Michelle Pfeiffer, Supervolcanoes and the Yellowstone Fallacy
I recently watched the Discovery Channel’s Supervolcano, a docu-drama about what would happen if the massive caldera underneath Yellowstone National Park were to erupt.
The program had been sitting on my TiVo for a while, because it’s hard for me to commit to an hour of Alias, much less a three-hour made-for-cable movie. But I knew I’d eventually watch it, because from the moment I first heard about the Yellowstone supervolcano, it was one of those nagging, back-of-my-mind fears. So much so, that I actually included a lengthy monologue about it for a script I was writing. (That scene got cut, so feel free to write your own.)
For those who don’t know, Yellowstone National Park, home of the Old Faithful geyser, is actually the caldera of a massive volcano. And not just a “theoretical” volcano: it’s erupted at least three times before: 2.1 million years ago, 1.3 million years ago and 640,000 years ago. Which, if you do the quick math, suggests that it erupts every 800,000 to 660,000 years. Which means it’s due to erupt now.
Which is very, very bad.
When Yellowstone erupts, it will be big-summer-disaster-movie apocalyptic. Think Armageddon x The Day After Tomorrow. Twenty feet of jagged volcanic ash strewn across the Midwest, tapering down to a centimeter on the East Coast. Global temperatures will fall. The monsoon will fail. Drought, famine, starvation lasting for years. As Discovery says on the website:
A modern full-force Yellowstone eruption could kill millions, directly and indirectly, and would make every volcano in recorded human history look minor by comparison.
Suddenly, moving to Australia looks a lot more enticing. Yes, there’s the global famine, but at least you don’t have ash falling on your head.
But here’s the thing: Yellowstone is not actually “due” to erupt. That’s a logical fallacy. And the celebrity spokesperson who proved it to me is Michelle Pfeiffer.
Let me provide context.
In 1994, my friend Elizabeth and I went to see The Madness of King George at a movie theatre on the Third Street Promenade in Santa Monica. The movie theatre was pretty full, so we ended up sitting next to a man and woman — who turned out to be Michelle Pfeiffer and her husband David E. Kelly. Being good Los Angelenos, we pretended we didn’t know they were beautiful and famous. We just ate our popcorn, watched the movie, and gossiped after they left.
The following Saturday, Elizabeth and I went to a different movie theatre and ended up sitting directly behind the Pfeiffer-Kellys. Now, at this point, one could argue we were stalking them. But it wasn’t intentional. It was just a coincidence that we all had the same taste in movies. Still, I felt guilty on some level. I considered saying something, but every scenario I played out in my head ended up with Michelle Pfeiffer looking disgusted, and David E. Kelly getting pushy. And he’s taller than me. So instead, Elizabeth and I just ate our popcorn, watched the movie, and gossiped after they left.
Then a week or two later, Michelle Pfeiffer drove past me on Montana Blvd. We didn’t have a big emotional moment, or bond over a shared reaction to a film, but it happened. Another Michelle Pfeiffer sighting.
I had seen her three times in the course of a month. When I’d tell friends about it, they’d rightly acknowledge it was quite a coincidence. But that’s all it was — a coincidence. Had I seriously said to my friends…
“I’m due to see Michelle Pfeiffer in the next week to ten days,”
…they would have slowly backed away. The fallacy is obvious: three sightings isn’t a pattern, and the two intervals between those sightings is even less of one. The most precise guesstimate a reasonable person could make is that I would see Michelle Pfeiffer again sometime between Eventually and Never. (It turned out to be Never.)
Michelle Pfeiffer is the supervolcano.
All of this is a very name-dropping way of explaining why I shouldn’t lose any sleep over Yellowstone erupting. You can’t say something is “due” just because a similar interval of time elapsed in the past.
That’s really obvious when the subject is a beautiful actress, but harder to grasp when you’re dealing with cataclysmic events. I think one difference is scale: When you’re dealing with millions of years, your brain gets overloaded. You lose touch with common sense, and start thinking purely by numbers. But the more important factor is stakes. If you never bump into Michelle Pfeiffer again, your life won’t be significantly worse. However, if the Yellowstone supervolcano erupts, you die. Or at least suffer horribly.
All of this is not to say we should ignore Yellowstone. Scientists might one day tell us that the area is due to erupt based on studies of magma, crust density, earthquake activity and their correlation with historical patterns. But they’d be looking at the real causes, not guessing based on a two-sentence summary of the situation.
Which leads to the final mystery: Why did I see Michelle Pfeiffer three times in a month, but never since? Well, she used to live in Brentwood, near those movie theatres. But in 2004, she and her husband moved to Northern California.
For the record, closer to the Yellowstone supervolcano.
Articles
May 27th, 2005 at 3:32 pm
Very entertaining John! Might I also add that even at the shorter end of the scale of gaps (660,000 years), that would still leave us 20,000 years shy of its “due date.” Of course, one could counter-argue that since the previous gap was 140,000 years shorter than the one before it, the gaps may be growing ever shorter and we’re actually overdue for the cataclysm. Me, I’ll just go on living and enjoying until that “act of God” ends it for me!
May 27th, 2005 at 3:45 pm
As an FYI for all the name-lovers out there, the supervolcano below Yellowstone is called Bruneau-Jarbridge. Ever since I heard it, I’ve always loved that name.
May 27th, 2005 at 3:48 pm
At least they’re not calling it Jar Jar Binks. That would be a total disaster.
May 27th, 2005 at 4:22 pm
Just in case you change your mind, they’re making a lot of movies down here in New Orleans. You could beat the rush.
And tell Michelle I said hi tomorrow.
May 27th, 2005 at 5:15 pm
Hey, what’s wrong with “Alias”?!
May 27th, 2005 at 5:16 pm
But your Michelle Pfeiffer sightings were truly random (assuming you two live in the same town and both have similar tastes in movies). (similar taste v. similar tastes? off to the grammar shelf for me!) The period between volcano eruptions has to do with the amount of pressure that builds up below the volcano over time. Although the last blast may have caused changes below ground, for most volcanos there does appear to be a fairly regular (if 250,000 years is regular) cycle. It’s like snow. Sometimes there is snow before Thanksgiving, sometimes not until after Christmas. But snow does come about once a year.
Was terrified of volcanos as a kid. Does anyone remember the children’s book about a family with a tiny volcano in the basement? Freaked me out.
May 27th, 2005 at 6:03 pm
Mary:
What’s interesting about the Yellowstone situation — and probably other volcanoes as well — is that it’s not simply cyclical. Scientists can’t say that, oh, the pressure builds up by X amount each year, until it blows. There are all sorts of possible outcomes, including the possibility that it may never erupt again.
Because of the way plates are moving, the hot spot might end up underneath a mountain range, where it couldn’t do the same things.
May 27th, 2005 at 6:06 pm
By the way, if anyone actually knows the proper name for “trying to form a pattern out of too little data,” I’d love to hear it. The closest I could Google was the Fallacy of Hasty Generalization, but I’m pretty sure that’s not right.
May 27th, 2005 at 8:15 pm
I think it’s called “guess-timating”.
May 27th, 2005 at 9:52 pm
I didn’t see VOLCANO or DANTE’S PEAK. But if you write this one and include another great role for Timothy Olyphant, I’ll be first in line!
May 27th, 2005 at 9:58 pm
I learned about his last year from reading, okay listening, to an audiobook called, “A Short History Of Nearly Everything” by Bill Bryson. An incredibly entertaining look at science in the past, present and future for people that aren’t scientists. If you get the audiobook, make sure you get the 6 hour abridged version because Bryson himself reads it and he’s got a great sense of humor. The 20 hour version is read by a really dull dude.
May 27th, 2005 at 10:14 pm
I love the Discover Channel because it feeds into my intense fear that the world is on its way to it’s final curtain. Now maybe it’s because I’m a bisexual puerto rican catholic screenwriting playwright in the east coast with a deep intense need to have the “bejesus” scared out of me twenty four hours a day. “I LIVE IN NEW YORK.” But the Discovery Channel seems to be where I get my fix of terror these days. In the last few months alone, they have covered; floods, tsunamis, fires, earthquakes, but it is this SUPERVOLCANO that sent me over the edge. My agent says that I have to move west eventually. But I am convinced that on the day I get on that plane, The SuperVolcano will erupt, creating a chain reaction of earthquakes along the California coast, that will lead the plane to crash in a mysterious island with a french lullaby singing woman who wants to steal my baby. I am staying PUT. As far away from that Supervolcano and Michelle Pfieffer as possible.
May 27th, 2005 at 10:27 pm
John wrote: “By the way, if anyone actually knows the proper name for “trying to form a pattern out of too little data,â€? I’d love to hear it. The closest I could Google was the Fallacy of Hasty Generalization, but I’m pretty sure that’s not right.”
Maybe the gambler’s fallacy would be closest (it hasn’t happened yet, so therefore it’s “due”).
That’s not exactly right either, but still…
May 27th, 2005 at 11:24 pm
Cascadia fault has the collapsing pattern (curve bell and all) that could cause a real giant quake inflicting as much damages as predictable by SuperVolcano theory or any other geophysics modeling created from scientific evidence.
Magma moves and all that iron 125 kms down spins orbital magnetic fields. Knowing how gravity can spray up comet material when it nears solar pull, one can always speculate these forces can shift a shearing gap to continental crusts planetwide.
Monitoring odds can only warn in advance. Infinitesimal data lacks precision, in fact.
Now, had ourselves a rogue comet slamming into Jupiter… barely made a scratch?
Earth trajectory itself has velocity and thus, molds a reactive “double-coat” magnet. And then, volcanoes erupt. Lava is attracted upwards – by the local star bending space (and momentum of nearby planets… even far away pluton is prisoner of that drag) which in turn is caught on a self-spinning spiral galaxy.
Plus, the closer Moon!
May 28th, 2005 at 5:51 am
According to my research, I am due to bump into Michelle in 140,000 years from last saturday.
May 28th, 2005 at 7:01 am
Matt:
Nothing’s wrong with Alias. I guess. I hit two bad (read: dull) episodes of the season pretty early on, so they’ve been stacking up on the TiVo. I’m at least four episodes behind the finale.
May 28th, 2005 at 7:57 am
I was wondering what Old Faithful was up to (David Kelley, not the volcano)
May 28th, 2005 at 8:30 am
Don’t move to Oz until you have checked out the Lake Toba supervolcano. It’s approximately 5 times the size of Yellowstone. It erupts on a 74 000 year cycle. Last eruption was 74 000 years ago. Did I mention it’s a subducion volcano and sensitive to major earthquakes?
May 28th, 2005 at 10:11 am
John, Besides Alias and Supervolcano what other T.V. shows do you watch or try to watch?
Cheers— Susan
May 29th, 2005 at 5:10 am
Would the tsunami cause enough shift to make Lake Toba blow?
May 29th, 2005 at 12:33 pm
well, i don’t know the “proper name for it”, but i believe david hume was the first one to describe it in detail. (the wrong idea that because something happened several times, it has to happen again.) of course, his conclusions were quite drastic: he basically questioned causality as an objective phenomenon.
new-art.blogspot.com
May 29th, 2005 at 8:03 pm
August:
Behold my Googling powers.
The key to the hit was remember the classic “Hot Hand” studies from my college cogpsych class. “Hot Hand Phenomenon” wasn’t really good enough, but I like what this guy calls it.
May 30th, 2005 at 12:15 am
I think we’re looking for the West Wing episode title Post Hoc Ergo Propter Hoc. And Lake Toba, like other subduciton volcanoes, is more likely to erupt after large earthquakes. It sits on the fault line that generated the Aceh and Nias earhquakes.
May 30th, 2005 at 8:37 am
Meteor impacts, volcanoes, climate change, biological catastophe. It seems more likely that the human race isn’t meant to survive forever. Could have an interesting impact on entertainment choices, until it’s over.
May 31st, 2005 at 9:49 am
Alan, I think post hoc ergo propter hoc has to do with bad assumptions about causality. For instance, if I light a match and then a helicopter goes flying by, post hoc ergo propter hoc thinking would lead me to believe that lighting matches attracts helicopters.
May 31st, 2005 at 9:51 am
Here’s a better link on the Clustering Illusion.
May 31st, 2005 at 10:09 am
Good link, Craig. Particularly, “Belief in the Law of Small Numbers.”
To me, there’s still sort of a difference between the “true” clustering illusion, which tries to tease a pattern out of a set of random occurences (“XOXXOXOXXXXO”) and the Yellowstone Fallacy (“XOXO..?”), which is predicting the future based on too little information.
But it’s certainly the closest match I’ve seen.
June 3rd, 2005 at 4:34 am
Vermont Troll weighs in: Very clever guys and gals. Is ClusterF…. the phrase we seek?
June 3rd, 2005 at 7:10 am
Along the same line, somewhere in the nineties, during the same week I almost ran over Tim Burton on the Hollywood-Warner’s lot where he was making Edward Scissorhands and almost impaled a very pregnant Michelle Pfiffer while shopping for a shower curtain rod at Liz’s Hardware on LaBrea…needless to say, for the rest of the week I hid in my closet.
June 6th, 2005 at 8:11 pm
I stumbled upon John’s site through one of my writer zines and I must say, this has been one of the most amusing series of posts I’ve read in a while. I didn’t get a thing written tonight, but I was entertained here anyway. I will have to come back more often.
June 13th, 2005 at 12:02 pm
A problem with comparing periodic Michelle Pfeiffer sightings and periodic volcanic activity, is that a supervolcano doesn’t drive around in a sporty car with tinted windows. It sort of stays put.
Regardless of the number of samples, trying to figure out the next sample outside the data range is “extrapolation.” Doing so inside a range of incomplete data is called “interpolation.”
As for what its called when it is obviously done incorrectly because of low sampling, that would depend on what you are doing. It could be called an “aliasing artifact,” “government policy,” or maybe the above suggest cluster-f is better.
Ted
June 20th, 2005 at 12:43 am
I would call it “Forecasting using a statistically invalid sample”, but I am not popular at parties.
July 29th, 2005 at 9:47 am
Here’s an extra piece of information. It appears that ‘Yellowstone’ is a hotspot (perhaps on a different scale) similiar to the Hawaian islands. They’ve found record of a chain of 7 calderas from the Nevada/Idaho border to the current Yellowstone. It appears that as N. American continental plate moves the hot spot burns through, makes a caldera (that may or may not be on a regular cycle) and then the continent drifts off of the area and a new caldera has to be burned through. The delay between each new cauldera (between when the old one last erupts and the new one starts appears to be millions of years. Old Supervolcanoes
September 20th, 2005 at 2:28 am
I checked your URL about vulcanoes and i really enjoyed the story about yellowstone! It is useful for those who really are interested in this stuff.
January 6th, 2006 at 9:02 am
We are all going to Die…!!!!! 2009
April 19th, 2006 at 11:38 pm
So, did ya ?
August 17th, 2006 at 5:10 pm
Do you think Daniel Ash is susceptible? I’ll write the post-fallout screenplay in NY while all you mofos dance in a burning hot foam party– ahhhhh!
May 22nd, 2007 at 2:59 pm
Take your pick- http://en.wikipedia.org/wiki/Listofcognitive_biases
November 6th, 2007 at 6:52 pm
hahahaha im still waiting and still nothing hapened
November 24th, 2007 at 11:17 am
Seeing Michelle would be better than winning the lottery. I would be eternally grateful