One of my favorite contemporary philosophers, Nick Bostrom, once wrote the captivating paper *Why I Hope the Search for Extraterrestrial Life Finds Nothing*. After reading both Max Tegmark's *Our Mathematical Universe* and Paul Davies *The Eerie Silence*, I have joined Bostrom's seemingly contrarian camp.

To assign credit where credit is due, however, this view is not Bostrom's idea. Rather, it originates from Robin Hanson's *Great Filter*, which is a possible resolution to Enrico Fermi's eponymous paradox: if life is so easy, then where are all the aliens?

According to the GF, even if abiogenesis is easy, then there is plausibly some extremely unlikely event between abiogenesis to the "techogenesis" of a species as, or far more, technically literate as our own. Indeed, on Earth a lot had to happen before the most primitive prokaryotes evolved enough to send a descendent to the moon. Perhaps one of those events, such as the evolution of multicellular eukaryotes, is the impossibly lucky roll our great ancestors had. Of course, who's to say there aren't many GFs?

It is self-evident, I think, that just because the descendents of some extremely primitive prokaryotes sent a man to the moon, it does not imply that a later descendent of us humans will be sent to another galaxy. We know this because we humans are far from this, and we don't seem to be much closer as the years go by. But of course I hope, one day, we will be there.

This hope of an awesome, space faring civilization is at the center of Bostrom's argument. However, because we're not there yet, it raises the question of whether we ever will be. Is there another GF between where we humans are now and the time when humans could be a space faring civilization, where we colonize planets like in *The Expanse*? What would such a GF look like? If you think about it, a GF is really just a euphemism for an existential risk. In this vein, Bostrom writes:

Perhaps the most likely type of existential risks that could constitute a Great Filter are those that arise from technological discovery. It is not farfetched to suppose that there might be some possible technology which is such that (a) virtually all sufficiently advanced civilizations eventually discover it and (b) its discovery leads almost universally to existential disaster.

Where, then, is the GF? We hope it's behind us, so that laid before us is a more or less existentially risk free future where the path to a space-based future is not improbable. But how can we reason about this? No one knows for sure, but we can put on our Bayesian hats and see what comes out.

Let \[P(\text{GF is at or before }t)\] be the probability that the GF is at or before some "time" \[t\], where \[t\] is simply a placeholder for some epoch making event that is plausibly necessary to become space faring, such as the transition from single- to multicellular organisms, or the harnessing of nuclear energy (or maybe, even, discovering to effectively control AI).

How can we estimate \[P(\text{GF at or before }t)\] given the observations of life that we can do today. Well, absent any radio signals from intelligent species, let alone in-person "hellos" from already space faring civilizations, the best we can do is look at ourselves, as well as the planets and moons and other natural satellites in our solar system. Following Bostrom, let's take Mars as our case study.

Suppose we think the GF is near some time \[t = t'\], so that the probability distribution \[P\] is somewhat peaked around \[t'\]. And now consider what happens if on Mars we were to discover a lifeform, independent of all life on Earth, at the stage \[t'\]. What happens to the distribution \[P\]? Well, simple Bayesian updating tells us that \[P(\text{GF at or before }t') \mid \text{Mars data})\] equals \{\frac{P(\text{Mars data} \mid \text{GF at or before }t')P(\text{GF at or before }t')}{P(\text{Mars data)}}.\} But because we thought the GF was at \[t'\], then, by definition, \[P(\text{Mars data} \mid \text{GF at or before }t')\] is an *extremely* small number, much smaller than the other probabilities in this equation. The net effect, then, is to lower our credence that the GF is actually at or before time \[t'\], and must instead be later.

So you see, the more advanced species we discover independent of ourselves, the more the GF is shifted to the future. The best way to keep the GF in our past, then, is for all searches for extraterrestrial life to turn up empty handed.

To give the final say to Bostrom:

So this is why I’m hoping that our space probes will discover dead rocks and lifeless sands on Mars, on Jupiter’s moon Europa, and everywhere else our astronomers look. It would keep alive the hope for a great future for humanity.