Slashdot: Are we alone in the Universe?

posted Aug 06, 2004

Slashdot points to a CNN article that has been justly taken apart for falling prey to a biased sample fallacy; we are currently only capable of detecting planets in close solar systems such that the planets we can detect preclude the formation of planets conducive to life as we know it. (The gas giants we can detect are too close to their respective stars to even have a habitable moon, as I understand it.) Ergo, of all the solar systems we have detected planets in, 100% of them are unsuitable for life.

I wanted to talk about a comment a lot of the posters have made in the Slashdot discussion, where the posters fall into a varient of the Big Number Fallacy. It is trivial to observe that there is a lot of stuff in the universe, which is then used in an appeal to intuition... typically couched in insulting terms like so: "Most people are ignorant of how big the universe is. Only someone really parochial could think we are the only intelligent life in the universe." (Note the inclusion of "intelligent"; when we talk of "being alone" in the universe, we are not talking of being kept company by five billion planets inhabited by life that makes the simplest algae look advanced.)

But matching and exceeding the ignorance of how big the universe is ignorance of what the odds of life are. Appealing to the sheer count of stars in the universe, very conservatively a hundred billion galaxies of a hundred billion stars each, implicitly invokes the image that each star gets a fresh chance to have life.

But that's not true; we have knowledge of stars and so that affects the odds we must assign to those stars for life. Assuming that "life as we know it", broadly speaking life based on carbon and water, is all that is viable for independently-evolving life, which based on the complete lack of any other type of life or any reason to believe any other substrate can house the necessary complexity, we can eliminate huge swathes of the universe from consideration. In other words, we know enough about many stars that the odds are zero, which does little to help the "life out there" argument.

(The italicized phrase is added to forstall the inevitable objection that life need not be like our own. While it is true that Science can not conclusively prove such alternate life is impossible, there is no good scientific reason to believe in the validity of other substrates, and while it doesn't add up to "proof", a lot of reasons to believe all other proposed substrates in science and science fiction, including silicon, gas clouds, and the almost mythical "pure energy" beings living in stars or even somewhere vaguely "in space", can't support structures of significant (in a statistical sense) complexity. Thus, while referring to such life forms is not technically "invalid", you need to be aware that you have left the domain of science and have entered science fiction. There's nothing intrinsically wrong with that but one should not pretend one is making any sort of scientific argument thereby.)

Temporally speaking, the entire first and second generation of stars had insufficient complex elements like metals and such to support anything beyond vast hydrogen clouds. Spatially, there is a whole boatload of conditions that need to be met for intelligent life.

A good starting point for such a discussion is probably this page. Yes, that is from a creationist site but remember that logically, one must consider only the argument, not the source, and in this case it is the form of the argument I wish to highlight more than the specific details and numbers on that site. The form of the argument is a perfectly rigorous scientific argument. That link points to a list of various conditions thought to be necessary for life, assigns rough (and generally very conservative, at least in the case of the good points) odds to them, and then notes the total at the end.

Again, I want to emphasize I am not condoning every line of the list; as you get towards the bottom I find it highly likely that the page is really computing the odds of a planet almost exactly like Earth appearing by chance, and the odds of any given planet can be made as small as you like simply by tightening the requirements. However, the stuff towards the top, conditions for a life-bearing solar system, are quite sound. I can't find a cite online, but entire classes of galaxies are thought to be unsuitable for life, because on galatic time scales every star ends up passing through a region of space that will sterilize any planets it may have. (I have seen the idea that only spiral galaxies can contain life, but that is the most common type of galaxy, although not all spirals may be appropriate as some are dim. I can't find a layman-link for this but here's an astronomer link, see Figure 3. The Milky Way is type SBb or possibly Sbc; I am not entirely certain how SBb maps onto that figure.) Entire classes of stars can't contain life, being either too cold, too variable, or too "something else" that makes them unsuitable. Multi-star systems, which IIRC the majority of stars in the universe are involved in, are unsuitable because it becomes effectively impossible to have a stable orbit in such a system over cosmological time, and the planets are subjected to widely varying climates or are ejected from the system entirely after one pass too close to one of the component stars. Many stars are in areas that haven't had enough supernovas to make the elements above iron that we need. Many other stars are in areas that are too exciting, like the galatic core, where supernovas happen way too often for life, or pass through such areas too often. (Supernovas can sterilize a surprisingly large area; here's some descriptions of what supernovas may do to us in the future and have done in the past.)

Moreover it is certain that some or most of the lines are correct, and many lines obscure detailed scientific arguments that require some knowledge to know about, but can be quite confidently believed in after that. For instance, the line that reads "tectonic activity", rated 0.05, is based on the observation that if it wasn't for subduction the carbon balance would be off. (The page doesn't go into how the balance becomes off but if I recall correctly, the problem is that all the carbon on the planet gets deposited in the oceans as sediment over geologic times and becomes unavailable for either life or climate control as carbon dioxide.) That page also points out that it is necessary for any land, or you get a global ocean. (We pretty much know that that could support life, all other things being like Earth and assuming a magical carbon pump (maybe the biosphere gets lucky and stumbles onto a process that keeps the carbon moving, much like we got unbound oxygen floating in our atmosphere), but it is not so obvious that intelligent, tool-using life could evolve under such circumstances (no fire, and all that entails); all the intelligence on Earth seems to have evolved on the land, and the most intelligent creatures in modern oceans are dolphins and whales, both animals that migrated back into the water after evolving as mammals on the land.)

Another thing that you may not immediately realize is the importance of the Moon. Along with the stabilizing effects cited in that article, I seem to recall reading about how the tides help stir up the oceans near the edge, and tidal pools are generally thought to be where life most likely started.

It is not hard to stack up a list of odds for various independent factors needed for life, and since we have to multiply them all together to compute the odds of all of them coming in to place at once, it is not really all that difficult to build a list up that matches 1 in 1022, which is our conservative estimate of the number of stars in the universe. (I have seen a version of the list where there is a final fudge factor divided out for non-independent variables; for instance, the Moon accounts for a lot of things, including some I haven't listed and quite possibly some nobody has ever thought of, but you don't get all of them at once. On the other hand, every theory of planetary formation I am aware of makes a double-rocky planet like the Earth-Moon pair exceedingly unlikely, so the Moon still ends up accounting for a huge chunk of improbability.)

It is also worth pointing out that on cosmological scales, there is a time limit on any given planet for intelligence to develop before the star or the planet expire, and it is commonly believed that the particular balance that Earth has acheived between varying conditions (ice ages vs. heat ages) that don't vary enough to kill all the advanced life on the planet has gone a long way towards accelerating evolution's "progress" (in human terms). Along with all of the other bands Earth had to be in the middle of to match the conditions for us, there may also be a relatively narrow "placidity/activity" band necessary for periodic purges of the dominant form of life, giving new meta-stable ecologies a chance to develop, without wiping the slate too clean. And there are any number of other reasons to believe that Earth's rate of evolution itself may be unusually fast; perhaps there are planets out there that could in theory harbor intelligent life, but for the fact that nothing can gain any traction against the Trilobite of the planet.

Obviously other estimates can generally be matched by discovering one or two other factors; it is better to think of the log of these numbers because in the probability domain we are dealing in "multiplication" is the primary operator, and the log domain will help our numerical intuition apply better.

The number of stars in the Universe is a Big Number. The odds against life is a Big Number. My point is actually not that intelligent life must be rare (though that is my personal belief), but that since our ignorance of the size of the odds against life dominates our ignorance of the size of the Universe, it is a scientifically defensible position that life may be very rare, and Earth may indeed be the only planet in the Universe that can independently evolve intelligent life. The mere existance of a Big Number representing the number of stars in the universe is not enough to establish the proposition that "there just gots ta be more intelligent life out there!"; that's falling for the fallacy. It must also be established that it dominates the odds against such life, and we simply don't know enough to have a good enough idea what those odds are. Semi-detailed knowledge of nine-ish planets (all in one possibly unusual solar system and certainly not independent of each other at all) and a handful of size and orbit parameters for super-Jovian planets in the immediate vicinity just doesn't cut it.

Therefore, it is not sound to mock people (like me) who believe that there may not be a whole lot of intelligent life out there, and that we may even be the only intelligent life in the universe. You may just find that we believe that because we have a better grasp on the size of the universe than you may.

(One last comment: It is often pointed out that it sure would be some sort of "waste" if this whole universe only contains us. Since the term "wasted" implies some sort of scarcity, and the desirability of allocating the resources more efficiently for some purpose to avoid said wastage, what scarce resource is being misused? Especially that given that in most religious answers for how the universe was started and all current viable physics answers, the amount of matter produced and the size of the final universe is to all appearances an unbound quantity; there is no reason God could not have created a universe a hundred times smaller, there is no reason the initial singularity could not have resulted in a mass explosion a hundred times bigger, in either case needing no more or no less of any "resource" or "effort". I think this attitude itself shows a sort of self-centeredness, demanding that the universe be dedicated to some sort of "meaning" or "task" comprehensible or acceptable to the speaker, instead of existing for unknown, unacceptable, or even no reasons.)


Site Links


All Posts