The core issues in the Biosafety Protocol: An analysis

Only the core issues of the Biosafety Protocol were negotiated at Montreal. The remaining provisions were basically the same as those negotiated at the earlier talks in Cartagena. Lim Li Lin analyses and comments on these core provisions.

'IF you are destined to die, and you become lame, you are lucky'. So goes a Ugandan saying that was quoted by an African Minister to describe the Cartagena Protocol on Biosafety.

That, perhaps, is the best summation of the final agreement that was reached at 4.45 a.m. of 29 January in Montreal, Canada. The Protocol that was adopted by more than 130 countries was a heavily negotiated deal struck at the eleventh hour, in a desperate attempt to avoid a repeat of the collapse of negotiations at Cartagena, Columbia in February last year.

Six countries led by two major producers, the US and Canada, were opposed to an international law to regulate genetically engineered (GE) organisms. The majority of the rest of the world wanted a strong and comprehensive system to cover all GE organisms and products made from or containing GE organisms.

With every attempt at regulating big business including even voluntary codes of conduct at the United Nations, having been defeated over the past decades, the Biosafety Protocol was indeed an achievement.

But it is not a fully robust agreement.

Many concessions were made, primarily by the Like-Minded Group of developing countries, in an attempt to reach agreement in Cartagena. However, the talks collapsed due to the refusal of the Miami Group (the US, Canada, Australia, Argentina, Chile and Uruguay) to subject GE commodities intended for food, feed and processing to some form of regulation. At that point, the Like-Minded Group at the closing session, declared they were 'putting back on the table all the critical elements for a strong Protocol'. This meant that whatever compromises that had been agreed to by the Like-Minded Group in order to secure agreement were effectively withdrawn.

The concluding Montreal talks used the draft Protocol from Cartagena. That Cartagena version itself was never negotiated.

It was produced by the Danish chairman of the Biosafety Working Group which had met six times over almost three years to work on the Protocol. The last meeting was in Cartagena, and his draft was a desperate attempt to produce a Protocol from an unwieldy text with more than 670 brackets.

This itself is a peculiar feature of the Protocol.

Governments, individually or collectively (as in the Africa Group), had submitted detailed drafts of what they wanted as a Protocol. A large number of developing countries were the most active, providing the most detailed provisions because of their desire for a strong biosafety agreement. In contrast, most industrialised countries submitted general, even meaningless proposals. Many issues covered by developing countries, such as socio-economic considerations, liability and redress, and transit, were rejected by them.

Make or break on three issues

At 'informal consultations' in Vienna in September last year, the Chairman of the negotiations, Juan Mayr Maldonado, Minister of Environment for Colombia, identified three key outstanding issues to focus the consultations on.

These were the most contentious issues at Cartagena - the general scope of the Protocol; the application of the Advance Informed Agreement (AIA) procedure to GE commodities intended for food, feed and processing; and the relationship of the Protocol to other international agreements, especially the trade agreements.

The consultations were not intended to negotiate actual draft text, but to explore the possibility of reaching a conceptual compromise on the three issues. Little headway was made at the informal consultations in Vienna, apart from the display of  more political will by the other countries to conclude an agreement.

When talks resumed in Montreal in January this year, negotiations turned on these three core issues. Given the political and time pressures to reach agreement at the end of the one week scheduled for the talks, negotiating groups were left with little opportunity to address other issues and provisions which were still of concern to them.

The Chairman reiterated the common understanding that 'nothing is agreed until everything is agreed', but in the end, negotiating groups were resigned to the fact that few provisions apart from the three areas (and those specific provisions with direct bearing on these three) would be re-opened for negotiation.

Thus, a large part of the Protocol was never conclusively negotiated, and remains as of the Cartagena version. The Protocol's conclusion was really dependent on the trade-off reached on the three core issues and some related provisions.

Advance Informed Agreement for first import only

At the heart of the Biosafety Protocol is the prior informed consent system since transboundary movement is the focus of the agreement. Called the 'Advance Informed Agreement' (AIA) procedure in the Protocol, this applies only to the first intentional transboundary movement of certain GE organisms for release into the environment.

The large bulk of GE organisms are not covered, including commodity seeds for food, animal feed and processing; and organisms which are merely transiting through the territory of a Party. In addition, Article 7(4) states that the meeting of the Parties to the Protocol can identify living modified organisms (LMOs) which are not likely to have adverse effects on the conservation and sustainable use of biological diversity, taking into account risks to human health, in order to exclude these LMOs from the AIA procedure. This 'exclusion list', however, must be agreed upon by all parties.

The Like-Minded Group of developing countries had wanted AIA to apply to the first transboundary movement of all GE organisms. This itself was a compromise, since the nature of GE organisms as living and evolving organisms requires a case-by-case approach, thus a fresh consent for each import.

This was a principle for AIA which was dropped when the major producer countries did not support it.

The AIA procedure provides for notification (the notification must include the relevant  information about the LMOs) by the exporting Party to the importing Party for the first intentional transboundary movement of LMOs. This triggers a process of decision-making based on full information and risk assessment, according to the Precautionary Principle. The primary responsibility is on the exporter to notify and provide information. The basis of the risk assessment is set out in Annex III to the Protocol.

The inclusion of the Precautionary Principle is very significant. Under the Protocol's AIA procedure, this means that lack of scientific certainty regarding the adverse effects of a LMO on the conservation and sustainable use of biodiversity, taking into account human health risks, can be a basis for prohibiting an import, imposing conditions on the import, or requesting more information.

However, time limits are imposed on importing countries that are Parties to the Protocol. Thus, within 270 days of the receipt of notification by the exporter of an intended transboundary movement, the Party of import must declare its decision. This is a step back, as countries are now not bound by any time limit and considering that the European Union took almost two years to approve GE maize.

Though there must be express consent before any export can take place, the time limits may now be used against a country.

The US and industry have always demanded 'timely' decision-making, and in this respect, they have gained an advantage.

If it is a conditional consent to, or a rejection of, the import, then reasons for that decision must be given.

Decisions may be reviewed in light of new scientific evidence, and the procedure is set out in Article 12.

General scope of the Protocol

The general scope of the Protocol applies to 'the transboundary movement, transit, handling and use of all living modified organisms that may have adverse effects on the conservation and sustainable use of biological diversity, taking also into account risks to human health.'

The term 'living modified organisms' (LMOs) is used throughout the text of the Protocol instead of the more commonly used terms, 'genetically modified organisms' (GMOs) or 'genetically engineered organisms'. The term was adopted at the insistence of the United States in the course of the negotiations of the Convention on Biological Diversity, under which the Protocol was negotiated.

The US had opposed the use of the term 'genetically modified' and wanted to downplay, semantically, the connotations of the term. The US was adamant that GE food and crops were no different from conventional varieties, and should be accorded the same treatment, as they introduced no special hazards. It was, however, always understood that LMOs are GMOs, and the definition of the term in the Protocol is clear on this point.

A key fight during the course of the five-year negotiations was for the inclusion of 'products thereof' in the scope of the Protocol. This was strongly advocated by the Like-Minded Group. 'Products thereof' would include products derived from LMOs such as tomato paste made from GE tomatoes, and soy proteins, a product of transgenic soya beans. However, 'products thereof' are now excluded entirely from the scope of the Protocol and, as such, remain unregulated internationally.

In the Cartagena draft text of the Protocol, exclusions to the general scope were contained within Article 4. Article 4(2) excluded the transboundary movement of LMOs which are pharmaceuticals for humans, and LMOs 'that are not likely to have adverse effects on the conservation and sustainable use of biological diversity, taking also into account risks to human health'. These LMOs would be listed in an annex to the Protocol.

Transit (passing through the territory of a third party) of LMOs and LMOs destined for contained use were also effectively excluded from the general scope of the Protocol. Certain exceptions, however, were made, for example, in the case of unintentional transboundary movements which may require emergency measures to be taken.

At the informal consultations in Vienna, the Like-Minded Group submitted a proposal for a comprehensive scope of the Protocol. That proposal brought all LMOs under the general scope of the Protocol, with AIA required for the first intentional transboundary movement of a specific LMO intended for a specific use. Only a written notification from the exporting Party would then be required for subsequent transboundary movements.

The proposal also allowed the importing Party the right to decide whether or not to require AIA for LMOs which are pharmaceuticals for humans and LMOs destined for research in contained use.

In addition, the exporting Party would be under an obligation to inform or notify the Party of transit if LMOs will pass through the territory of that country.

This proposal was obviously unacceptable to the other negotiating groups and no detailed discussions ensued thereon.

In Chairman Mayr's 'non-paper', which was a draft proposal on actual language for the three core issues, and which was distributed prior to the resumption of negotiations in Montreal, the recommendation for the general scope of the agreement was to retain Article 4 of the Cartagena draft, with all its exclusions.

Thus, the Like-Minded Group faced an uphill battle to re-open the scope of the Protocol. This battle was crucial to the Like-Minded Group as all LMOs carry the same risks and hazards whether they are used in agriculture, medicine, or research, and regardless of whether they are classified as commodities or pharmaceuticals. An international agreement to regulate LMOs should not arbitrarily exclude any category of LMOs on any basis.

Science backed opening of scope

Fortunately, a lot of new scientific evidence had emerged since Vienna, which bolstered the arguments of the Like-Minded Group for a comprehensive scope.

A scientific report by the Institute of Science in Society, UK, which was released at the negotiations in Montreal, revealed that a huge variety of artificial genetic material (naked/free nucleic acids) are made in the laboratory by genetic engineering biotechnology. These are being released without controls into the environment. They are used as research tools, in industrial productions and in medical applications. (Under the Protocol's loose definition of 'contained use', most of these applications would be considered to be contained.)

According to the report, these genetic material are potentially the most dangerous xenobiotics (substances which are foreign to nature) to pollute the environment as nucleic acids can be taken up by all cells to multiply, mutate and recombine indefinitely. There is abundant evidence that extraneous nucleic acids taken up can have significant and harmful biological results, including cancer in mammals.

Another report on the environmental risks of genetically engineered vaccines, published by the Directorate for Nature Management, Norway, was also released in Montreal. The report concludes that 'from an ecological and environmental point of view, many first generation, live, genetically engineered vaccines are inherently unpredictable (and) possibly dangerous' and should not be used on a large scale until the problems identified have been addressed and clarified.

However, in practice, the risks are considered non-existent from the medical and scientific points of view simply because no investigations addressing them have ever been made.

In the final text, the general scope of the Protocol in Article 4 provides for a comprehensive scope, covering all LMOs, and does not specifically exclude any category of LMOs.

However, the two articles that follow (i.e. Article 5 and 6) limit the application of the Protocol with regard to pharmaceuticals, transit and contained use. These are new compromise provisions that only surfaced in Montreal, and an improvement on the Cartagena text.

Importantly, the provision to exclude from the entire scope of the Protocol, LMOs that are 'not likely to have adverse effects on the conservation and sustainable use of biological diversity, taking also into account risks to human health' was dropped, closing a very large loophole. However, it must be noted that Article 7(4) could potentially exclude some LMOs, on the same basis, from the AIA procedure.

Pharmaceuticals: In and out

Article 5 states that, 'Notwithstanding Article 4 and without prejudice to the right of a Party to subject all LMOs to risk assessment prior to the making of decisions on import, this Protocol shall not apply to the transboundary movement of LMOs which are pharmaceuticals for humans that are addressed by other relevant international agreements or organisations'.

LMOs which are pharmaceuticals for humans include GE vaccines and insulin which, according to growing scientific literature, have environmental and health risks.

In other words, some LMOs which are pharmaceuticals for humans may be excluded entirely from the scope of the Protocol depending on whether or not they are addressed by other international agreements or organisations. But a lot depends on the interpretation of the terms used in this provision, for example, 'addressed' and 'relevant'. The former can mean anything, including legal regulation, codes of conduct or guidelines and standards.

It is envisaged that 'other international organisations' is meant to refer to the World Health Organisation (WHO).

However, the WHO does not deal with GE pharmaceuticals as such. Furthermore, it only sets standards for human health and safety and does not take into account impacts on the environment and biological diversity.

In any case, it is merely a standard-setting body which prescribes standards that members are encouraged to adopt.

Ceding jurisdiction from a legally binding international agreement to such a non-legally binding standard-setting organisation would result in a much lower obligation on Parties with respect to LMOs which are pharmaceuticals for humans. It also leaves open the question of whether or not standards which have been or might be set by such international organisations or any other international agreements are able to or will take into account the special hazards and risks of LMOs which are pharmaceuticals for humans.

In addition, leaving some LMOs which are pharmaceuticals for humans out of the entire scope of the Protocol also means excluding this category of LMOs from a liability and redress framework. (The framework for liability and redress under the Protocol is meant to be concluded within four years from the entry into force of the Protocol.)

So, we now have a situation where the regulation of some GE pharmaceuticals falls entirely under other international agreements and organisations which would set standards for domestic laws. The remainder will be covered by the Protocol.

This is a clear result of the resistance by some industrialised countries to protect the interests of the pharmaceutical sector from comprehensive international rules.

By separating the strands of biosafety oversight, there is a retreat from the rationale of the Protocol in the first place: a hard-fought recognition that there was no international legal regulation of genetically engineered organisms. While there is a role for the WHO, the Protocol should be the overarching legal standards-setting forum.

Yet, a victory to some degree was achieved by the Like-Minded Group in that there is no total exclusion. The debate also emphasises the urgency for national and regional regulation of this group of GE organisms and materials (see box ).

Why GE pharmaceuticals have to be regulated
Potential Environmental and Health Hazards

These constructs intended for therapeutic and prohylactic use are general expression vectors and LMOs and carry exactly the same environmental and health risks as any other constructs and LMOs.

      Examples                                                          Rationale

Edible vaccines (i.e. potatoes, bananas) These are LMOs by definition
Viral vaccines Viruses are 'organisms' as defined in Protocol and hence viral vaccines are LMOs
Naked viral genome vaccines Naked viral genomes are often more infectious than the intact viruses
Naked viral vectors for gene therapy They are intended to be incorporated into the genome of human cells and be replicated
Naked DNA vaccines & vectors They are known to incorporate into cellular genomes to be replicated, whether intended or not
Naked nucleic acid vectors and vaccines RNA can be reverse-transcribed into cDNA and become incorporated into cellular genomes to become replicated. RNA is the genetic material of the majority of viral families, and RNA recombination is much more common than previously thought
Unregulated Hazards: 'Naked' and 'Free' Nucleic Acids, Ho, M.W., Ryan, A, Cummins, J, and Traavik, T, ISIS-TWN Report, Jan 2000.
An Orphan in Science: Environmental Risks of Genetically Engineered Vaccines, Traavik, T, Directorate for Nature Managemant, Norway, 1999.

Transit countries win some protection

Article 6(1) on transit states that, 'Notwithstanding Article 4 and without prejudice to any right of a Party of transit to regulate the transport of LMOs through its territory and make available to the Biosafety Clearing-House, any decision of that Party, subject to Article 2, paragraph 3 of this Protocol, regarding the transit through its territory of a specific LMO, the provisions of this Protocol with respect to the AIA procedure shall not apply to LMOs in transit'.

What this means is that the AIA procedure, a system of prior notification by the exporting Party to the Party of import which triggers a decision-making process by the importing Party based on risk assessment and the Precautionary Principle, will NOT apply to LMOs that are merely passing through the territory of a third party.

The regulation of the transboundary movement of LMOs in transit will be left to domestic laws and regulations, and the Protocol does not set internationally binding minimum standards.

All other aspects of the Protocol which are unrelated to the AIA procedure will still apply. However, in effect, few operative provisions in the Protocol will apply to LMOs in transit, since the Protocol deals almost exclusively with the transboundary movement of LMOs from one Party to another, and the AIA procedure is at the heart of the Protocol.

The substantive provisions that do apply to LMOs in transit include risk management, unintentional transboundary movements and emergency measures, and liability and redress. Since documents accompanying transboundary movements of LMOs now require LMOs to be identified, transit states can have access to this important information as part of the normal access to such documents.

Again, though transit states do not have an international right to advance notice and prior consent as they had wanted, they obtained more than in the Cartagena draft. The mood then was to effectively exclude transit.

Contained use which is uncontained

Article 6(2) on contained use states that, 'Notwithstanding Article 4 and without prejudice to any right of a Party to subject all LMOs to risk assessment prior to decisions on import and to set standards for contained use within its jurisdiction, the provisions of this Protocol with respect to the AIA procedure shall not apply to the transboundary movement of LMOs destined for contained use undertaken in accordance with the standards of the Party of import'.

The language in this provision is highly ambiguous, particularly the last proviso, 'undertaken in accordance with the standards of the Party of import'. One reading of the phrase 'standards of the Party of import' could be domestic standards for contained use, for example, certification of a laboratory or other installation or facility as having met standards of containment, which may have nothing to do with the approval procedure for imports. These standards may even be non-legally binding guidelines.

This reading would mean that LMOs that are being exported for contained use will be exempted from the AIA procedure by the mere fact of existence of domestic standards on contained use.

Exempting LMOs for contained use from the AIA procedure on the basis that there are such domestic standards would go against the whole rationale of biosafety regulation for the transboundary movement of LMOs. This would leave developing countries, especially those without the necessary laws or biosafety capacity, in a vulnerable position when the reverse is needed, i.e. that the responsibility for care starts with the exporter.

Another reading of the phrase 'standards of the Party of import' could mean domestic regulatory procedures which includes provisions for the import of LMOs for contained use.

This reading is a rational interpretation of this provision.

In other words, the AIA procedure will not apply to the transboundary movement of LMOs destined for contained use, where there are procedures laid down by the importing Party for the import of LMOs for contained use, in addition to biosafety standards for the actual use, and these standards must be adhered to by the exporting Party. If an importing Party has not set such standards for contained use, the AIA procedure will have to apply to transboundary movements of LMOs destined for contained use in that country.

Unfortunately, a fundamental problem lies with the definition of 'contained use' itself.

In the Protocol, 'contained use' means 'any operation, undertaken within a facility, installation or other physical structure, which involves LMOs that are controlled by specific measures that effectively limit their contact with, and their impact on, the external environment'.

This is an extremely loose definition that allows many kinds of deliberate releases into the environment to qualify as contained use. For example, open field trials with fencing or other physical barriers; caged transgenic fish or other aquatic LMOs in ponds or other marine environments; and the deliberate release of liquid and solid wastes of laboratories creating GE organisms, would qualify as contained use under the Protocol (see box ).

Why 'contained-use' regulation must be tightened up

THE definition of 'contained use' in the Biosafety Protocol includes many kinds of deliberate release (see Box 1). The current regulation of contained use, as for example in the EU Contained Use Directive, actually includes 'tolerated releases' which should be prohibited in the light of existing scientific knowledge (see Box 2). In short, 'contained use' is not contained.

Regulatory oversight on contained use was established in the 1970s largely on the basis of assumptions, every one of which has been contradicted by scientific findings since. The major inadequacy is that the regulation takes no account of the ability of transgenic DNA and RNA to persist in all environments and to transfer horizontally to unrelated species. Furthermore, biologically 'crippled' strains of transgenic microorganisms, assumed not to survive in the environment, are now known to persist or to proliferate. The ecological impacts due to potential changes in the composition of natural microbial populations are unknown and unpredictable. In all cases where the release of transgenic microorganisms into the environment has been monitored, effects on the composition of soil microbial communities have been found.

Instead of tightening the regulation to reflect the new knowledge, and in accordance with the Precautionary Principle, there has been progressive deregulation of 'tolerated releases' of potentially the most hazardous LMOs and transgenic DNA.

It is imperative that the regulation of contained use is tightened up to reflect existing knowledge and in accordance with the Precautionary Principle.


Ho, M.W., Ryan, A., Cummins, J. and Traavik, T. (2000). Unregulated Hazards. 'Naked' and 'Free' Nucleic Acids. TWN Report, Penang.
Ho, M. W., Traavik, T., Olsvik, R., Tappeser, B., Howard, V., von Weizsacker, C. and McGavin, G. (1998). 'Gene technology and gene ecology of infectious diseases'. Microbial Ecology in Health and Disease 10, 3-59, also TWN Report.
Tappeser, B., Jager, M. and Eckelkamp, C. (1999). Survival, Persistence, Transfer: An update on current knowledge on GMOs and the fate of their recombinant DNA. Third World Network, Penang.
Traavik, T. (1999a). Too early may be too late: Ecological risks associated with the use of naked DNA as a biological tool for research, production and therapy. Report to Directorate of Nature Management, Norway.
Traavik, T. (1999b). An orphan in science: Environmental risks of genetically engineered vaccines. Report to the Directorate of Nature Management, Norway.

Box 1: Examples of 'contained uses' in the Biosafety Protocol which are deliberate releases

* Caged transgenic fish or other aquatic LMOs in open ponds, lakes and marine environments
* Vacinations with transgenic viruses and naked nucleic acid vaccines
* All forms of gene therapy
* Xenotransplantation using transgenic animal organs
* Open field-trials with fencing or other physical barriers
* Transgenic organisms enclosed in cages or other containers and destined for deliberate release
* Liquid and solid wastes of transgenic livestock contained in the laboratory
* Liquid and solid wastes of laboratories creating transgenic organisms destined for deliberate releases

Box 2: Examples of hazardous 'tolerated release' from contained users

* 'Diluted' air and liquid streams of live transgenic micro-organisme greatly in excess of the minimum infective dose of virulent bacteria such as E.coli 0157
* Undiluted liquid wastes from commercial and other users containing live transgenic microorganism not considered to be pathogenic, though we now know that non-pathogens can be converted into pathogens by acquiring unit-blocks of virulence genes or 'pathogenicity islands'
* Solid wastes of killed transgenic microorganisms and viruses, whether pathogenic or not, with large amounts of free or potentially free transgenic DNA containing antibiotic resistance genes, virulence and other genes which are likely to transfer horizontally to other microorganisms in the environment and create new bacterial and viral pathogens
* Solid wastes of killed cell cultures containing pathogenic viruses or transgenic DNA with viral oncogenes and other genes with harmful biological effects which may transfer horizontally to human cells and contribute to cancer
* Amplified nucleic acid sequences constructs, plasmids, transposons, artificial vectors which are known to be invasive to genomes of all cells

This loose definition of contained use may also create a 'double exclusion' for LMOs which are pharmaceuticals for humans. Even though some LMOs which are pharmaceuticals for humans may fall within the scope of the Protocol, they may nevertheless still be excluded from the AIA procedure on the basis that they are destined for 'contained use', i.e. that the human body qualifies as contained conditions. GE pharmaceuticals that are consumed by humans are, however, deliberate release into the environment as human waste can still contain genetic material originating from the GE pharmaceutical.

Parties may and should set higher domestic standards for contained use than that provided for under the Protocol. As a minimum, what is effectively a deliberate release into the environment should be clearly considered as such and thus should be subject to the full AIA procedure.

The right to set higher domestic regulatory standards is entrenched in Article 2(4), which states that nothing in the Protocol shall be interpreted as restricting the right of a Party 'to take action that is more protective of the conservation and sustainable use of biological diversity than that called for in this Protocol, provided that such action is consistent with the objective and the provisions of this Protocol' and is in accordance with its other obligations under international law. This would also extend to formulating definitions in the Protocol which are more appropriate for biosafety.

As with transit, all other provisions in the Protocol, apart from the provisions relating to the AIA procedure, will still apply to LMOs destined for contained use. But again, the exemption of LMOs for contained use from the AIA procedure shifts the regulation of the transboundary movement of such LMOs to national laws and does not set internationally binding minimum standards. For developing countries, this means the burden of care is fully on their shoulders.

LMOs intended for direct use as food, feed or for  processing

The main disagreement that sank the negotiations in Cartagena last February was the refusal of the Miami Group (almost all are major exporters of genetically engineered commodities) to accept any measures that would hinder the trade of GE commodities intended for food, feed or processing (LMO-FFPs).

LMO-FFPs are clearly within the general scope of the Protocol, but the Miami Group was opposed to the application of the AIA procedure to LMO-FFPs. These make up more than 90% of the Miami Group's exports of LMOs.

Biosafety Protocol - main provisions

THE Biosafety Protocol is a historic agreement, being the first time that international law recognises GE organisms as distinct and inherently different thereby requiring a separate regulatory framework.

Only the core issues were fully negotiated in Montreal, while the rest of the Protocol (which was drawn up in Cartagena) was accepted as part of the package without further negotiation. The Protocol provides minimum standards, and reaffirms a government's right to take national action.

The main provisions are:

* * Advance Informed Agreement:
The Advance Informed Agreement (AIA) procedure applies automatically to the first shipment of LMOs to be introduced intentionally into the environment, with decisions to be made according to the Precautionary Principle. The obligation is on the exporter/exporting Party to give notice to the importing Party, thus triggering the AIA.

Decisions relating to approvals of LMOs under the AIA must be preceded by risk assessment. The Protocol specifically provides that an assessment can also be required under domestic law for subsequent shipments. Risk assessment criteria are listed in the Protocol.

Parties must then implement risk management measures to regulate and control risks identified in the risk assessment.
* An information exchange system, basically on the Internet, is provided separately for LMOs for food, feed or processing instead of the AIA.
* GE pharmaceuticals for humans are covered by the Protocol, except where they are dealt with in other international agreements or organisations.
* No AIA for LMOs for 'contained use' and LMOs in transit. This means there is no automatic international obligation to inform the importing Party of a specific movement.

**  Identification and segregation:
a) Documents accompanying transboundary movements of LMOs for intentional introduction into the environment and those for contained use, must identify the organisms as LMOs. b) Documents accompanying LMO-FFPs need only to state that the shipment 'may contain LMOs'.

* * A system for liability and redress where the transboundary movement of LMOs causes damage or harm to biodiversity and human health, is to be worked out within four years after the Protocol comes into force.

** Parties can enter into 'side agreements' on issues under the Protocol, as long as these are consistent with the objectives of the Protocol and do not result in a lower protection than that provided for by the Protocol. There are fears that this may still weaken the Protocol as Parties may be tempted to avoid or dilute certain obligations through these agreements. This provision was not fully negotiated even though many developing countries did not want it at all.

** Parties can trade with non-Parties, under agreements consistent with the objectives of the Protocol. The major concern here is that a powerful non-Party, such as the US, may insist on weak agreements. This will amount to non-Parties enjoying the benefit of trade without the responsibility of biosafety.

** Parties are obliged to promote and facilitate public awareness, education and participation concerning the safe transfer, handling and use of LMOs by, inter alia, providing access to information on LMOs that may be imported, consulting the public in decision-making processes regarding LMOs and making the results of such decisions available to the public.

** In making import decisions, Parties can take into account socio-economic considerations arising from the impact of LMOs on the conservation and sustainable use of biodiversity, especially with regard to the value of biological diversity to indigenous and local communities. However, this has to be consistent with their international obligations. Thus, the issue of the relationship with the WTO agreements rears its head again.

** In cases of unintentional transboundary movements of LMOs (e.g., an accident) which may cause harm, there are obligations to alert affected Parties and to take emergency measures.

** Information submitted to a Party of import, as required by the Protocol, can be claimed to be confidential by the exporter. Thus the public's right to know is restricted. However, the following information cannot be considered to be confidential: name and address of the exporter; general description of the LMO; summary of the risk assessment and any emergency response plans.

** Parties can apply the Precautionary Principle, i.e., where there is no scientific certainty after a risk assessment is carried out, a Party can take the necessary steps to stop or restrict the import of an LMO. This is because an LMO, once released, can multiply and mutate, and cannot be recalled. able.

But as the Like-Minded Group explained at the consultations in Vienna, there is no difference between a LMO earmarked for planting and a LMO-FFP, and they have identical impacts on human health and the environment. Particularly in many developing countries, given the lack of capacity for enforcement, regardless of what a seed is intended for, such a distinction may not be able to be maintained once it enters the country. Hence, only a system 'as robust' as the AIA procedure would be acceptable to the Like-Minded Group for the regulation of LMO-FFPs.

(A recent study commissioned by the UK Ministry of Agriculture, Fisheries and Food concluded that DNA is not degraded under most commercial processing conditions, and that current animal feed is likely to contain substantial amounts of undegraded DNA. Thus, the secondary horizontal transfer of intact antibiotic resistance genes to bacteria from transgenic crops processed into feed cannot be ruled out, and other components of transgenic DNA may also have significant health impacts on livestock and human beings up the food chain.)

A proposal put forward by the Compromise Group - Switzerland, Norway, Mexico, Japan and South Korea (Singapore only joined the Compromise Group in Montreal) - in Vienna formed the basis of the discussion for an 'alternative AIA' which would apply to LMO-FFPs. However, after some headway made in these discussions, the Miami Group was forced to admit that it was not prepared to accept any new obligations that might limit or prevent exports of LMO-FFPs. They made it clear they were unwilling to cause any major disruption to their agricultural trade system. As such, they were only prepared to discuss mechanisms based on information sharing and insisted that the only way to address the range of views on LMO-FFPs is through domestic regulatory frameworks. This declaration effectively stymied the discussions on this issue in Vienna.

Pressure on the Miami Group to agree to some system for LMO-FPPs continued, and in Montreal a procedure was finally agreed upon. Article 11 requires a Party to inform all other Parties of an approval for domestic use, including introduction into the market, of a LMO-FFP that may be subject to transboundary movement.

This information must be communicated to the Biosafety Clearing House (basically a website administered by the Secretariat to the Convention on Biological Diversity) within 15 days of the final decision and, at a minimum, must include the information specified in Annex II. This information is less than that required for the AIA procedure. Although additional information may be requested, it falls on the potential importing Party to know what to seek. A better approach would have been to at least require the same information as in the AIA.

If a Party does not have access to the Clearing House (e.g. due to lack of Internet facilities), it must inform the Clearing House in advance. In such a case, the relevant information has to be provided in writing to the national focal point of that Party.

From there, a Party may take a decision on the import of that LMO-FFP under its domestic law. This procedure applies only to the first import of LMO-FFPs.

In the absence of a domestic regulatory framework, developing countries and countries in transition may follow the procedure in Article 11, after it makes a declaration in the Clearing House to do so. Again, this procedure applies only to the first import of the LMO-FFP. A Party can, before deciding on the first import, undertake a risk assessment in accordance with Annex III of the Protocol, which sets out the basis for risk assessment.

But a decision must be made within 270 days from the Party's declaration to the Clearing House that it will be making its decision as provided for in the Protocol for developing countries and countries in transition. (But developing countries and countries in transition are not bound, under the Protocol, by the 270-day time limit if they are applying their domestic laws and regulations.)

For this category of countries, however, express consent for the import of LMO-FFPs is still required.

This system for the transboundary movement of LMO-FFPs is in no way as robust as the AIA procedure for LMOs for intentional introduction into the environment. Rather, it is what the Miami Group wanted - for the transboundary movement of LMO-FFPs to be regulated through domestic regulatory frameworks and for the international framework to only lay down mechanisms for general information sharing.

The special provision for developing countries and countries in transition only applies so long as these countries do not have domestic regulatory frameworks. Though still lacking, the AIA procedure at least provides for notification by the exporting Party to the importing Party for the first intentional transboundary movement of LMOs. This triggers a process of decision-making based on full information and risk assessment, according to the Precautionary Principle. There must be express consent before any export can take place. The primary responsibility is on the exporter.

Notification of domestic approvals via the Clearing House mechanism shifts the burden to importing Parties to initiate procedures for assessing whether or not a particular LMO-FFP should be admitted into its territory. There may be a considerable time lag between the domestic approval and the actual transboundary movement of that LMO-FFP, if the LMO-FFP is even finally exported. A potential importing Party cannot even be sure whether or not the LMO-FFP will actually be exported to its country. All this increases further the burdens of finance and capacity of potential importing Parties, which are largely developing countries.

This system also basically shifts the regulation for the transboundary movement of LMO-FFPs to domestic regulatory frameworks. It does not set internationally binding minimum standards. Nevertheless, like the AIA procedure, it does operationalise the Precautionary Principle in import decisions.

Article 11 assumes that developed countries have adequate domestic regulatory frameworks in place, which is not necessarily the case. On the other hand, developed countries are not bound by the 270-day time limit.

Segregation and labelling

In the final hours of negotiations in Montreal, the Miami Group held out to the last on the provision that would most severely disrupt their exports of LMO-FFPs. Article 18 on handling, transport, packaging and identification became the final bargaining chip for whether the Protocol would be concluded or not. The Miami Group was only willing to accept that LMO-FFPs would be identified in shipping documents as 'may contain' LMOs and as not intended for intentional introduction into the environment. All the other negotiating groups had no choice but to give in to this demand in order to see the Protocol finally concluded.

In addition, a further clause was added - 'The Conference of the Parties serving as the meeting of the Parties to this Protocol shall take a decision on the detailed requirements for this purpose, including specification of their identity and any unique identification, no later then two years after the entry into force of this Protocol'. This is expected to be on the agenda of the meeting to be hosted by France at the end of this year as part of the preparations to implement the Protocol.

What this provision now means in effect is that exporters will not have to segregate GE commodities from non-GE commodities, as potentially, all commodities may contain LMOs. But as public pressure and consumer demand grows for segregation and labelling, this final trump card merely stalls an inevitable process.

Relationship with WTO

The draft Cartagena text contained a provision that would have effectively subordinated the Protocol to other international agreements, primarily the World Trade Organisation (WTO) agreements. Article 31 stated that, 'The provisions of this Protocol shall not affect the rights and obligations of any Party to the Protocol deriving from any existing international agreement to which it is also a Party, except where the exercise of those rights and obligations would cause serious damage or threat to biological diversity.'

Under international law, the interpretation of treaties is governed by the Vienna Convention on the Law of Treaties. The rule is that a later agreement supersedes an earlier one, and an agreement on a specific subject prevails over a general one. Since the Biosafety Protocol comes after the trade agreements and deals specifically with biosafety, in a conflict of laws, the Protocol has to be given priority.

Having a savings clause in the Protocol would have meant pre-empting the application of the Vienna Convention in interpreting the relationship of the Protocol to other international agreements. Even then, the Miami Group had wanted to delete the qualifying phrase which referred to harm or threat to biodiversity. The EU favoured a clear provision on the precedence of the Biosafety Protocol, while the Like-Minded Group said no provision was needed as the Vienna Convention on the Law of Treaties would apply.

In the September meeting in Vienna, the discussions on this issue went round in circles, and finally concluded with four statements of agreed concepts. Based on that, Minister Mayr proposed a Chairman's solution to delete Article 31 and add the following in the preamble to deal with this issue - (Recognising that there are other international agreements relevant to sustainable development with rights and obligations; ( Recognising further that trade and environment agreements and policies should be mutually supportive; (Emphasising that this Protocol and other international agreements are of equal status.

In the final text, Article 31 was deleted and the following language was added to the preamble: 'Recognising that trade and environment agreements should be mutually supportive with a view to achieving sustainable development,  Emphasising that this protocol shall not be interpreted as implying a change in the rights and obligations of a Party under any existing international agreements, Understanding that the above recital is not intended to subordinate this Protocol to other international agreements'.

This language relegated to the preamble carries far less weight than a substantive provision. Preambular language in international agreements, however, sets the framework for their interpretation. The effect appears to be a return to the general international law of interpretation.

However, this position is still vulnerable, as there are specific provisions in the Protocol that also refer to other international obligations. Article 2(4) on the right of Parties to take more protective domestic biosafety action qualifies this right - such action has to be 'in accordance with its other obligations under international law'. The provision on socio-economic considerations (Article 26) also makes reference to the other international obligations of Parties.

In the end, a lot will depend on the forum where any dispute is arbitrated. The WTO need not be the only forum where biosafety disputes are settled, as the CBD itself provides for a dispute resolution procedure, which is also applicable to the Protocol. Therefore concerned countries should, as a matter of priority, explore ways and means of defending the integrity of the Protocol in its implementation stages.

Dispute options under the CBD need to be further explored.

However, the US, the largest producer of GE organisms and products, cannot be a Party to the Protocol as it is not a Party to the CBD. Since it is unlikely that it will join the CBD, any dispute initiated by the US will ultimately be brought to the WTO, where dispute settlement decisions are increasingly in favour of the interests of big corporations and powerful countries. If this trend continues without urgent reform of the WTO, the embryonic Protocol will be stillborn.


Viewed from the perspective of biosafety and what most developing countries had wanted, many key issues had been compromised, and even lost. On the other hand, industry backed by a few governments had not wanted any Protocol at all, and fought against crucial principles and issues. They have had to accept the beginnings of international legal regulation.

The inclusion of the Precautionary Principle for decision-making on imports is a significant step forward. This was adamantly rejected by industry until the end.

So, while the Protocol may be starting with a limp, it is nevertheless the first step towards a system that can be strengthened. The spreading and intensifying of public awareness, more open scientific debate and political commitment by more governments can well nurture a robust international legal framework.

Lim Li Lin is a researcher at Third World Network.