Sunday, September 14, 2008

A 1-in-1,000 Chance of Götterdämmerung

Source: Reason online

Will the world come to an end on September 10? That fear is motivating two lawsuits—one American, another European—that aim to stop the physicists at the European Center for Nuclear Research (CERN) from switching on the Large Hadron Collider (LHC) on that day. The LHC is a $10 billion 17-mile long particle accelerator lying in a circular tunnel beneath the border of France and Switzerland. Its massive superconducting magnets cooled with liquid helium accelerate two beams of protons and lead nuclei to nearly the speed of light. These particle beams will eventually be crashed into each other to produce temperatures and particles not seen since microseconds after the Big Bang 13.7 billion years ago.

One of the chief goals of the LHC experiments is to find the elusive Higgs boson, the only fundamental particle predicted by the Standard Model of particle physics that has not been directly observed. The Higgs boson plays a key role in explaining the origins of mass in other elementary particles. Exciting, if esoteric research, to be sure, but why oppose it?

Walter Wagner, a former nuclear safety officer, and Spanish science writer Luis Sancho, have filed a civil suit in federal district court in Hawaii asking for a temporary restraining order to stop the researchers at CERN from switching on the LHC until further safety analyses are completed. In Europe, Professor Otto Rössler, a chemist at the Eberhard Karls University of Tübingen in Germany filed a similar suit with the European Court of Human Rights.

These LHC opponents fear that the Earth could be destroyed by vacuum bubbles, magnetic monopoles, microscopic black holes, or strangelets produced by the high-energy proton-proton collisions planned by CERN physicists. Vacuum bubbles have been described as a kind of "cosmic cancer." If it turns out that there is a lower energy state into which the universe could settle, then the LHC might produce "bubbles" of such a state which would then expand, ripping apart the Earth and eventually the entire universe. If magnetic monopoles were produced they might induce protons to decay and thus destroy normal matter. Microscopic black holes might grow by gobbling up the Earth. And strangelets are combinations of quarks that theoretically interact with normal matter and transform it into strange matter.

At the Global Catastrophic Risks conference at Oxford University this past July, CERN's Michelangelo Mangano described the findings of a report released in June by the LHC Safety Assessment Group (LSAG). The bottom line: "There is no basis for any conceivable threat from the LHC."

While the LHC safety report goes through a number of scenarios, its chief point is that the energies produced in the LHC are "far below those of the highest-energy cosmic-ray collisions that are observed regularly on Earth." In fact, cosmic rays produced by phenomena in the universe "conduct" more than 10 million LHC-like experiments per second. If such energies actually produced vacuum bubbles, microscopic black holes, magnetic monopoles, or strangelets that could destroy planets and stars, physicists wouldn't be here to perform experiments in the LHC now.

At the Global Catastrophic Risk conference, Future of Humanity Institute research associate Toby Ord asked an interesting question: How certain should we be about safety when there could be a risk to the survival of the human species? As Ord argued, "When an expert provides a calculation of the probability of an outcome, they are really providing the probability of the outcome occurring, given that their argument is watertight. However, their argument may fail for a number of reasons such as a flaw in the underlying theory, a flaw in their modeling of the problem, or a mistake in their calculations."

In other words, for the argument that the LHC poses no existential risk to humanity to be sound, the theory underlying it must be adequate. But physical theories have been upended in the past. Ord pointed out that Lord Kelvin had calculated the age of the sun. Using the best physics of his time, Lord Kelvin concluded that the sun was 100 million years old. It was not until the discovery of radioactivity that the current estimate of 4.6 billion years could be calculated. So Ord argued that it's not unreasonable to think that there is a 1-in-1,000 chance that the theories underlying the LHC are flawed in some important details.

In addition, the model of the problem itself could be flawed. As an example of how flawed models can impact the real world, Ord cited the Castle Bravo 15-megaton thermonuclear bomb test in 1954, the explosive yield of which was two and half times what had been calculated by the bomb's designers at Los Alamos National Laboratory. Those experts had missed the fact that the lithium-7 isotope, when bombarded by high energy neutrons, decomposes into tritium and boosts neutron production. As a more recent example, Ord claimed that Lloyds of London's insurance models for New Orleans had failed to consider the risk that the city's levees might fail.

And finally, it's possible that errors in calculation could slip into errors of analysis. Ord cited the frequency of miscalculations in medication dosages as an example of such errors. To get an estimate of argument failure, Ord cited survey evidence which found that 1-in-1,000 to 1-in-100 articles are retracted from high-impact scientific journals. For an article to be retracted something must be found to be seriously wrong with it. "If the probability estimate given by an argument is dwarfed by the chance that the argument itself is flawed, then the estimate is suspect," argued Ord. He suggested that multiplying the probabilities that the theory, model, and/or calculations on which the operation of the LHC rests are wrong dramatically increases the probability estimates that switching it on will destroy the world. Thus Ord concluded that the LHC should not be switched on.

Mangano from CERN objected furiously to Ord's presentation, arguing, "I can apply that estimate of a 1-in-1,000 chance to everything." Ord responded that his analysis should only apply to experiments that pose an existential risk to humanity, not to experiments whose outcomes can be ameliorated later. I asked Ord if he could think of another experiment or situation to which he would apply his analysis. He looked surprised for a moment and then reluctantly said, "No." Over canapés after Ord's talk, several of his colleagues expressed glee at the prospect that a philosopher's arguments might derail a $10 billion physics experiment. Personally, I estimate the probability of that happening at less than 1-in-1,000.

As intriguing as Ord's argument is, I am ultimately unpersuaded by it. Why? Largely because the empirical evidence is that the universe has been running trillions of these high-energy physics "experiments" for billions of years without disastrous results. In fact, Ord's colleagues Nick Bostrom and Max Tegmark from the Massachusetts Institute of Technology calculate that the empirical evidence suggests a conservative estimate of the annual risk that LHC-like experiments would destroy the earth is 1-in-a-trillion. At the end of his talk, Mangano reminded the Oxford conferees, "Jeopardizing the future of scientific research would be a global catastrophe." Any theory, model, or calculation that suggests otherwise is clearly flawed.

Ronald Bailey is reason's science correspondent. His book Liberation Biology: The Scientific and Moral Case for the Biotech Revolution is now available from Prometheus Books.

No comments:

Rate this blog