A couple of weeks ago the Associated Press ran a story that seemed to confirm the world’s fears about Iran’s suspected nuclear weapons program. Central to the story was a plot (shown here) that showed, over a period of a few microseconds, energy and power output from a nuclear weapon. While everyone seems to agree that Iran does not yet have a nuclear weapon, calculations such as those plotted in this graph are an important (dare I say “critical”?) step in the process of designing these devices. And the scale on the axes was certainly thought-provoking – the right axis showed an energy of 50 kT; more than three times the energy released by the Hiroshima device. According to this plot, not only was Iran developing nuclear weapons, but they were looking at something with the output of our second-generation (i.e. post-WW II) nuclear weapons.
When I first saw this plot I was surprised at the calculated output – our first nuclear test was a fraction this size, as have been the initial nuclear tests of every other nuclear power. To think that the Iranians were already looking at a 50 kT device was a surprise. But the shape of the graphs seemed reasonable, the time scale seemed in line with what I’d expect, and the article quoted some well-known scientists and organizations who took the plot seriously. I looked at the plot, read the commentary, and it reinforced my suspicions that the Iranians are continuing to work on nuclear weapons. In my mind, it all fit together into the mental picture I’ve been assembling. And then I went back to work.
Yesterday, then, I was sort of surprised to read a snippet that said the graph was wrong. Skimming the article I read a line that mentioned one of the scales was off by a factor of a million and, looking at the plot, I immediately saw the problem (for those of you who are wondering…the numbers on the left axis are in units of kT per second – you get these numbers when you divide the yield (in kT) by the time shown on the x-axis (in seconds). And the numbers are about a million times higher than they ought to be. Dividing 50 kT by 2 microseconds should give a value of about 25 million – not 25 trillion as shown on the plot. Blindingly obvious when you’re looking at the plot and checking it for flaws – but nothing that’ll jump out at you unless you’re looking for mistakes. What I saw – and I’m willing to bet that what most people saw – were two lines that seemed to have the right shape, a surprisingly high (but by no means implausible) yield, and about the right time span. Since so many things seemed right about the plot – and because it fit my assumptions – I never thought to check the math. Bummer.
So – one thing that’s interesting here is that I (and apparently a number of others) were inclined to accept this plot without a whole lot of checking because it fit our expectations. Consider – if I see a horse-drawn carriage in Central Park it fits my preconceptions of the sort of thing you see around Central Park, and I’m likely to assume that whoever’s in the carriage is a tourist. If I take a quick glance and see a couple snuggling under a blanket it only confirms my preconceptions – but I’ll probably never go up to ask to peek at their drivers’ licenses to make sure. We are all very ready to gloss over details as long as the big picture matches our expectations.
Part of this is almost certainly due to our evolution as a species that profited from pattern recognition – recognizing patterns helped our earliest ancestors to make sense of the world and to survive. So recognizing that really big bears lived in caves helped them to avoid becoming cave bear chow, just as recognizing that inserting a spear into in the right spot on a mammoth could lead to dinner. Once we have recognized a pattern, that expectation becomes a sort of mental short-cut that lets us save time (or to stay alive) – pattern recognition is engraved in our DNA.
The problem is that it works too well and sometimes leads us to see patterns that don’t exist. Our expectations about Saddam Hussein helped us to see evidence of a WMD program that didn’t exist. And my expectations that Iran might be developing nuclear weapons inclined me to believe the authenticity of this plot.
Our expectations aside, it’s still interesting to think about the plot, where it came from, and what it might mean.
Take the simplest thing – this was almost certainly not the product of a computer program because a computer would never have made the mistake of being off by a factor of a million in a simple calculation. The plot looks to have been printed with a computer and printer, but the values in the left y-axis were probably entered by hand rather than by calculation. This, in turn, suggests that whoever put this plot together (and leaked it to the AP) was likely not a scientist or an engineer – it’s not as hard as you might think to be off by a factor of a million, but that sort of mistake usually doesn’t make it past the first line of review. So this plot seems most likely to have been put together by a non-scientist, perhaps using information provided by scientists or maybe one who read enough to get the superficial generalities (the stuff that caught my eye) right. If this plot was really constructed by Iranian scientists then I’d have to question their ability to design a working nuclear weapon – but it seems more likely that this was never in close proximity to a scientist. Or at least, not one who bothered to check the math (and if this is a “real” plot then we might not have to worry as much as we have been about Iranian nuclear weapons!).
Of course, if the plot wasn’t developed by a scientist then we’ve got to wonder who made it up, and why. Was the intent to discredit Iran? To justify an attack against Iran? To push the US and Europe into additional sanctions? To help Iran dissuade potential attackers, or to help Iran look stronger and more advanced than they really are? To make it looks as though Iranian nuclear scientists are inept? To put it another way – was this plot propaganda against Iran, disinformation put out by Iran, or something else by a person (or nation) with a political agenda? Not sure that we can pin that one down now, but it would be nice to know.
Whatever happened – and whoever was responsible – this gives me pause. We have made similar mistakes in the past – letting our tendency to see the patterns we expect, swallowing information that fits these biases, and drawing conclusions about a nation we really don’t like much – and made some mistakes. If this plot is authentic and there’s a simple mistake in the plotting then we have to worry about Iran; if the plot is fabricated to help nudge us along a path to war then we have to worry about whoever is doing the nudging, and about our seeming inability to learn from past experience. But for me, seeing this plot makes me wonder if maybe Iran isn’t running into some snags – are they trying to convince us that they’re farther along than they really are, or is some other nation so disappointed in their ability to find a smoking gun that they’re manufacturing “evidence” of wrong-doing? Either way, it makes me more inclined to hold fire until we can find something more definitive.