), evaluation of these risks must be made on a case-by-case basis, rather than assuming their general safety based on a few controlled field tests (Tiedje et al., 1989). The ESA has published tables of significant types of methods, organism attributes and environments for possible consideration in risk evaluation, which future meetings on a bio-safety protocol may wish to use as a starting point for discussion on how to develop indicators and measures of risk. The typology comprises broad categories of analysis, similar to, though more specific than, those used in this paper.
Knowing whether a particular engineered trait has the potential to confer the characteristic of weediness to a new organism is not enough to predict whether it may happen. Traits may have different effects in different circumstances, while novel traits, such as virus-resistance, could have unknown effects. Therefore, lists of organism traits are not sufficient for risk assessment. Containment of transgenes
is impracticable when GMOs are produced commercially for global
distribution. For this reason, in managing risks, regulators cannot rely on containment as the sole strategy of risk avoidance or reduction. Rather, they must address the
ability of transgenes and transgenic plants to escape the agricultural
setting for which they were designed--their "invasibility." (Kareiva and
Parker, n.d.) The problem of "scale-up" is already well known in the chemical and pharmaceutical industries. Like a chemical pollutant, a genetically engineered organism that requires only a modest degree of caution in test-tube quantities in the laboratory could become an environmental disaster if released into the wild in vast amounts.
Products to be covered by a protocol [p. 29]
As stated above, a bio-safety protocol should regulate all LMOs produced by nontraditional breeding methods. Therefore, its provisions and stipulations should extend only to those "alien species" arising from modern biotechnological techniques--techniques not based on natural mechanisms, such as fertilization and propagation. The protocol should be technology-specific. Attempts to conflate modern biotechnology and traditional breeding methods are political tools that serve to weaken the justification and development of a protocol.
The scientific analysis of transgenic organisms has and will have unique difficulties. For instance, one may already know a great deal about the biology of a nonindigenous species or can research it, but in the case of an organism with genetically engineered novel traits, accumulated natural history cannot provide useful information. Thus one must make rather abstract inferences about the novel trait, based on contemporary theory of how genetic engineering influences the biology of the host, as well as its expression in previous donor organisms. Lastly, not only has it been demonstrated that the ecological assessment of risk is complex, due to the multifarious nature of ecosystems, but also that a thorough understanding of all the factors may require unprecedented interdisciplinary efforts.
* Numbers refer to pages in the Report of Expert Panel IV (UNEP/Bio.Div./Panels/Inf.4)
Next section/The need for a protocol