On the decision to use nuclear power, the erstwhile and long-time director of Oak Ridge National Lab, Alvin Weinberg, warned that the society had made a “Faustian bargain”. Atom provides a limitless supply of energy, which is not only cheaper than fossil fuels, but is also less polluting. But the risk from nuclear power plants demands vigilant inspection and careful control. We cannot treat nuclear power as casually as some other technologies, such as coal-fired power plants.

Although Weinberg’s statement of the 1940s might be outdated today, the issues that he raised are current and alive. The big question is, where should the society draw a line while making these technological Faustian bargains? Since Weinberg’s statement, the society has witnessed at least three major technology-related accidents, such as the one in Bhopal, India, the Three Mile Island in the US and the Challenger space shuttle.

It is evident that newer technologies hold both the promise and the potential for far greater disasters. With the advent of the telecommunication system and computer technology, a major breakdown, such as those that occasionally hit long-distance telephone systems, could cost billions of dollars and even kill some people, depending upon the type of devices the networks use in the future. This raises the question as to how we can make today’s technologies safe.

Today, the engineering view of a machine is that which is “designed, built and maintained well and as long as operators follow correct instructions, and should perform well”. This is a very naive view of looking at things. The safety of complex systems depends not just on their physical characteristics but also on the people and organisations operating them. Operators who are trained to go by the book may freeze at the time of calamity or misinterpret a situation and thus take the wrong action. This is what happened in the Three Mile Island incident, where the operators shut off most of the emergency cooling water going into the core for what seemed like a good reason at the time of the incident. Thus, an organisation’s reliability is as important as the reliability of the equipment under consideration.

Here there are two schools of thought. Yale sociologist Charles Perrow argues that accidents are unavoidable in certain hazardous technologies, such as nuclear power plants, nuclear weapons and so on. According to him, these technologies place mutually contradictory demands upon the organisations running them. So the organisation, as a whole, requires centralised control. Perrow’s arguments have been influential, but they are not the last word, as a group of researchers believe that organisations perform better than what Perrow thinks.

As per Berkeley political scientist Todd La Porte and colleague Paula Consolini, these organisations “work in practice but not in theory”. La Porte and Consolini studied the so-called high-reliability organisations or HROs and found that there were reasons that such organisations work so well. If Perrow is right, then it is best if some technologies are left unused. However, if La Porte is correct, then we may not be able to avoid accidents altogether but learn to make them as unlikely as possible and, hence, keep our end of the bargain. However, if we follow La Porte, it won’t be easy, as managing complex hazardous technologies demands making a sustained and conscious effort far greater than most organisations are accustomed to.

To take an example of what happened in the early hours of December 3, 1984, in Bhopal, when the lethal methyl isocyanate gas leaked out of the Union Carbide plant. Thousands of people died and many were injured, not due to the mistake of one or two people but of the organisation as a whole. It was the system that had failed. Similarly, the Challenger disaster was not the result of the O-ring’s failure but that of NASA, which did not attend to the warning of engineers from Thiokol.

NASA’s problems with the space shuttle programme are far removed from Union Carbide’s troubles at the Bhopal plant, but the stories do have several common features. Both institutions were much less concerned about safety than they should have been, thanks to complacency or a failure to appreciate the risks. Both dealt with complex technologies whose behaviour was impossible to predict, although in each case, the accident could have been avoided with a little extra caution on the part of the organisations concerned. Both organisations were facing financial pressures that pushed them to attempt more with less. What these organisations failed to realise is that risky technologies are a recipe for disaster.

The need of the hour in the advent of new and rapid technologies is to build high reliability organisations. Such organisations can be seen in the nuclear aircraft carriers of the US Navy and air traffic control on airports, which have learned how to deal with complex and hazardous technologies by effective communication. Such organisations not only provide a counter to Perrow’s argument but also inculcate a culture among its employees that fosters a spirit to deal effectively with crisis whenever it arises.