For 40 years I've written about nuclear power, defended
it and believed, as I still do, that it offers the best
signpost to a great future; to what Churchill called the
"sunlit uplands"; in short, utopia.
I regard electricity as one of mankind's great achievements,
saving people from the menial, painful drudgery that marks daily
existence without it. Growing up in Africa, I'd see men and women
walking miles, many miles, barefoot across the savanna, looking
for a few pieces of wood to burn for cooking and hot water.
Electricity, I've believed for these four decades, is assured
for thousands of years through nuclear. With advanced breeder
reactors and with the energy stored in weapons plutonium, it
comes close to perpetual motion: So much energy from so little
The alternative choice is to burn up the earth, fossil fuel by
fossil fuel, until we are searching, like the people of the
African savanna, for something that is left to burn.
Wind and solar are defined by their geography and limited by
their scattered nature. Their place at the table is assured but
not dominant. Industrial societies need large, centralized energy
Yet a nuclear tragedy of almost immeasurable proportions is
unfolding in Japan. The sum of all the fears about nuclear is
being realized. Hades and Poseidon have joined to cut nuclear
Do disasters, like the Japanese nuclear one, really kill
technologies? Mostly, obsolescence does that; but their demise
can be accelerated by a last huge mishap.
While the Hindenburg disaster at Lakehurst, N.J., in 1937
didn't end lighter-than-air aircraft for passenger travel, it
drew the curtains: Fixed-wing airplanes were doing a better job.
The Concorde supersonic jet didn't leave the skies because of a
fatal accident at Paris-Charles De Gaulle Airport in 2000; but it
did make the Concorde's planned retirement immediate.
Conversely, Titanic's sinking in 1912 didn't put an end to
ocean liners: They got safer. Throughout the 19th century boilers
were constantly blowing up, not the least on the stern-wheelers
plying the Mississippi. Boats kept working and the technology --
primarily safety valves -- got better. Bad technologies are
replaced by safer ones; and good ones with flaws were improved
That is the history of boats, cars, planes and, yes,
resoundingly yes, of nuclear power.
After the Three Mile Island disaster in 1979, a new word,
"passive," began to dominate reactor design and construction, but
maybe too late for the General Electric Mark1 plants ordered so
long ago. Passive, as it sounds, is a design in which cooling
pumps are not as important. The idea is to depend more on gravity
feeds and convective cooling. These are featured in newer
designs, and there has been some back fitting. Things were moving
in the right direction, but not fast enough.
The story of the reactors at the Fukushima Daiichi site is a
story of success and failure. They were designed 40 years ago to
meet what in advanced design is known as a "maximum" credible
accident. That was, in that location, an earthquake of a
magnitude which had never occurred there. Excluded from this
calculation of credible -- i.e. it could happen -- was the
That exceeded the imagination of catastrophe to that point in
time. Within the credible design envelope, the plants performed
flawlessly. They shut down; the emergency cooling pumps started
up in fractions of a second; and when they failed, batteries took
over. The problem was the tsunami destroyed the diesel
generators, and the whole sequence of disaster began.
The opponents of nuclear power -- and they have been
pathological in opposition for more than 40 years -- have their
footwear on and are ready to dance on the grave of nuclear. They
might want to unlace and take a seat: Nuclear power does not have
an alternative about to retire it.
Big demand for new energy (ideally carbon-free energy) around
the globe, and especially in India and China, can't be satiated
without nuclear. Abundance of natural gas in the United States
already has reduced the demand for new nuclear to four or five
reactors. We'll be okay for a while.