Mon, 02 Dec 1996

The threat of biological weapons

LONDON: In the early 1990s, it became public knowledge that the former Soviet Union - despite being a depositary state, along with the U.S. and UK of the Biological and Toxin Weapons Convention (BTWC) - had a large-scale offensive biological- weapons program following the 1975 entry into force of the Convention.

Furthermore, by the time of the 1991 Gulf War, Iraq had developed a comprehensive range of biological and toxin agents and munitions. More recently, it was revealed that the Aum Shinrikyo sect which used Sarin nerve gas to attack the Tokyo underground system on March 20, 1995 had experimented with anthrax bacteria. Along with significant advances in microbiology, genetic engineering and biotechnology, these revelations have placed the threat of biological weapons in a new and deeply worrying perspective.

Following the 1991 Gulf War, major Western states, including the U.S. and the UK, recognized this new danger, and placed greater emphasis on developing better means of field detection. International efforts to restrict the proliferation of biological weapons were primarily focused on the BTWC. The key issue now facing the Fourth Review Conference - scheduled to take place in Geneva, Switzerland this week - is whether the Convention can be effectively strengthened to prevent the proliferation and potential use of these deadly weapons. Given the complexity of the issues involved, a concerted international effort to achieve this objective will be required.

Historically, there has been sporadic use of biological weapons, but only in this century have large-scale, offensive biological-weapons programs been undertaken by major states. It is known, for example, that the U.S. had one such program for over 25 years from the World War II until the end of the 1960s. During that period it weaponised bacteria such as anthrax, viruses similar to that which causes Venezuelan equine encephalitis (VEE) and toxins including botulinum.

Bacteria are tiny, single-celled living organisms; viruses are very small elements of hereditary material, surrounded by a protective protein coat, which cannot replicate except inside a living cell. Because these agents multiply rapidly in the victim's body, very small amounts are required to achieve a major effect. Toxins are chemical substances, not always derived from living organisms, but they can also be deadly in extraordinarily small quantities.

Nature provides a weapons designer with a wide choice of agents: VEE, for example, has a low lethality in humans but it is regarded as an incapacitating disease. A huge variety of agents are available for attacking livestock or plants - to inflict economic damage, rather than to affect personnel directly. For the latter, a number of distinct target types and methods can be envisaged, but one particular characteristic of biological and toxin weapons is prominent. It is clear that in attacking humans via their lungs, by spraying from an aircraft and allowing the wind to carry the agent to the target, biological weapons can achieve an area coverage which puts them in the class of weapons of mass destruction - rivaled only by nuclear weapons. While some agents (such as plague) could cause uncontrollable epidemics, others would spread very little from the first victim and their use would thus have more predictable outcomes.

Although both are weapons of mass destruction, there is one very obvious difference between nuclear and biological weapons. Producing fissile material requires a massive industrial effort which few potential nuclear proliferators have achieved. In contrast, biological weapons are relatively simple to produce. The United States Office of Technology Assessment stated in a 1993 report that, with a small vial of seed culture and a simple fermenter, kilograms of an agent like anthrax could be produced in as little as four days.

Developing effective weapons that disperse agents in particles of the right size and viability is more difficult - but not impossible. The UN Special Commission (UNSCOM) has reported that Iraq had filled shells, bombs, rockets and missile warheads with biological and toxin agents. In 1995, UNSCOM noted that the program covered a whole variety of biological-weapon delivery means, from tactical weapons (such as 122mm rocket and artillery shells) to strategic weapons (including aerial bombs and Al- Hussein warheads filled with anthrax, botulinum toxin and aflatoxin) and 'economic' weapons (for example, wheat cover smut which affects crops).

The modern revolution in biotechnology has had a major effect on production capabilities. For example, the ability to introduce a foreign toxin gene into a bacterium has allowed very much larger amounts of toxins to be easily produced than was previously possible by extraction from natural sources. Additionally, the growth and spread of the biotechnology industry has led to much more sophisticated capabilities being developed and used around the world. There is clearly now a much greater potential for the rapid production of agents in smaller, less easily identifiable, facilities.

In comparison with arms-control treaties developed since the 1970s, the BTWC is a slight document of just 15 articles. The heart of the Convention is embodied in the general purpose criterion of Article 1 which states that:

'Each State Party to this Convention undertakes never in any circumstances to develop, produce, stockpile or otherwise acquire or retain:

Microbial or other biological agents, or toxins whatever their origin or method of production, of types and in quantities that have no justification for prophylactic, protective or other peaceful purposes...' (emphasis added)

No verification provisions were included which can help to ensure that the States Parties - now 139 - are abiding by their undertakings. As concerns grew in the 1980s over possible proliferation and the rate of change in biotechnology, an initial attempt was made to tighten up the Convention. At the Second Review Conference in 1986, a series of confidence-building measures (CBMs), in the form of politically (though not legally) binding annual data exchanges, were agreed. These were refined and extended at the 1991 Third Review Conference, but the overall outcome has been disappointing. Far too few of the States Parties have submitted regular data and the information supplied has often been inadequate.

Partly in response to this lack of success, the Third Review Conference also initiated attempts to determine whether some tougher system of verification might be found and agreed. As a first step, a series of meetings were held to examine possible verification measures from a scientific and technical viewpoint. The findings were considered at a Special Conference in 1994 which mandated a further series of meetings to consider proposals for a legally binding instrument. These meetings continue.

The whole verification enterprise is bedeviled by a variety of differences. One view is that a restricted technical concept of verification, similar to the one used in nuclear arms control (that is, counting individual systems), should be applied to the problem of containing the proliferation of biological weapons.

A second argument is that the international community should concentrate on deterring militarily significant offensive biological-weapon programs. These programs will still involve production of relatively large amounts of material, in facilities different from those of purely civil enterprises, and end in operational testing and training. Rather than allowing the international norm against the use of biological weapons to be eroded, this wider view urges mobilization around the norm to develop a 'web of deterrence'. Such a web would depend on enhanced intelligence, export controls, detection and protection capabilities and a willingness to respond effectively to the use of biological weapons.

Crucial to this view is the notion that the BTWC must also be strengthened by adding a legally binding protocol. Proponents accept that total transparency is unattainable but, given the characteristics of a significant program, they consider that a well-constructed protocol could, even without 100 percent technical certainty, greatly enhance the possibility of detection and therefore deter states from embarking on proliferation.

Proponents also argue that this can be done in a cost- effective manner, certainly well below the verification costs of the Chemical Weapons Convention (CWC), by carefully choosing parameters of the protocol. Opponents of this view remain unconvinced, however, arguing that the deterrent value of a protocol will be small unless it is very intrusive and if it is effectively intrusive it will be very costly and an unacceptable threat to commercial proprietary rights.

Some hold the view that the introduction of a verification protocol could damage rather than enhance security. UNSCOM's experience in Iraq has shown how difficult it is, even with an extraordinarily intrusive inspection system, to discover elements of a biological-weapons program. Some see a danger in introducing a verification protocol in that a global system would not be intrusive enough to deter cheating and indeed may even provide a cover to mask illegal programs. However, countries such as the UK, through a program of trials, have sought to demonstrate that managed access techniques can be employed to achieve effective on-site inspections while protecting commercially confidential information. Work in other countries has led to support of the UK's evaluation, and is likely to lead to practical developments in the verification sphere.

The Fourth Review Conference will follow the pattern set at previous Conferences, reviewing the operation of each of the Articles of the Convention. Although Article 1 specifies that purposes are prohibited, not things, and thus that all new technologies are covered, previous Review Conferences have sought to emphasis the point in regard to potentially dangerous new developments.

Consequently, toxin production was singled out for special mention in the Final Declaration of the Second Review. Modern genome research and its possible implications for example, the production of 'ethnic weapons' which would be based on the increasing understanding of the diversity of the human genetic make-up in different populations, could similarly be mentioned in the Final Declaration of the Fourth Review. Attempts are also likely to be made to secure more universal adherence to the Convention, to develop some details of the agreed CBMs.

Given the clear dangers to international security from the proliferation of biological weapons, the success or failure of the Fourth Review Conference will be judged mainly by what it does in regard to verification. Nevertheless, it is unlikely that the differing views will be resolved at the Review Conference. The most crucial issue is whether the States Parties can agree to endorse progress made since the 1994 Special Conference on a legally binding instrument and set an early date for its completion. The debate on verification seems set to continue in 1997.

Window A:

Producing fissile material requires a massive industrial effort which few potential nuclear proliferators have achieved. In contrast, biological weapons are relatively simple to produce.