The Great Filter, Part I

Go watch this video

Its only 15 minutes long. Or don’t, but the rest of this post will be a commentary on the video, so there wouldn’t be much point to reading this.

Robin Hanson argues that there is a “great filter” something which is killing everything in the universe. The argument goes that, because we don’t see any aliens, we can be reasonably sure that there aren’t any, and therefore we have to figure out why.

A couple of notes on the Fermi Paradox:

It’s probably true; we can be reasonably sure that, if life anywhere got to the point of distributing itself across the galaxy, it would very quickly get everywhere. Basically, it’s exponential growth; when civilizations get to the point that they can colonize star systems, even if each star system can colonize another once every 100,000 years, that means that galactic population doubles every 100,000 years, which means it goes to one star system to every star system in the entire galaxy within 3.6 million years (log 2 of 100,000,000,000 * 100,000), or about .03% of the lifetime of the galaxy. So within galactic time, if any civilization is willing and able to colonize the galaxy, it will do so within the blink of the galactic eye. That means we can be pretty sure that there haven’t been any such civilizations within our galaxy yet.

What it doesn’t mean is that there aren’t one off civilizations on various planets. Yes, a single planet could be populated by a civilization which doesn’t want to colonize the galaxy.  There could be dozens of these, perhaps hundreds.  But not thousands and certainly not millions.  If intelligent civilization is common in our galaxy, then either galactic colonization must be impossible, or the galaxy most already be colonized.

If we assume that a Galactic civilization would, in some manner effect the galaxy itself (a common candidate for this is the Dyson sphere*, which would encompass an entire star system for the purpose of extracting energy) we would theoretically be able to notice it in other galaxies (in the case of Dyson spheres, we would notice bodies which emit no visible light yet emit large amounts of infra red radiation); which we do not notice. So it doesn’t appear that there have been colonizers at any point in the near universe, (which Robin Hanson estimates covers 10^18 planets).

There are couple of other explanations for the Fermi Paradox.  Most likely that there are civilizations our there, but we don’t see them, because they use something better than radio to communicate, and don’t bother building Dyson spheres because a single zero point energy source has twice the power of a star, or they all quickly migrate to subspace, because nobody would ever want to live in boring space if they can help it.  Or maybe there is a sort of prime directive, and that advanced civilizations aren’t allowed to contact unadvanced ones.  These are all possible, but not certain.  We still must give some possibility to the theory that the universe is just as it looks: dead.

So what prevents life from populating the universe is called the great filter. It can be something in the past, such as the development of single celled life (if this is really “hard” then maybe about of a billion planets we would only expect three or four to develop single celled organisms, of which we are one). They could also be in the future, for instance maybe we will destroy the environment, or maybe a giant pandemic will destroy humanity.

In order for something to be the great filter, it must posses these three traits:

1: It must prevent galactic colonization
2: It must be stable
3: It must be universal (or near universal)

Item one is self explanatory. Item two simply means that in order for a thing to prevent colonization, it has to be long lasting; at least in terms of galactic time (even setbacks of thousands of years don’t count for a galaxy that is ten billion years old).  Finally, it must be universal. That is, it has to be something that effects every candidate for colonization (or in the case of multiple filters, each filter must affect enough civilizations that the combination of filters affects all civilizations).

Robin Hanson proposed a list of possible filter candidates for the future:

Robot Rebellion
Totalitarian world
no starships?
lose desire

Lets examine them

Asteroid and  Supernovae: Hanson dismisses these; we can see that they are not universal. That is, we are reasonably knowledgeable about how often Asteroids and Supernovae happen, and they aren’t common enough to be universal. There may have been individual civilizations destroyed by either of these things, but we wouldn’t expect every civilization to be destroyed by them. Earthquakes and supervolcanoes can also be dismissed for the same reason (or can be dismissed as a future filter.  Perhaps supervolcanoes are common enough in the typical planets that they prevent civilization from starting on almost every planet.  But we can be reasonably sure that the Yellowstone supervolcano won’t go off for tens or hundreds of thousands of years.  Unless we think we are 100,000 years away from colonizing the galaxy, then we don’t have to worry about the supervolcano stopping us).

Robot rebellion is another filter that Hanson rejects, Robots may destroy us but then they would colonize the galaxy (and would probably do a better job at it!).

Totalitarian world is an interesting one, but for various reasons I think we can reject it.  Maybe a despotic government will rule the world, blocking all progress. This actually fails on all three levels. Looking at 20th century dictatorships, we can say two things. First, that they aren’t very stable. Of the totalitarian governments, most of them no longer exist. Totalitarian governments in Germany, Russia, Italy and Spain have been dissolved one way or another. We can maybe put an expected lifespan of 150 years for a totalitarian government, not exactly very stable. Even if I’m grossly underestimating the stability, it doesn’t really matter, these governments would have to existing for hundred of thousands of years to be a filter.  Thank God that Nazi Germany lost, but even if they won and established a thousand year Reich, it would have set humanity back, well, one thousand years.  Again, a horrible thing from a human perspective, but in galactic time it would be inconsequential.

Secondly, based on our history, totalitarian governments aren’t exactly bad at space exploration. Nazi Germany made great advances in rocketry, the Soviet Union had Sputnik and Mir, China has a not unimpressive space program. By all accounts North Korea is a living nightmare, but they still have if not a space program, then a rocketry one. One may argue that these governments aren’t as good as the free world at space exploration (and I think the evidence bears this out), but even if we assume that a totalitarian government would progress only 10% as fast as a free one then we would expect to colonize the galaxy, maybe instead of 500 years from now then 5,000 years from now.

Finally, in order for Totalitarian government to be the filter it must be universal.  Such a government might occur on earth, but for it to be the great filter, it has to occur for all civilizations; regardless of the underlying biology (or culture or history) of the species.  Basically, if you believe that totalitarian government is a real filter; then you’d have to basically believe that there is a form of government such that it will occur in every intelligent species, once established it will last for the duration of habitable epoch of the planet on which it occurs, (in our case, that’s in the hundreds of millions of years), and which absolutely prevents space colonization.

For berserkers, we can also think that there is reason to doubt. There are, from what I can tell, two subsets of this.

The first idea is that nobody wants to begin colonizing the galaxy because if they did they will be “found out” and then destroyed by all the other civilizations. I find this very hard to believe, that every civilization is in fear of every other civilization. If the fears are grounded, then colonizing space would lead to their destruction. But whoever destroyed them, well, they would be revealing themselves; leading to their destruction. The cycle repeats itself until every civilization but one has been destroyed, and the one left has effectively colonized the galaxy. If the fears aren’t grounded, then it only takes one civilization trying otherwise to break the who system.  To me, it looks like a very fragile equilibrium.

The second berserker hypothesis is that there is a single, dominant civilization that destroys all other civilizations who approach space colonization. There are basically two problems with this. The first is that we should expect to see signs of the berserker civilization and we don’t. In fact, the fact that we still exist is a good indication that there aren’t any berserkers out there. The second problem is the universality of the berserker. Maybe the berserkers have effectively stopped space colonization in this and in neighboring galaxies. But when we look at galaxies farther away, we should expect to see the end of the berserker influence.  If we look a billion light years in one direction, and a billion light years in another, then we should expect, even if the berserkers have been going at it for 4 billions years and expand at half the speed of light (both generous estimates, in my opinion), to see the entirety of the berserker influence. There’s a good falsifiable test here. That is, after building a better telescope, if we see signs of life in galaxies farther away (perhaps looking 2 billion years back), then we can expect that there may be something preventing life in our neighborhood, and that can be berserkers.  But for now, since the dead spots in the universe don’t seem to be local, we can reject (or at least reduce the possibility) of berserkers.

Finally, that leaves us with the following:

War/Pandemic/Environmental destruction
No Spaceships
Lose Desire

I will talk about the No Spaceships and the Lose Desire in two other posts, but for now I will talk about war/pandemic/environmental destruction.

For pandemic, there are two types. The first is naturally occurring pandemic, the second is human designed pandemic. We can reject naturally occurring pandemic due to universality. Yes, we might be destroyed by a virus, but we wouldn’t expect every civilization to randomly be destroyed by viruses, any more than we expect the human race to become extinct by everyone having a heart attack at the same time. The second is more worrisome, that we will design a supervirus (or superbacteria or superfungus) that will destroy us. This would be stable and prevent colonization – but is it universal?

Let’s change focus briefly to talk about war. Well, we’ve had plenty of wars in human history, but I guess the kind of war we’re talking about here is nuclear war. Nuclear war would certainly stop space colonization so it would be effective, but would it be universal?

Well, the best way to determine how likely things are is to look at how often they happen (this may seem like it is obvious but it isn’t. I will have to write something about it at some point). But in this case, it’s hard, because we don’t have multiple examples of civilizations; and the one example we do have (us) it is kind of necessary that we can only have observations before it happens.

However, there is a way we can look at this. Lets assume that we are in the xth percentile of luckiness. Then, see how long we’ve been lucky for, and we can calculate the yearly odds of destroying ourselves.

Lets give an example; say we’re in the luckiest 1% of the galactic population. Therefore, we can assume that at most, 99% of the galactic population has destroyed itself by this point in its history. Its been 62 years since the Soviet Union developed hydrogen bombs (I’m assuming that we only have world destroying possibility if two parties have the bomb). If we are in the luckiest one percent, then we can do some math, and assume that at most, there is a 7.16% chance the average civilization will destroy itself via nuclear weapons in any given year. (the math for this is (1-x)^y = z, where y is the number of years since self destruction became possible, z is the percentile of luck that we are, and x is the chance that we will destroy ourselves in any given year.) If we assume there is a 7.16% chance any given civilization will destroy itself in a given earth year after possessing the bomb, then we can expect that if there were 10,000 civilizations, the longest one would last 124 years after building the bomb.  This might be  a little too short to colonize the galaxy. Anyway, instead of writing more, let me add a chart:

How long the last civilization will last based on…
How Lucky We are (percentile) Worst case Chances of Self Destruction per Year 10,000 civs 100,000 civs 1,000,000 civs 10,000,000 civs
1% 7.16% 124.00 155.00 186.00 217.00
5% 4.72% 190.62 238.27 285.93 333.58
10% 3.65% 248.00 310.00 372.00 434.00
25% 2.21% 411.92 514.90 617.88 720.86
50% 1.11% 823.84 1,029.80 1,235.76 1,441.72
90% 0.17% 5,419.88 6,774.85 8,129.82 9,484.79

If we assume we’re in the 50th percentile (which might be the most reasonable assumption), we can see that even if there are only 10,000 civilizations in the galaxy, we can expect at least one of them to last about 824 years after the invention of the bomb, which might be enough time to begin to colonize the galaxy.  If we assume the galaxy produced 10 million civilizations (about one intelligent civilization per ten thousand star systems), we see that we can expect one civilization to last 1,442 years.  Also, since we’re so close to the invention of the H bomb, these numbers will become out of date quickly. If you’re reading this in July of 2015 (assuming that you’re not reading it from a fallout shelter), you can replace the 824 with 830 in this paragraph.  If we’re still around in 20 years, that number will be 1,089.

Finally, let’s think about the universality of this: this analysis assumes that we all civilizations are like us. In reality, if there are many civilizations out there, we can assume a diversity of characteristics. We may be in the luckiest 1%, but I find it very hard to believe we are in the most peaceful 1% of intelligent species in the galaxy. Also, think about how much less likely nuclear war would be if we had a single world government (or single superpower, as opposed to the four or five we have now). A single world government would have little reason to ever launch nuclear weapons or do anything else to end the species. If we assume that there are other civilizations out there which are, either significantly more peaceful or significantly more likely to have a world government than we are, then we can revise those numbers above upward. If anything, we are in the worse case scenario; we had the invention of nuclear weapons coincide with the development of a major ideological divide (communism vs democracy/capitalism), so there perhaps reasons to think ourselves further down on the chart in terms of luck (paradoxically, this is a case where we want to be less lucky; if our current success in not killing ourselves depends less on luck, then it depends more on “skill”, which means we are likely to last much longer).

I think we can assume the same for the intentional pandemic I mentioned above. The motives for pandemic are similar, but if anything the execution is more difficult. Its very easy to understand how a nuclear winter could literally kill everyone, but with a pandemic, all you have to have is .01% of the population survive, and within a few thousand years we’ll be right back where we are. If it only kill 95% of the population, we might be talking about only a few hundred years to get back to where we are (if that, as our technology wouldn’t necessarily be destroyed).

Finally, on the topic of war, lets talk about stability of destruction. I don’t know exactly how destructive a nuclear war would be (although I don’t want to find out). If it blasts us back to the stone age, then it only sets us back a few thousand years, which isn’t much in galactic time.   We could potentially see 10,000 civilizations from a single planet alone (although that would assume that the negative effects of nuclear wars wouldn’t begin to accumulate, or that the effects wouldn’t retard civilization development). If destroys all big animals, leaving only rats and pigeons and cockroaches, then it very well may be that there will be another intelligent civilization on earth within a few million years (lets say 50 million years). If we have another 400 million years left of habitable earth, then we have 8 more chances just from one planet. Even if nuclear war destroys everything but bacteria on this planet, it is still conceivable that Earth would eventually get another chance at colonizing the stars.

There is one last category, which is environmental damage. This is a very hard one to speak about, because, even if we assume that global warming is real, it’s difficult to really tell how harmful it will be. It may cause the ice caps to melt and for low lying populations to be displaces and for famine to break out, it may kill billions. This would be very bad, but it wouldn’t be a filter. It would literally have to destroy the human race to really be a filter. Also, we can look at the same logic we talked about earlier. We are at some level a responsible species, after all we were threatened by the destruction of the ozone layer, but we pretty much solved that through the Montreal Protocol.  Again, I find it very hard to believe that there have been thousands or millions of civilizations in the Galaxy, and we just happened to be the most responsible one.

Now, there can always be something that I’m missing, and I will discuss the other filters at a later point, but for now I’m ready to at least temporarily dismiss the above as causes of the great filter.
*I once met Freeman Dyson’s Granddaughter and Son-in-law.

Leave a comment

1 Comment

  1. The great filter Part 2 | a gallant chrome tiger

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: