Bryan Caplan  

Planets, Life, and the Fermi Paradox

The Real Concentration of Powe... For Students...
Since junior high, I've believed that the galaxy is full of planets and intelligent life.  But it wasn't until I was in my thirties that I heard of the Fermi Paradox.  You can boil the paradox down to a sentence: "Where is everybody?"  In other words, "If there's intelligent life anywhere else in the galaxy, why don't we see any sign of it?"

It turns out that at least the first of my junior high guesses was right on target.  The galaxy is indeed full of planets.  The New York Times:
Three studies released Wednesday, in the journal Nature and at the American Astronomical Society's conference in Austin, Texas, demonstrate an extrasolar real estate boom. One study shows that in our Milky Way, most stars have planets. And since there are a lot of stars in our galaxy -- about 100 billion -- that means a lot of planets.

"We're finding an exciting potpourri of things we didn't even think could exist," said Harvard University astronomer Lisa Kaltenegger, including planets that mirror "Star Wars" Luke Skywalker's home planet with twin suns and a mini-star system with a dwarf sun and shrunken planets.

"We're awash in planets where 17 years ago we weren't even sure there were planets" outside our solar system, said Kaltenegger, who wasn't involved in the new research.

Astronomers are finding other worlds using three different techniques and peering through telescopes in space and on the ground.

Confirmed planets outside our solar system -- called exoplanets -- now number well over 700, still-to-be-confirmed ones are in the thousands.

These discoveries seriously undermine the Fermi Paradox.  If we've only recently confirmed the existence of extrasolar planets, why on earth should we be surprised by the fact that we've failed to confirm the existence of extrasolar intelligent life Clearly our powers of detection remain extremely limited if we could overlook the existence of hundreds of billions of entire planets.  To refuse to believe that any of these planets sustain intelligent life dumbfounds me.

Several people have objected, "The Fermi Paradox doesn't ask why we haven't found extraterrestrial life.  It asks why they haven't found us."  Maybe because they face the same problem we do: Detecting intelligent life is extremely difficult.  But shouldn't they already be here?  Not if space travel (including the value of time) permanently remains extremely costly relative to the value of raw materials.  It's a lot easier to believe that space travel will forever remain a rare luxury for intelligent life than that intelligent life exists on Earth alone.

When people discuss the Fermi Paradox, a common rhetorical tactic is to say, "Whatever intelligent life usually does, surely one species of intelligent life would be the exception that proves the rule."  Facile.  When you multiply independent, rare events together, you quickly reach situations with zero examples.  There are seven billion people on earth.  But how many people named Bryan Douglas Caplan are watching Citizen Kane and eating butterfingers right now?  Almost certainly zero.  As Google often tells me, "Your search did not match any documents."

Similarly, even if there are seven billion species of intelligent life in the galaxy, there could easily be zero species that entered our solar system during the last century, approached the earth, and stayed long enough for the scientific community to detect and confirm.  Your search did not match any documents.

I don't expect science to vindicate my other junior high guess in my lifetime.  Our powers of detection are bad enough to overlook hundreds of billions of planets.  Detecting intelligent life will be vastly harder than detecting planets - maybe prohibitively harder.  But until we actually do the leg-work of examining a thousand random Earth-like planets, concluding that we're the only intelligent life in the galaxy is an astronomically implausible leap of logic.

Comments and Sharing

COMMENTS (39 to date)
Luke G. writes:

One thing I don't get about the Fermi Paradox: why are we so certain that life is so abundant? Could it be that the Solar System is a one in a zillion anomaly? Is there good reason to assume life exists outside our zone at all?

Lars P writes:

I've discussed this a number of times now, and this is the first time I've seen the idea that space travel is simply too expensive even for advanced species.

Economists *do* have different smart ways of looking at things!

But the strongest form of the Fermi paradox doesn't deal with simple travel between worlds, but with colonization. Even if if takes a million years to colonize a nearby star system, with exponential growth and assuming other species started out a few hundred million years ago - not long by galaxial age standards - the galaxy should be well covered by now.

My favorite explanation to Fermi is that There Are Rules out there, and whoever is in charge of this corner of the galaxy doesn't want us to be disturbed yet.

Carl Shulman writes:

A colonization probe that reaches another star can build up a whole new civilization that can build many new probes without further taxing the resources of the home system (and they could even pay their own way by sending back valuable information, like the short results of enormously energy-intensive computations). Natural selection will then tend to give you a galaxy filled with colonizers (even "environmentalists" would still need to send out replicating seeds to enforce their environmentalism): even one colonizing civilization per galaxy is enough to colonize it within a few million years.

Pandaemoni writes:
To refuse to believe that any of these planets sustain intelligent life dumbfounds me.

I wouldn't go that far...

It's fine to have a hunch on the matter (and I have the same hunch you do), but we should be somewhat careful, as it's no more than a hunch.

First, no one has any sense at all of what the odds are of life evolving on a planet (even assuming the planet in question is capable of supporting life). We have one example of a life bearing planet, our own, and we have a limited understanding of how it arose here. We have an hypothesis on how life arose from non-living matter, but even if we assume the truth of that (as I tend to do) we haven't worked out many of the details.

Second, there is some reason to believe that all multicellular life on Earth may have been an unusual accident. Most of the history of life on Earth featured no multicellular life, and only about 500 million years ago it seems to have suddenly arose in what may have been a single event. There is the theory that the mitochondrion (aka, the "powerhouse of the cell") was once a separate organism that somehow settled inside a larger cell and somehow embedded it's own genetic code in that of the cell so that the two merged into a single cell. The additional energy granted to the cell by the hitchhiking mitochondrion may have been what made multicellular life possible.

If that's correct, it may be unfathomably unlikely that multicellular life develops on otherwise life-sustaining worlds. (Or it may be that life elsewhere would find some other way of reaching a multicellular status...notwithstanding that that only seems to have evolved here on Earth once. Again, we just don't know.)

A related point is that, by biomass alone, single celled organisms are far more successful that multicellular ones in evolutionary it's not clear that evolution favors the big guys.

Third, it's not clear that evolution favors intelligence beyond a certain level. Human brains, at least, are massive calorie hogs, and require precious protein and regular food intake to grow properly. While there are advantages to being intelligent, we are the first animal of high intelligence to evolve, so why did nature take so long? It could be random chance that it took 500 million years after the development of the first multicellular life, but here again we have a single data point, humanity, and we really can't generalize from that in a rigorous way. Human intelligence is, right now, a mere anecdote in the history of life on Earth, and so far as we know life on Earth is an anecdote in the history of the universe.

You also have to consider that technologically sophisticated aliens would all, presumably have access to technologies that could kill them. Even if the odds of them starting their own version of a nuclear war are 1 in a million in a given year, the odds of them lasting a million years is only 37%, the odds of them making it 10 million years are nearly zero. (So one hopes the odds of an accident triggering a nuclear war are less than 1 in a million per year on average, for our sakes).

We all like to imagine what numbers would be filled into the Drake equation and see what results our guesses yield. Even though some small part of me does want to say, "I can't imagine the odds of life evolving on an Earthlike planet are less than 1 in X," I have to be intellectually honest and admit that I have no rigorous basis on which to estimate that X, it could be that X = 10, it could be that X = 10 googol.

If it were the latter, then it would be extraordinary that we exist, but that's not evidence that X = 10 googol is wrong. Vesna Vulovic holds the record for surviving the highest fall without a parachute and surviving, at 33,000 feet. If that factoid were my one and only data point, I would probably think that a 3,300 foot fall from the sky is easily survivable even though (as it turns out) that is a near certain death.

I have a hope that intelligent life is plentiful...but if I look at it objectively, the most I can say is we don't have sufficient data. That I hold onto my hope doesn't improve the quality of my limited data, in fact it likely distorts it.

Jim Glass writes:

The Fermi Paradox is based on the observation that our galaxy is not only "big in space", with countless planets that may be home to life, but also hugely deep in time, giving any intelligent species "like us" plenty of time to explore it end-to-end even at rates of speed even we can easily concieve. People quicky grasp "big in size" but are very slow to grasp the full implications of "deep in time". Hugely deep in time.

The Paradox is not a simple "where is everybody?". Fermi asked, since the galaxy is so old, if it holds so many civilizations like ours, then why haven't they been here many times over?

"Lots of planets equals lots of intelligent life" is the logic physical scientists who are impressed by very big numbers and basic probabilities. It is of course embodied in the Drake Equation.

Many evolutionary scientists have pointed out that to be a very naive approach. If natural selection selects *against* intelligence of the sort capable of developing technology as we know it (or higher) then the whole premise falls apart, and even with trillions of life-carrying planets such intelligent life may quite logically and predictably be vanishingly rare, the result of extremely long-shot odds.

The data on Earth show that adaptive evolutionary developments (eyes, herding/flocking/schooling, patterns of predation, etc. etc.) repeatedly arise over and over and over. "Intelligence"
as meant for the purpose of this discussion has arisen only once -- and in doing so escaped extinction by only the barest of margins more than once.

Many, many things select *against* big brain intelligence -- not least, the big brain itself that consumes massive amounts of fuel, limits breeding numbers, delays development, prolongs dependence, results in high rates of maternal death ... all for what benefit(?) before it much later invents a technology that gives it an advantage? The big brain must come first, and survive. The technology is only a later happy accidental byproduct of whatever process created the big brain for some other purpose. That makes intelligence seem pretty remotely contingent (lucky!) right there.

Calculus provided no evolutionary advantage to our distant ancestors. Having a brain big enough and wired complexly enough to do calculus was a very costly handicap, other things equal. That they nonetheless survived leaving us able to do calculus is a truly remarkable happenstance, not repeated once among all the billions of species that have evolved on this planet. To think that evolution naturally drives to create species capable of calculus, just like us, is ... optimistic.

I could go on about this and the other pressures potentially selecting against intelligence, but Carl Sagan and the great evolutionary biologist Ernst Mayr debated this at the Planetary Society some years back, so I will leave the discussion to them.

I'm with Mayr, at least in that it does seem very unscientific to jump to big conclusions about the likelihood of higher life in the galaxy without any consideration of the patterns of evolution, merely considering the numbers of astronomy.

Personally, IMHO, I'd guess that life of the simplest form on other planets is exceedingly common, I'd take a 50-50 bet that it will be found on Mars -- but that the "higher" technology-capable intelligent life of space operas and targeted in the SETI search is so vanishingly rare that humans will never encounter it.

But that's just a guess, of course. We shall see. Or, in our lifetimes, not. In which case the Fermi Paradox holds for us.

Essen writes:

What if we are actually colonizers from a distant galaxy who have come in a giant Noah's arc with all species packed together?
Maybe that giant ship has disembarked us here and gone off for further colonization.

Lord Kelvin writes:

So much text and so few numbers.

Bruce Cleaver writes:

@ Lord Kelvin -

Yes, I agree. Another way of saying we have insufficient data.

Here is a way I like to think of it. Each of the following probabilities has been glibly assigned as 10E-04 (1/10,000):

1) Probability of Life on a planet
2) Probability of multicellular life
3) Probability of Intelligent Multicellular life
4) Probability of Technological Intelligent Multicellular life.

The product of these is 10E-16, which is very small even for a galaxy full of planets. The point being that anyone can argue the actual figures of these (or even the constituent staements themselves), but a few small probabilities rapidly erodes the final product to nothing.

Greg writes:

you are WRONG. interplanetary star-trek type travel is perfectly possible, just like flying is possible.

what you FAIL to understand is the vastness of the galaxy. a light year is incomprehensible. put simply the chances of star trek type civilization stumbling upon the solar system (which is not central to the galaxy) is simply VERY VERY VERY small.

also consider this: how would roman empire soldiers detect a spy plane? rome was certainly advanced but they were not as advanced as us.

a star trek type of civilisation could simply fly by at the edge of the solar system, scan everything and leave. we would NOT be able to detect it. in fact we have trouble detecting huge rocks floating around the solar system which can destroy earth never mind tiny spacecraft which may be routinely cloaked or more likely made from materials which are simply incomprehensible to us.

there are millions of termite mounds on earth yet most termites have never seem humans, but humans are perfectly aware of them and simply show no interest...

fundamentalist writes:

Seems that the rational response until powers of detection improve would be simply "I don't know". What is the point of unfounded leaps of faith?

Tom writes:

The Singularity ideas of recent decades has given rise to another possibility that I've seen kicked around -- maybe advanced civilizations ultimately "digitize" themselves and then have no need for colonization schemes. They merely cluster around their home star to power their computers while the inhabitants generate whatever digitial worlds they care to live in.

Brian writes:

If there is life out there intelligent enough to contact us, and we are incapable of finding them, they're probably debating whether or not we qualify as "intelligent life".

Seth writes:

If intelligent life learned to carry signals on electromagnetic spectrum anytime in the last several hundred to several million years, then we might have a chance of listening to some of their early radio broadcasts any day now.

Finch writes:

There is a literature on this. It seems to come down to, either, intelligent life like us really is rare even at the scale we're talking about, or, something really weird is going on.

I don't think anyone who's thought about it carefully believes what Bryan seems to suggest, that intelligent life like us is common but we just haven't spotted it yet. The 1950s SETI view, if you will.

Finch writes:

I don't mean to rule out strong explanations for silent intelligent life, like Berserkers, in my second paragraph above. I mean to rule out the Star Trek explanation.

Bob Murphy writes:

Can someone explain this statement for me?

Detecting intelligent life will be vastly harder than detecting planets - maybe prohibitively harder.

jb writes:

I'm with Bob Murphy - you declare 'Detecting intelligent life will be vastly harder than detecting planets', but provide no supporting facts, or even hypotheses that this is true.

Silas Barta writes:

@Tom: Digitization of life would increase the incentives for colonization because it would become trivially inexpensive (a rounding error in terms of resources) to, once you're at that point, create a self-replicating "brain in a box" and "pitch it" to the rest of the universe.

Yancey Ward writes:

If I threw a pack of cards into the air and they landed forming a six-story card house, what could I say about the probability of it happening again? Probably not much. I think people who plug in values into the Drake equation are in the same boat I would be, and I will note that it is probably possible to give a pretty good estimate of the likelihood of my accidental card house since it is a more malleable physical calculation problem. I assert there is simply not enough information to do the Drake calculation- if even one of the proposed values turns out to be essentially a 1 in a 15 billion year event, then humans may well be the most intelligent species in this universe.

Yancey Ward writes:

And the next 20 most intelligent may reside on the same planet.

Chris T writes:

Jim Glass does an excellent job explaining the reasoning behind the Fermi paradox and why intelligent life might be vanishingly rare even with large numbers of planets.

If the galaxy has spawned numerous civilizations in its 13.2 billion year history, it is difficult to see how not a single one succeeded in the obvious survival step of spreading itself out (and a drive to survive is one trait all life will share). Far from undermining the paradox, the high frequency of planets actually makes it considerably more vexing.

roystgnr writes:

Even if creating new habitats in space doesn't produce a positive economic return to the creators, there is a form of life which has an evolutionary propensity toward making many copies of itself even at the expense of the original; this form is called "life".

If the solution to the Fermi Paradox isn't "something stops most stars from developing human level civilizations", then it's "something stops most human level civilizations from developing technology which doesn't seem too far-out in our own future". "Considerably more vexing" is an understatement. It would be the existential equivalent of the "28 Days Later" opening: awakening in a place that ought to be crowded and wondering what the hell happened to everybody else...

Hasdrubal writes:
Bob Murphy writes: Can someone explain this statement for me?

Detecting intelligent life will be vastly harder than detecting planets - maybe prohibitively harder.

Non-directional electromagnetic communications (radio, TV, etc) follows an inverse cube law (I think, though it might be inverse square) meaning we can't detect anything at what we would consider realistic power levels from more than a light year or so away. So, in order for us to detect intelligent life, they would either have to come here or send a coherent beam communication directly at us, one of 100 billion stars in the galaxy. Furthermore, it's entirely possible that their communication isn't even something we would recognize as a signal, like a stream of neutrinos using muon neutrinos as 0 and tau neutrinos as 1 the way we use high/low current for digital communications today. Not to mention that they might not be as outgoing as we are.

So detecting extrasolar life is hard because a.) they have to be intentionally trying to talk to us (among the 100 billion other possible stars) and b.) we would have to recognize that they are trying to do so.

We wouldn't be able to detect ourselves from any of the planets we've found so far.

diablo writes:

@Pandaemoni That is an interesting story about Vesna Vulović, and it illustrates your point well (that a single piece of data doesn't tell you much about a distribution)... Unfortunately it seems that this particular story is false:
"A spokesman for Guinness World Records told German paper Taz: 'It seems that at the time Guinness was duped by this swindle just like the rest of the media.'"

It seems that Ms. Vulović survived a fall from a much lower, although still impressive, 800 meters.

Jason Malloy writes:

"Facile. When you multiply independent, rare events together, you quickly reach situations with zero examples. There are seven billion people on earth. But how many people named Bryan Douglas Caplan are watching Citizen Kane and eating butterfingers right now? Almost certainly zero."

The formation of life on earth plausibly involved many independent, rare events together.

Life itself is like a man named Bryan Douglas Caplan watching Citizen Kane and eating butterfingers. Just because there's one example in 7 billion (or 100 billion), does not imply there are two.

Finch writes:

> So detecting extrasolar life is hard because
> a.) they have to be intentionally trying to
> talk to us (among the 100 billion other
> possible stars) and b.) we would have to
> recognize that they are trying to do so.

This is true only if their civilization had stopped growing and stayed stably small. Which makes you wonder why that happened to everybody. When you have a hard time explaining why it happened to everybody, you've got a poor resolution of the paradox.

Starships, for example, ought to be visible across the galaxy. But we don't see any at all. We don't see nuclear waste in stellar spectra. We don't see Dyson spheres. As you note, we don't see any sign of anyone trying to communicate. There is a conspicuous lack of evidence that we would reasonably expect to see if the galaxy had another sentient species.

Finch writes:


It's inverse square. Think of the antenna as having fixed area and sitting on the surface of a sphere with the center of the sphere at the broadcaster's location. Double the radius of the square, divide by four the fraction of power received by the antenna, assuming the broadcaster is evenly spreading his power over space.

David P writes:

There's also the possibility that there's more than one way evolution would make intellgent life and that their technology is based around something other than wires and metals.
In the discussion of life we have a tendency to look at ourselves and then modify a few characteristics but construct a relatively human creature, especially in mind. Who's to say that given proper building blocks for life (which may or may not be the same set or subset of building blocks used for life on earth) that it would behave in any way we would expect? Maybe the nearest life evolved in gas giants and is looking for other life in other gas giants. Maybe the star in their system has a lot of solar flares making radio communication not feasible. Maybe they can't even see the stars. Who knows, but without some kind of survey of the stars all we can do is speculate.

Of course, you do not believe in God, but you believe in aliens. Those who know the conditions for life dont so much agree. I am specifically talking about the book "Universes" written by someone explaining that the only reason we see life is that there is multiple "dimensions", thereby increasing the chance of it happening. I also refer to the Creationist movie "the Privileged Planet". Life does not happen by chance.

Tom McKendree writes:

I do not understand how you can have this essay, and 29 comments, all in this GMU-related corner of the universe, and not have anyone reference Robin Hanson's "The Great Filter" (click Obviously, intelligent entities, even in quantity, can behave in unexpected ways.

[comment url edited slightly for clarity--Econlib Ed.]

Evan writes:

My current "most wrong belief" is that the Fermi Paradox might be explained by the possibility that the Many Worlds Interpretation of Quantum Mechanics is mostly correct, but the part of the theory that says it is impossible to travel between the many worlds is not. Colonizing space seems pretty wasteful if you have infinite parallel versions of your world where intelligent life never evolved.

The Many worlds interpretation might also explain things if it turns out we are in a fantastically (un)lucky branch of universes where life never evolved on any other planets, but there are far, far more branches with multiple inhabited planets on them.

Roger Sweeny writes:

This puts me in mind of Stephen Baxter's wonderful--if somewhat depressing--fiction Evolution. The book is a series of interconnected stories of human ancestors and imagined human descendents back to the time of the dinosaurs and well into the future. Intelligence turns out to be a very mixed blessing that eventually is lost.

David MacRae writes:

If you want the answer to this so-called paradox, ask yourself the following question:

Why is there only one form of intelligent life on this planet?

Sounds silly doesn't it? Here's a few more:

Why is there only one form of life with a spinal cord?

Why is there only one form of multicellular life?

Why, in fact, is there only one form of life period?

The answer to all these questions is simple. One form of life got there first, which eliminated any chance for other forms to do the same. Elephants are close to intelligence. The Neanderthal may have been only few thousand years off. Too bad. We got there first.

The answer to the Fermi paradox is the same: somewhere in the far reaches on the universe there probably is intelligent life but, in this galaxy, we are almost certainly first. The time it will take us to colonise the entire thing is nothing on the timescale of the universe. Even with technology only slightly more advanced than we have now, it would only take a few million years.

To flip this around, if there were intelligent life elsewhere in the galaxy, we wouldn't be here. Same reason.

Dan Weber writes:

This concept that other civilizations will be so much more advanced than us that they don't consider us intelligent is bollocks. There are people from the most technologically developed cultures that want to talk to people from the least technologically developed cultures. There are people who devote their entire research careers towards proving (futilely, I'll add, but they still try) that we can communicate with dolphins, or gorillas, or dogs, or bees. If they are out there and know about us, someone would be trying.

Unless, as Finch said, Something Really Weird Is Going On.

Richard Hunter writes:

I'm surprised no one has cited David Brin's 1983 review on the subject in the Journal of the Royal Astronomical Society:

Though some of it is dated by new findings, much of it still holds. One of the most convincing arguments for the 'Great Silence' as he calls it is the possibility of malignant/xenophobic Von Neumann/Tipler replicating colonizers. These would operate like the normal self replicating colonizing probes many have proposed, but have the added program of finding and destroying any alien source of intelligent communications. Even at a very low rate of creation, these alone could explain the silence without recourse to novel physics. He doesn't add the obvious point that follows from this that wise civilizations would mask their communications just in case. We have only been making noise for about a century, which is only on the cusp of the time it would take for a malignant replicator to get here from the a nearby solar system traveling at say 1% of the speed of light.

Finch writes:

> If they are out there and know about us, someone
> would be trying.

Turning this around, despite the presumptiveness and inherent danger, there have been several credible attempts at active SETI, the sending of messages from Earth that realistically could be detected and understood at interstellar distances.

I.e., despite this being an obviously bad idea at our level of development and understanding of the universe, even we are doing it.

Marcus writes:


I had never actually thought of it that way but that honestly makes perfect sense.

Nick writes:

The OP and most of the commenters above are profoundly underestimating the amount of information astronomers actually have about the universe. We have spectroscopic data on not only billions of stars in our own galaxy (along with associated dust clouds and the like), but also on billions of other galaxies. They all look perfectly natural. Galaxies where most of the surfaces and illumination have been efficiently engineered would stick out on the spectrograph like sore thumbs.

Since out of billions of galaxies we've found none converted to artificial surfaces or illumination methods, something we should expect a significant fraction of ETI to have done over the many billions of available years, it's astronomically unlikely that we share our own galaxy with any other ETI.

Either that or all over the universe the aliens are all very good at hiding, like elves and dwarves.

JayMan writes:

Well this is a change in direction from the usual topics discussed here. Perhaps all good scientists begin as aspiring astronomers.

The bottom line with the Fermi Paradox is we just don't know. Every inquiry into its implications is based on extrapolating from what we know about human behavior, something that may be largely irrelevant to advanced technological species.

For example, how do we know that the idea of population growth and expansion works the same way for a technologically advanced species? What if they have greatly extended their lifespans such they have abandoned the need to reproduce, hence their population remains static? Why colonize new worlds then, or even if such colonization begins, why continue to do so indefinitely?

We can't rule out the "digitization" concept, either.

What if their understanding of physics is such that physical concepts (and physical constraints) familiar to us no longer apply to them? Perhaps advanced species have shed their physical existence altogether?

These are on top of a host of other scenarios one could imagine that are virtually impossible to adjudicate given our level of ignorance about the universe.

That said, all present research indicates that planets are plentiful in the Galaxy (indeed, it's entirely possible, if not likely, that planets of ~1 Earth mass are universal around stars in the galaxy). Planets with Earth-like conditions (liquid water, relative stability in terms of surface conditions) may also be abundant. It's possible, and I would say quite likely, that life arises whenever the conditions are right for it.

We have no way of knowing whether technological life—species capable of advanced civilization—are rare or common, and it's impossible to determine this from using Earth as an example (because N = 1). As well, we have no idea how long such civilization typically lasts once it arises (could be brief, could be indefinitely), but it is entirely possible that such civilization is abundant. We just don't know, and that's what makes it impossible to seriously draw conclusions about Fermi paradox one way or the other.

Comments for this entry have been closed
Return to top