Tomorrow's People

Home > Other > Tomorrow's People > Page 28
Tomorrow's People Page 28

by Susan Greenfield


  The most obvious image that springs to mind, thanks to the series of Star Wars films and other science fiction, is the ray gun, but surprisingly it will not appear in any futuristic armoury. The use of light to project and direct energy actually began in 1958. Lasers are good for communications guidance and range-finding, but the realization of the vision of ray guns, the direct lethal application of lasers, still has very serious obstacles. In order to destroy its target such a gun would have to shoot off large amounts of energy in short bursts. This enormous draw on energy reserves would necessitate a bulky and potentially vulnerable power plant. Moreover, the turbulence of the atmosphere would pose problems for the final accuracy of laser direction.

  But nanotechnology has clear potential in a large range of military uses. One idea is that supersensitive nanosensors will improve the scanning of remote landscapes like Afghanistan, whilst a very different application of nanoscience to warfare is in new types of materials for equipment; as one military man enthuses, ‘with nanotechnology, we can add properties to materials that weren't there before’. By 2025 soldiers under attack will be able to change their appearance to blend in with their surroundings: nanoparticles will change the properties, such as colour, of the materials of their uniforms. In addition, helmets will be 40 to 60 per cent lighter, and tent fabric will self-repair when ripped.

  Nanotechnology is currently bandied around as an umbrella term that covers the notion of merely very small devices right through to the manipulation of atoms that would defy the laws of chemistry. Inevitably, the wilder leaps of imagination create black scenarios of massive arsenals of self-assemblers churning out weapons of destruction. But this idea is not just technically difficult; it is, as we saw in the previous chapter, actually conceptually impossible. On the other hand, we should be wary of focusing our horizons exclusively on what is potentially feasible at the moment. In the future, even though the non-negotiable chemistry laws are hardly likely to be violated, there may be paradigm shifts of different kinds in military applications of fuel cells and other basic means of speeding up production that could all amount to a frightening escalation in the rate of weapon production.

  But the physical sciences cannot claim a monopoly in destructive potential. Even in ancient times armies relied on the power of illness to decimate the enemy, by routinely throwing the bodies of plague victims over the walls of besieged towns. And a particularly shameful incident in British history occurred when the USA was still under the King's rule, and the occupying force aimed to quell ‘disaffected tribes of Indians’: the British Commander-in-Chief, Jeffrey Amherst, distributed ‘gifts’ of blankets contaminated with smallpox. Much more recently, in 1984, the followers of the guru Bhagwhan Shree Rajneesh contaminated salad bars with salmonella as a deliberate act of biotechnological terrorism intended to reduce voter turnout at an upcoming election; although there were no fatalities, 751 people were poisoned.

  In 1995 approximately 5,000 commuters were injured on the Tokyo subway, and twelve were killed, as a result of nerve gas unleashed by a religious sect, Aum Shinrikyo. Nerve gas, originally developed in the Second World War, is not a gas as such but is either ingested as a solution or inhaled as a vapour. The toxin then works by blocking an enzyme (acetylcholinesterase) that otherwise breaks down a widespread and potent transmitter (acetylcholine) between nerve and muscle, which also operates between nerves and vital organs: the net result of the gas, therefore, is the excessive accumulation of acetylcholine at these crucial sites. Although many different processes in the brain and body depend on acetylcholine in normal amounts, the Dr Jekyll transmitter can also transform into a biochemical Mr Hyde if it is not removed from the site of the action by its special enzyme. If acetylcholine is allowed to continue stimulating its target muscles, then it is a little like pressing on the accelerator so much that the engine stalls. The muscles all over the body stop working. The results of inhalation of nerve gas include tightening of the chest with more widespread symptoms setting in, if vapour concentrations are high enough, in less than a minute. A mildly exposed casualty will experience anxiety, headache, pain behind the eyes and restlessness; a higher level of exposure can cause twitching, cramps and general muscle weakness. A victim exposed to very high levels will have slurred speech, confusion, then slip into a coma and finally die from a complete shutdown of the respiratory system. A further concern, however, is that even those surviving modest levels of exposure might still be at risk in the long term. The effects that even low concentrations of nerve gas could have on the delicate and dynamic connections between neurons, which constitute the essence of an individual, might accordingly lead to much longer-term changes in one's state of mind.

  Another, very different form of chemical warfare that was also developed in the early 1940s as part of the war effort, but is only now realizing its hideous potential, is anthrax. The recent anthrax attacks in the USA, in the wake of the 9/11 tragedy, left five dead and seventeen infected. However, further exploitations of nanotechnology could eventually make agents like anthrax harmless or detect it in public areas before it could do any harm. For example, at Swinburne University of Technology, in Melbourne, artificial ‘muscles’ are being built that could give early warning of biological weapons.

  The principle is ingenious, yet simple. The contraction of muscle needs two proteins to slide past each other: actin and myosin. Anthrax will slow down and eventually stop this sliding. In order to monitor this action, some myosin is attached to a biochip, so that any decrease in the movement of adjoining actin will register. The chip is coated with a polymer to prevent any random sticking; a laser then etches straight lines on it, enabling the myosin molecules to stick to the chip in the tracks. Now actin molecules are added. Normally they slide along the myosin tracks, but if they come to a halt, then anthrax must be present.

  Anthrax is a bacterium that, when inhaled at a rate of some 2,500 spores an hour, causes respiratory distress and death a few days later. Its action is based on two enzymes (lethal factor and oedema factor) which break down the defence mechanisms of cells. However, a third component is necessary for the toxin to do its work: a protective antigen (PA) that enables the enzymes to gain entry to cells in the first place. Genetic modification of this chemical can make the toxic enzymes useless, as they cannot gain access to the target of their action; so far this mutant PA has protected rats from an otherwise potentially lethal injection of anthrax. This is only one example of how the genetic technologies are not, as they are often simplistically perceived today, automatically bad. Yet though they may be used to combat biological or chemical warfare they can also, of course, inspire a whole new range of deadly substances.

  In the future as the synthetic genome becomes better understood it may be pressed into service to manufacture completely novel bugs with lethal actions. Whether artificial or modified from natural genomes, ‘stealth viruses’ could be covertly introduced into the genome along with ‘designer’ diseases. Even before molecular biologists are able to produce diseases to order they could still set in train a course of events that has disastrous consequences – by accident or by design.

  For example, Australian scientists Ron Jackson, of the Commonwealth Scientific and Industrial Research Organisation, and Ian Ramshaw, of the Australian National University, in a novel attempt at pest control, recently inserted the IL-4 gene into the mousepox virus to boost rodent antibodies, making the rodents infertile. Mousepox normally has only mild ill effects but the new gene converted it into a killer disease. How easy to imagine, then, the possibility of making a similar genetic modification to human smallpox to produce a new and even more deadly form of the virus.

  Even more sinister still is the prospect of exploiting particular genetic profiles of different races for selective destruction. The issue of genocide as a conspicuous factor in modern-day warfare arose in the late 20th century, for instance in Rwanda and Bosnia. Now ‘ethnic cleansing’ has unfortunately become a familiar term in our vocabulary. If, as cu
rrent trends suggest, war in the future will no longer be fought for territory so much as for cultural or racial homogeneity, then we might expect ever more focus on ethnic targeting. Indeed, several governments may already have worked on the creation of race-selective diseases. For example, in the apartheid South Africa of the 1980s, it later emerged, scientists studied the possibility of a disease that would render only black women infertile. Fortunately, however, the creation of such conditions or diseases would be very difficult; as the Human Genome Project showed, the genetic differences between races are far slighter than they might on the surface appear.

  Because of this underlying genetic similarity it is hard to see how such a strategy could be highly effective on a wide scale. Given that most of us have a complex genetic cocktail, it would be difficult to select a sufficiently conserved yet highly selective gene that could be targeted within any one race. True, there are certain societies, in Iceland or indeed the Amish community in the USA, that have already helped the study of genetic-based disease by virtue of their relative, and very unusual, genetic sameness. But even so, there is a big difference between tracking the appearance of a disorder through generations and identifying a special gene that is common to, and operative in, the majority of any one population – above all exclusive to those particular people and only thus appropriate as a target. Such ethnically selective genes, for example for sickle-cell anaemia among Afro-Caribbeans or Tay-Sachs disease within the Jewish population, occur only in a small minority and thus, from a highly cynical stance, would still only result in the deaths of relatively few if targeted.

  Indeed the very opposite of such precision targeting looks to become more likely if current trends in patterns of terrorist attacks continue; a primary feature is the indiscriminate choice of victim. The terrorist aims simply to deploy chemical or biological agents that do harm. In laboratories the manipulation of many biological agents is now routine. But for bacteria there appears to be a pay-off: as one feature is improved another usually declines – as scientists enhance infectivity, for instance, it may be at the expense of the organism's ability to survive in the environment. Nevertheless we cannot afford to be too complacent that the prospect of new toxic substances will not loom in the near future. Not only might we see more virulent viruses, such as the once mild mousepox, with new genes incorporated into their DNA but we should also expect the development of more effective delivery systems, such as micro-encapsulation.

  In any event, the ideal will always be to avoid contact with any toxins from the outset. Vaccination will never be a satisfactory defence, because many vaccines are only temporarily effective, and also because it is hard to know in advance what agents to protect against. Rather, not too far into the future, we shall probably see new air-filtration systems in offices and homes, along with some 21st-century version of the gas mask kept alongside the bathroom first-aid cupboard. Air quality might well be monitored, with online flashing read-out just as the temperature and time already blink from neon billboards in any city centre. Awareness of the air we breathe may soon become as routine as the sensitivity we have all developed to unattended packages – checking air quality from different monitors, perhaps even instruments on our bodies, may become as frequent and as automatic as checking the time.

  Bacteria will be more on the minds of our grandchildren than bullets. Pitched battles will become increasingly expensive, relatively inefficient and pointless – as territorial disputes are subordinated to ideological ones more suited to acts of terrorism; germ warfare is a far more obvious option. One crumb of comfort is that the Biological Weapons Convention of 1972 prohibits the manufacture or use in war of toxic substances – but then again, it can only urge signatories to respect the resolve. No one nowadays, as the power of the UN is challenged, feels reassured by toothless talking. Moreover, attack with chemicals does not necessitate a diversified industrial base nor is it expensive, and above all it allows the perpetrator to escape unidentified. The same ‘advantages’ also apply to cyber-crime and cyber-terrorism.

  A staggering two-thirds of the companies in the UK have been hit by cyber-crime in the last twelve months, the main threat coming from external hackers. Deriving as it does from computer use, cyber-terrorism in particular plays on our deep, atavistic fears of technology transforming the computer from slave into master. Yet, despite the plots that Hollywood weaves, the real controller is not, of course, a machine intent on taking over the world, nor a beefy combatant in khaki, nor a slick 007-style buccaneer but rather a Person of the Screen, a thumb-pointing, finger-quick youth. Interestingly enough, typical hackers are usually computer nerds, not necessarily of the same mindset as the terrorists. Apparently, a neutral IT-whizz is usually recruited by a fanatic, rather than the other way round. But as keyboards and screens become a feature of more and more homes throughout the world IT-agility will be a given rather than a selling point for a would-be cyber-terrorist.

  The particular power of cyber-crime, and especially cyber-terrorism, lies in its linking of the physical and virtual worlds. Ultimately, it is economic disruption or physical damage to people or places that will weaken the enemy; and you can steal or destroy in the physical world more effectively and with far less risk either to your person or of being identified with a keyboard than with a bomb or a gun. For example, in a Pentagon-run exercise hackers posed as a hostile Korean force waging a cyber-war against the USA. Using only equipment available on the shelves of computer stores or on the internet, and denied any advance intelligence or security clearance, these hackers were able, within several days, to shut down power grids and 911 call-centre systems to twelve major cities, including Washington. They also gained root access to six systems at the Department of Defense. From logistical disruption the hackers created communications chaos by sending fake messages from the President and Joint Chiefs of Staff: the result was to paralyse the American military, as every order now required visual or verbal verification.

  Clearly the potential for cyber-technology to offer a non-kinetic means of paralysing traditional, kinetic weapons is huge. Even conventional warfare in the Information Age needs complex planning and coordination, depending on access to massive databases. Information warfare is not really about the hardware, whether spy satellites or computers, but about changing their human operators – or, more specifically, the decisions they make. And then there are the facts themselves. A small change made to a salient fact can have devastating consequences. An astrophysicist once remarked that the worst that could befall him professionally would be for someone to alter the fifth decimal place in the constant pi: all subsequent calculations would then be flawed, and all his work useless. The mantra of the late 20th century, and the deadly reality in this one, is: information is power.

  Everyone has known since classical times that false information is a powerful weapon. But the destabilization of economies and of governments through the spread of false information has now acquired a more potent feature in that it is no longer obvious. Unlike the Nazi propaganda broadcasts of Lord Haw-Haw and the traditional strategy of dropping leaflets from a plane, a communications glitch in your IT-system may be the result of a deliberate, hostile attack, the work of a lone loony hacker, or simply an accidental technical malfunction. Another possible advantage for cyber-terrorists is the fact that computer systems can be doctored to garner data from other computer systems with which they interact. Some think that this is already happening.

  As terrorism supplants conventional warfare and nations are no longer safe behind fixed borders the all-pervasive cyber-world, oblivious to territorial boundaries, will prove the new battleground. Already conflict has started for real in cyberspace, or, more accurately, for virtual real. An ‘e-jihad’ broke out in October 2000 after Israeli hackers targeted a Hizbollah website that had been trying to incite anti-Israel violence. By the end of the month hackers on both sides had destroyed as many as thirty websites. But things could get far worse.

  Barry Collin, a senior rese
arch fellow at the Institute for Security and Intelligence in California, bleakly describes how wide the scope might be to fulfil the aims of the cyber-terrorist ‘to destroy, alter, acquire and retransmit’ information. For example, they might plan to access the control systems of a cereal manufacturer to change the levels of iron supplement in the cereal, so that the children of a nation were killed. Another of Collin's nightmare visions is that they could place rings of computerized bombs in a city: if one bomb stops transmitting to others, then the rest go off simultaneously. Another possibility is the disruption of banks and economic systems, so that subsequently destabilized societies lost all confidence in their financial infrastructure; or the cyber-terrorist might hack into air-traffic-control systems and cause two civilian aircraft to collide. They might alter formulae at pharmaceutical factories, or change the pressure in gas lines so that a suburb detonates and burns. In short, concludes Collin, the cyber-terrorist could prevent a nation from eating, drinking, moving or living – all without warning.

 

‹ Prev