Good technologies have few costs in the imaginary world in which we inhabit, bad technologies have no benefits, and all decisions are easy.
Daniel Kahneman, Thinking, Fast and Slow (TF&S)
Please don’t get me wrong: I’m not trying to be pro-nuclear. I’m just pro-arithmetic.
David J. C. MacKay, Sustainable Energy—Without the Hot Air
My interest in K-K and the psychology of nuclear power was first piqued by a fruitless search of the world’s favorite online bookseller for something, anything, in English and worthy of a read on the debacle of Fukushima Daiichi. Granted, it has only been 18 months since the events of 3/11 and the topic is a monstrous challenge, but what has been written is nothing but bilge. I’m going to pick on one book, Silence Deafening—Fukushima Fallout … A Mother’s Response, by one Kimberly Roberson. No, I haven’t read it, beyond what is available for free at the bookseller, but as the old saw has it, you don’t have to eat an addled egg to know it’s rotten. Here’s the beginning of the blurb on the back cover:
A CALL TO ACTION: Nuclear disasters and lessons learned. Facts are facts. There have been at least three major nuclear power disasters to date: Three Mile Island in 1979, Chernobyl in 1986, and Fukushima Daiichi’s unprecedented three nuclear meltdowns in 2011, the worst in history. … Do we wait for another life threatening catastrophic event, or do we act now?
So facts are facts, are they? What I adore about this pronouncement is that it is immediately followed up by a “fact” that is not a “fact”, but a highly contentious assertion, to put it mildly, that the meltdowns at Daiichi were “the worst in history”. The author’s “logic” appears to be that as there was only one meltdown at Chernobyl and three at Daiichi, Daiichi must therefore have been the world’s worst nuclear accident. It’s a matter of simple maths, you see. “Facts are facts” is one of my most beloved nonsensical expressions, right up there with “common sense”: it seems to obey the law of identity, that A is A, the first of the three classic laws of thought, known since at least the time of Aristotle; indeed it appears to be a tautology, but if your “facts” are not “facts” but “dubious assertions” or “downright lies”, you’re left with a very different pair of tautologies—and we haven’t even touched on the epistemological and historiographical slipperiness of facts. That “facts” might not be facts can be gleaned from the wondrous facts-are-facts.com, where we can learn from one Ursula Seiler that:
Jellyfish are essentially chiming bells that swim in the sea. Large jellyfish consist of entire melodies; small jellyfish individual notes … The increasing numbers of jellyfish appearing for example in the Baltic Sea is … a direct result of the ever-present music pumped out of our radios, department stores, etc. that makes up the soundtrack to our everyday life; this active music-making is chiefly what generates the existence of these creatures. Evidently, then, jellyfish epidemics are chiefly the result of mass-produced music.
Evidently. Or why not try this on for size:
Evidently, then, mass-produced music is chiefly the result of jellyfish epidemics.
It’s about as logical back-to-front as front-to-back. But to revisit the blurb:
There have been at least three major nuclear power disasters to date: Three Mile Island in 1979, Chernobyl in 1986, and Fukushima Daiichi’s unprecedented three nuclear meltdowns in 2011, the worst in history.
One of the fascinating consequences of Daiichi, and one that has gone wholly unremarked, is that it gave the world not one but two triptychs of calamity. To deal with this one first: notice how the word “three” recurs thrice in the sentence above—three disasters, one of which was Three Mile Island, and three reactor meltdowns at Daiichi. The triptych of calamity even embeds the word “three” within it: Three Mile Island, Chernobyl, and Fukushima. This is as fine an example as you’ll come across of the Rule of Three at work, a rule to which I ascribe an almost mystical power—take the Father, the Son, and the Holy Ghost, Brahma, Vishnu, and Shiva, and the world’s three monotheistic religions, for instance, or the three states of matter, gas, solid, and liquid, the three states of time, past, present, and future, the three primary colors, red, blue, and green, and the abundance of triumvirates, trilogies, and triunes, triads, troikas, and trinities, triplicities, tricoleurs, and hendiatris in our languages. The Rule of Three is, aside from my awful affection for alliteration (and a fondness for parenthetical asides), the only artifice I consciously employ in what I write, and if you’ve read this trio (so far) of posts from the start, you’ll have come across the Rule of Three at its merrily silent work, ooh, somewhere between 33 and 333 times already—although I’ll give you 3,333-to-one against that you’ll have noticed.
So, Three Mile Island, Chernobyl, and Fukushima—these are the Big Three nuclear calamities, then, this is a true triptych, yes? Ah, no, at least not if we accept the validity of the inevitably subjective International Nuclear and Radiological Event Scale (INES) of the IAEA, with the caveat that the IAEA is incorrigibly pro-nuclear. Chernobyl and Fukushima are rated “level seven: major accident”, the highest rating, but Three Mile Island is rated only “level five: accident with wider consequences”, a rating it shares with four other incidents, most of which will be unfamiliar to you in a way that Three Mile Island is not. But there’s a solitary “level six: serious accident”—the Kyshtym disaster in the (then) Soviet Union, on 29 September, 1957—so the true triptych should read Kyshtym, Chernobyl, Fukushima. Why don’t you—in all likelihood—know anything about Kyshtym? Well, first and foremost because the Soviet Union in the 1950s was not the most open—or safety conscious—of societies. Indeed, the accident is only known as Kyshtym because the east Ural city where it actually occurred, Ozyorsk (aka Chelyabinsk-40), was a closed city not on any maps, and while it now has a cartographical presence, it remains a closed city to this day, and it took some six months for news of the accident to filter out. Second, because 1957 is such an awfully long time ago, and because of the recency bias, our evolved human psychology is inclined to overweight the significance of recent events and underweight remote ones. And third, because unless you are a Russian speaker, you can’t pronounce “Kyshtym” (I believe it’s close to “Kuishtoim”), and words you can’t pronounce you can’t easily remember. Incidentally, you probably can’t pronounce “Chernobyl” either, but you think you can, and that’s good enough, whereas the unfamiliar Kyshtym, with its wall of consonants, looks unpronounceable, and that’s enough to intimidate.
So let’s compare the false triptych and the true:
Three Mile Island, Chernobyl, Fukushima
Kyshtym, Chernobyl, Fukushima
Do you notice anything? In the false, the accidents are spread geographically across three continents, the Americas, Europe, and Asia, and occur in two of the world’s three largest economies and (what was then part of) its largest country. In the true, two out of the three occur in the chronically closed and safety-contemptuous Soviet Union. Perceptions shift. These are, loosely and laterally, what Kahneman calls framing effects: put simply, the great difference in your reaction immediately before an operation you are about to undergo on being told either that that the survival rate is 90% or the mortality rate is 10%.
What if we add in the dates?
Three Mile Island (1979), Chernobyl (1986), Fukushima (2011)
Kyshtym (1957), Chernobyl (1986), Fukushima (2011)
Perceptions shift again: while the false triptych suggests a run-rate of a “serious” or “major” accident once a decade, as recency bias blocks out the nuclear quarter-century before 1979, the true suggests a run-rate of every couple of decades. This doesn’t mean that we should necessarily expect any run-rate, given the paucity of the denominator in the equation “cumulative years of safe global nuclear power plant operation divided by serious or major accident”, simply that this is how we psychologically perceive a run-rate.
An admittedly crude way of looking at historical accident rates is to assign the 25 INES level four to seven nuclear accidents that have involved far-field radiation releases since the dawn of the nuclear age a point score—say 5pts for a level four accident, 10pts for a level five, 15pts for a level six, and 25pts for a level seven—and break them out over the decades. Then we multiply by a thousand and divide by total net installed nuclear electrical capacity at end-decade to get a very rough metric of accident severity per megawatt, decade by decade.
1940s: 5pts (no commercial reactors in operation in 1949)
1950s: 50pts = 50,000 ÷ 548MW = 91.2
1960s: 40pts = 40,000 ÷ 14,121MW = 2.8
1970s: 40pts = 40,000 ÷ 117,814MW = 0.34
1980s: 40pts = 40,000 ÷ 311,942MW = 0.13
1990s: 5pts = 5,000 ÷ 347,368MW = 0.0144
2000s: 0pts NA
2010s (to date): 25pts = 25,000 ÷ 370,705MW (end-2009) = 0.0674
You don’t have to be Sherlock to notice a trend. However, if you don’t want to have any truck with the pro-nuclear IAEA’s INES, there’s an alternative: the nuclear accident magnitude scale (NAMS) developed by nuclear-skeptic David Smythe, professor of geophysics at Glasgow University. Here I add up the cumulative magnitudes, decade by decade, of the 17 INES level four to seven nuclear accidents for which he has been able to calculate a NAMS magnitude of severity and again multiply by a thousand and divide by total net installed nuclear electrical capacity at end-decade.
1940s: 3.8 (no commercial reactors in operation in 1949)
1950s: 21.4 = 21,400 ÷ 548MW = 39.1
1960s: 23.1 = 23,100 ÷ 14,121MW = 1.64
1970s: 16.9 = 16,900 ÷ 117,814MW = 0.143
1980s: 8.0 = 8,000 ÷ 311,942MW = 0.026
1990s: 4.8 = 4,800 ÷ 347,368MW = 0.0138
2000s: 0 NA
2010s (to date): 7.5 = 7,500 ÷ 370,705MW (end-2009) = 0.0202
The trend remains unchanged. (We might also observe here that there has never been an INES level four to level seven accident involving any reactor that started commercial operations after 1980 and only one that started after 1975 [Fukushima Daiichi No. 4].)
Let’s turn to the other triptych of calamity, this one wholly home-grown: Hiroshima, Nagasaki, Fukushima. Note the sibilant smoothness, punctuated by the repeated “shima”, with which the trio of four-syllable words trip off the tongue. This triptych is having a hugely potent effect on the Japanese psyche, with mayors from Daiichi-afflicted towns being invited last summer and this to address memorial services at Hiroshima and Nagasaki. Here’s an excerpt from a Japan Times article (August 28) by one Michael Hoffman, preposterously titled Complacency perished in the Fukushima nuclear disaster (as if the dirt of complacency had been simply wiped clean from the Japanese or human mind):
Neither the victims of Fukushima Prefecture’s triple meltdown in March 2011 nor the aging survivors of the world’s only two wartime atom bombings are letting that [the issue of nuclear devastation in peacetime] pass.
“In terms of being nuclear victims, we are the same,” Hiroshima survivor Sunao Tsuboi, 87, told the AFP news agency.
“In my mind, Fukushima is like a third nuclear victim, following Hiroshima and Nagasaki,” added Fukushima evacuee Sachiko Sato.
Nuclear devastation in peace is war, a 90-year-old Buddhist nun named Jakucho Setouchi goes so far as to say. Setouchi is a writer of considerable reputation whose collected works, published in 2002, run to 20 volumes. Speaking to Shukan Asahi magazine, she said, “The earthquake and tsunami were natural disasters, but (TEPCO’s Fukushima No. 1 nuclear plant) was a manmade disaster, and therefore the same as war.”
War happens when it is allowed to happen; ditto nuclear disasters. “The atmosphere today,” says Setouchi, “is exactly like 1941, ’42.” Back then, the public and mass media bought the official line that Japan’s victory was assured. In our own time, the public and mass media bought the official line that the safety of nuclear power was assured.
Leaving aside the grotesque syllogistic lapses (to say that all wars are manmade disasters, Fukushima was a manmade disaster, and therefore Fukushima was war is no different, (il)logically, from claiming that all cats are mammals, a cow is a mammal, and therefore a cow is a cat—this is the fallacy of the undistributed middle), it must seem the height of fatuity to the pro-nuclear power brigade—as it does to this more disinterested observer—to dare to mention Fukushima in the same breath as Hiroshima and Nagasaki. This is not for a moment to gloss over or belittle the very real torment, fear, and anxiety suffered by the 150,000 or so people whose lives and livelihoods were disrupted by Fukushima Daiichi, but this disruption lies at the far, far end of the wide, wide spectrum of human suffering from instant and involuntary vaporization. It just does. Nevertheless, instead of whining about this new triptych, it might be more productive to do some explaining of it. A useful pointer, I think, can be found in Stephen Pinker’s epic and hard to refute The Better Angels of Our Nature: Why Violence Has Declined, whose counterintuitive thesis is that we live in the most peaceful epoch of the history of our species.
Hiroshima and Nagasaki marked the culmination of the very interstate-violent short century of the Empire of Japan (1868-1945). It’s worth reviewing, briefly, that violence with some rough tallies of mortality. Killing kicks off with the civil strife of the Boshin War, (1868-1869, 3,000 deaths). A lull ensues until the First Sino-Japanese War and annexation of Taiwan (1894-1895, 50,000 deaths), followed by the Russo-Japanese War (1904-1905, 140,000 deaths) and the annexation of Korea (1910, deaths unknown). Then comes the Japanese involvement in World War I, 1914-1915, the seizure of German New Guinea, and the Siberian Expedition (1918-1922, 5,000 deaths). In the depths of the Great Depression come the invasion of Manchuria and establishment of the puppet state of Manchukuo, 1932. Then there are the two almost forgotten but history-changing confrontations with the Soviet Union, the Battle of Lake Khasan, 1938, and Battle of Khalkhin Gol/Nomonhan Incident, 1939, a “four-month long small war that … cost no fewer than 30,000 and perhaps as many as 65,000 casualties on both sides” (Alvin D. Coox, Nomonhan: Japan Against Russia, 1939). The Second Sino-Japanese War (1937-1945) bleeds into Pacific Theater in World War II (1941-1945), with around 18mn civilian and 6mn military deaths, of which Japan accounted for perhaps a million civilian and two million military deaths. So Imperial Japan was at war for 22 of the 52 years between 1894 and 1945 and preparing for, or recovering from war, for many of the rest, with a combined home islands death toll of maybe four million. And the death toll in interstate violence in the 66—and counting—years of peace between 1946 and the present? Zero. Very few major states have witnessed such a precipitous decline in interstate violence, and were we to investigate the plethora of forms of intrastate violence, from assassination as a political tool and state-sanctioned capital punishment to humdrum murder, rape, and robbery, we’d find similar pictures of precipitous decline.
With the gradual disappearance of violence, in all its forms, from Japanese society, there has concomitantly arisen the myth of safety—the myth that complex electromechanical systems such as trains, aircraft, and yes, nuclear power plants can simply never fail, despite being designed, built, and operated by ever-fallible humans. If the nuclear establishment was a willing purveyor of this myth, then the public was a willing buyer of it. Recently there has been a contretemps, with unpleasant nationalist overtones, about the deployment of the Bell Boeing V-22 Osprey tilt-rotor V/STOL aircraft in Japan in general but in Okinawa in particular, and the phrase of reassurance most routinely trotted out by politicians seeking to pave the way for deployment is that the Ospreys will not fly until “safety has been confirmed”. To be sure, the original Japanese expression, anzen wo kakunin, is an elusive one, and “kakunin” could be interpreted as falling just shy of “confirm”, but can there be anyone gullible enough to swallow the proposition that any aircraft could plausibly be declared unambiguously safe?
(to be continued)
(with thanks to A.E. for the Kyshtym tip-off)