Are we the only civilization-building intelligent species that has ever occurred in the universe?

It's one of science's oldest questions. Earlier this year, my colleague Woody Sullivan and I published a paper in the journal *Astrobiology* presenting new results that, I believe, throw new light on the ancient question. And, based on that work, last month I wrote an OpEd in *The New York Times* that ran with provocative title "Yes, There Were Aliens." The *Times* piece found a large audience and generated strong responses running from agreement to dissent to folks telling me I really should look into UFOs (sorry, not my thing).

Today, I would like, once again, to present our argument and dive a little deeper into its meaning and its limits. In particular, I want to address two excellent rebuttals written by Ross Andersen in *The Atlantic* and Ethan Siegel in *Forbes*. Neither Andersen or Siegel was buying some of my contentions and they both made good points. The thing about science (take note climate deniers) is that it's really a call and response. Both Andersen and Siegel are great writers. Their skepticism made me think even harder about the ideas in our paper and that was really helpful.

One note before we begin. This piece is a tad long because I need to introduce some of the background for the rest of my argument to make sense. Those familiar with the "Drake equation" and its history can skip the next section.

**The Background**

In 1961, astronomer Frank Drake was asked to convene a meeting to hash out the possibilities of interstellar communication. Drake decided to frame the question in terms of one simple one: What is the number of alien civilizations (let's call them exo-civilizations) existing now in the galaxy? To foster discussion at his meeting, Drake broke the problem up into seven pieces. Each piece represented a different aspect of the problem and each could be expressed as a factor in an equation for the total number of existing exo-civilizations (which we will refer to as N). Drake's equation looks like this:

In Drake's equation N_{*} is number of stars born each year; f_{p} is the fraction of stars that have planets; n_{p} is number of planets per star living on orbits in the right place for life to form (the so-called "Goldilocks" zone); f_{l} is the fraction of planets where life gets started; f_{i} is the fraction of life-bearing planets on which intelligence evolves and f_{t} is the fraction that go on to develop advanced technological civilizations. The final factor L is the most haunting, representing the average lifetime of a technological civilization.

The Drake equation has been hugely important for thinking about life in the universe. For the past 50 years, it's served as a critical guide for astronomers in organizing their thinking *and* their investigations of the subject.

What's important to note is that when Drake wrote his equation in 1961, only the first term, the number of stars forming per year, was even close to being known. Every other term was "data free." That meant through most of its history, scientists using the Drake equation could only provide educated guesses about the other terms. If you where *optimistic,* you argued for values that led to a large value of N. If you were *pessimistic,* you argued for values that led to tiny values of N. It was a free-for-all.

But that was before the *exo-planet revolution*. In the past 20 years, astronomical discoveries have transformed our understanding of planets orbiting other stars. In the process, they have nailed the second two terms in the Drake equation (f_{p} and n_{p}). *What we found was that there were planets everywhere*. Pretty much every star in the sky hosts at least one planet.

**The New Work**

In our paper, Woody and I realized that we could use this giant leap forward to do something that, to our knowledge, had not been done before. We used the new data to say something a little more definite about exo-civilizations.

To accomplish this, we first changed the question. We abandoned "How many exo-civilizations exist now?" and focused instead on "How many exo-civilizations have their ever been?" This approach allowed us to ignore the lifetime term L. It also allowed us to think differently about the three unknown probabilities involving life (f_{l}, f_{i}, and f_{t}). Rather that dealing with them separately, our approached focused on all three terms *together*. That means we were interested in the whole enchilada: the entire process going from the origin of life all the way up to an advanced civilization. We called our new term the "bio-technical probability," f_{bt}, and it's nothing more than the product of the usual life-centric terms in the Drake equation. In the language of math f_{bt} = f_{l} * f_{i} * f_{t}.

By looking at the problem this way — *and* using the new exo-planet data *and* rearranging things — our results provide an empirical *constraint* to a very different question than the one Drake's equation usually focuses on. Here is our question:

What would the bio-technical probability per planet have to be for us to be the only civilization that ever occurred in the entire history of the universe?

Putting in our exo-planet date, we found the answer is 10^{-22} or one in 10 billion trillion. We call this number the *"pessimism line,"* and you can think about its meaning in a bunch of ways.

First, imagine you had a big bag of Goldilocks zone planets (planets in orbits where liquid water can exist on the surface). Our results says that you'd have to go through 10 billion trillion planets and only find one with an exo-civilization for humans to be unique.

Another approach is to recognize that, until our work, no one really knew what pessimism meant. Were you a pessimist if, for example, you thought f_{bt} was one in a million or one in a billion? Before our paper, there was no way put a firm limit on which values for the life-centric terms in the Drake equation implied we were alone in the deepest sense of the word. What Woody and I found was that if nature, in its infinite wisdom, chooses a value below one in 10 billion trillion, then we're the only civilization ever. But if nature chooses a number bigger than one in 10 billion trillion then we (meaning life and intelligence and civilization) has happened before.

**The Critique**

One in 10 billion trillion is a pretty small number. My argument in *The New York Times* piece was that it's so small that the implication must be that exo-civilizations have probably happened before (possibly a lot). I considered it a kind of "argument by exhaustion."

But some folks disagreed. One of the principle objections raised to my piece was that the fact that just because 10^{-22} is small does not constitute *a proof* that exo-civilizations have existed before us. In particular Andersen took issue with this sentence: "... the degree of pessimism required to doubt the existence, at some point in time, of an advanced extraterrestrial civilization borders on the irrational."

It is here that *I have to agree* with the critique. I should not have used the word "irrational." That's because, in spite of the tiny size of the pessimism line, it's not "irrational" to doubt that we are unique in cosmic history. In fact, the only empirically valid claim Woody and I can make is as follows: We can say with certainty where the pessimism line lies (one in 10 billion trillion). In the absence of more data *it is rationally possible to construct an argument that claims nature's value for the bio-technical probability lies below 10 ^{-22}*.

Where I disagree with Andersen and Siegel, however, is how to interpret our result. First is an idea that our bio-technical probability, *f _{bt}*, somehow hides the fact that each of the life-centric terms in the Drake equation could be small on its own. The headline in

*The Atlantic*piece said "Fancy Math Can't Make Aliens Real" (though Andersen may not have had anything to do with the headline). I had to chuckle at that sentence, though, because the math I used was anything but fancy.

Even though I started out doing something more complicated, the result turns out to be embarrassingly trivial. It's just one over the number of Goldilocks zone planets in the observable universe. More to the point, we didn't establish our "pessimism line" by ignoring possibly small values of the individual life-centric terms. Instead we represented them all together.

Here is how it works.

First, let's say you think the probability of getting life to form on a Goldilocks planet is one in a million (*f _{l}* = 10

^{-6}). You might also think the probability that getting intelligence to form from life on one of those planets is one in a million (

*f*=10

_{i}^{-6}) too. Finally, you could also say there's a one in a million for one of the planets that formed life and then evolved intelligence to go on and create a technological civilization (

*f*= 10

_{t}^{-6}). That means the total bio-technical probability will be one in a million trillion (10

^{-6}* 10

^{-6}* 10

^{-6}= 10

^{-18}). There is no slight of hand here. Whatever arguments one wants to make about how improbable the formation of life

*or*the evolution of intelligence

*or*the creation of civilizations might be — they are all expressed within the bio-technical probability.

Note that the choices above, when compared with the pessimism line, lead to 10,000 exo-civilizations occurring over cosmic history.

Also, while its true that we can't say anything explicitly data-driven past our derivation of the pessismism line, the history of debate about the Drake equation provides ample material to think more deeply about our result. While many have argued that exo-civilizations would be rare, the sense of what rare means is rarely specified explicitly. If you scratch below the surface, rare often means orders of magnitude above our 10^{-22} pessimism line.

To see this point, let's take a particularly famous example. In 1983, the physicist Brandon Carter developed an absolutely ingenious argument against exo-civilizations based on the observation that the time for intelligence to arise on Earth was close the total age of the sun. Using this one fact, he further made the case that intelligence required evolution to pass through a series of "hard steps," each of which would be highly improbable.

Imagining there were 10 evolutionary "hard steps," he did a calculation where he found the total probability for exo-civilizations to form to be 10^{-20}. He then claimed, this value "is more than sufficient to ensure that our stage of development is unique in the visible universe."

But it's not! The pessimism line we derived shows that Carter's 1983 calculation still allows 100 exo-civilizations. Carter intended his calculation to be *hyper-pessimistic,* but it turns out to be optimistic instead. It should also be noted that researchers now believe only five hard steps exist (if they exist at all). This, along the other values in Carter's original paper, imply a probability of 10^{-10} which, along with our pessimism line, implies *a trillion exo-civilizations across cosmic history*. (It's also noteworthy that authors like Mario Livio present arguments that undermine the basis for Carter's work).

Of course, it's still possible to construct arguments leaving the probability far below our pessimism line, ensuring we're the only exo-civilization that ever formed. But it's here that, I believe, the most important *implication* of our result emerges.

Our probability is not an abstraction. It's not just a pure number. Instead, it represents something very real. It represents 10 billion trillion planets existing in the right place for nature *to have at it*. Each world is place where winds may blow over mountains, where mists may rise in valleys, where seas may churn and rivers may flow. (Note our solar system has two worlds in the Goldilocks zone — Earth and Mars — and both have had winds, seas and rivers). When you hold that image in your mind, you see something remarkable: *The pessimism line actually represents the 10 billion trillion times the universe has run its experiment with planets and life.*

That's why our result has implications worth exploring. For the first chunk of the 20th century, the dominant mechanism for planet formation was thought to be near collisions between stars. Now we know better, and we can empirically constrain the pessimism line. Because it turns out to be very small (or, conversely, the number Goldilocks zone planets is so large), it means *the burden falls on the hyper-pessimists*. The universe gets to run the experiment many, many times. So if you want to argue Earth is unique, then the onus is on you to show why technological intelligence is so strongly *selected against*.

And any hyper-pessimist argument will be balanced by the fact that there are many good new arguments that the emergence of life and intelligence may not be so hard to obtain. Many of these optimistic views come from advances in biology. For example Wentao Ma and collaborators use computer simulations to show that the first replicating molecules could have been short strands of RNA that were easy to form and which quickly led to a "takeover" by DNA. And, as neurobiologist Lori Marino has argued, human intelligence evolved on top of cognitive structures that already had a long history of life on Earth. Thus our kind of intelligence should no longer be seen as entirely separated from what evolved before. It's special but not that special.

Thus skeptics *are* entirely right that without any more data one *must* remain formally agnostic about exo-civilizations. You can't assign a probability to an unknown process. But to stop there misses a key point about our moment in science and in history. Astrobiology, the study of life in the universe, has made tremendous strides through studies of our world, the other worlds in our solar system and, famously, the newly discovered exo-planets. The study Woody Sullivan and I carried out is firmly situated in the midst of these expanding astrobiological horizons. Taken together, I believe our results mean that most pessimists (on the question we asked) are actually optimists and the remaining hyper-pessimists — well, they really have some 'splaining to do.

Finally note that our study said nothing about the existence of civilizations now. We were dealing with a kind of exo-civilization archeology. If that all important lifetime factor L is not long, then our neighborhood the Milky Way galaxy might be entirely empty (other than us) in the current cosmic epoch.

*Adam Frank is a co-founder of the 13.7 blog, an astrophysics professor at the University of Rochester, a book author and a self-described "evangelist of science." You can keep up with more of what Adam is thinking on *Facebook and Twitter:* *@adamfrank4.