this is mostly for Galactus(should e youtube again)

Oct 04, 2008 10:43

This is one of my favorite debate topics, and is phrased here much better then I ever would be able to.  Read it, and then respond with thoughts.

The Doomsday Argument is something that was first proposed by the Cambridge astrophysicist Brandon Carter in a lecture to the Royal Society in the early 1980's. It demonstrates from basic probability theory that the risk of our species going extinct in the near future is much greater than previously thought. The method has since been developed in a Nature article by Prof. Richard Gott (1993), and in several papers and a book by philosopher John Leslie (The End of The World, 1996). Being itself based on game theory, a simple game will help to establish the argument:

Imagine a sealed wooden box, large enough to hold maybe a gallon of liquid. The box has a single grooved outlet and a wooden lever alongside. In this box there are a number of small marble balls, all of them black -- except for one only, which is white. When the lever is pressed, a single ball at a time will be retrieved from the box into the slot, in random order. Now, suppose I promise you that there are either ten balls in the box, or a thousand -- and then ask you to press the lever.

You press the lever, and a ball pops out. Its color is black. You press again: another black ball. However, you try once more, and the third ball produced is white. Whoa! If I now ask you how many balls there were in the box -- ten or a thousand -- what is your answer? Give it some thought, the answer shouldn't be very difficult. :confused: Ready? OK. Your answer is ten; are you sure? Okay :D Your intuition is sound; if there were a thousand balls in the box, it wouldn't be likely that you would have reached the white ball so quickly. Based on Bayes' theorem, the probability that you are correct is two-thirds, on the basis of the white ball being third out.

Now, let's turn our attention away from the game, and let's think about the future. Specifically, how the human population might be expected to develop from its current point. There are three possibilities:

a) Expansion: the continued increase of human numbers, presumably requiring an eventual move off-planet. Perhaps even the colonization of other star systems, and far in the future, of the entire galaxy itself.
b) Stabilization: we settle for resources on the Earth, and find a way to manage our numbers and our planet indefinitely.
c) Extinction: for whatever reason -- asteroid impact, global war, biotech disaster, runaway nanotechnology -- the human race dies out.

It is quite a remarkable fact that at least ten percent of all the humans who have so far existed are alive right now, in our present time. However, in the first two possible cases stated (a and b), the vast majority of humans are still yet to be born. Even staying on Earth, there are by some estimates billions of years ahead of us before Earth's atmosphere is rendered uninhabitable by changes in the aging sun -- we have far more future than past. And if we -- or our engineered descendants -- expand to the stars, the possibilities are even grander. Suppose we colonize the galaxy; the total human population, over time, might then easily reach a billion times its present number.

Therefore, if the stars really have our name written all over them, we have to believe that we presently alive belong to the first one-billionth (or so) part of the total human population including past, present and future. So why are we alive here and now? How likely is that? Remember the game with the balls and the box: it's as if you drew the white ball out third of a thousand balls, instead of ten -- only even far, far more unlikely than that. It is more probable by many orders of magnitude that you'd have been born in the future instead of now.

The simple answer is: it is much more likely that we are among the last 10 % of all humans who ever lived, or will ever live, than the first 0.000000001 %. Suppose you were told that the world will end tomorrow; you might think yourself unlucky that your natural life span has been cut short so abruptly. But in fact, one in ten of all humans -- that is, the ones alive now -- would be in the same boat with you. Losing out to one in ten odds is unlucky, certainly, but not drastically so.

Oh-kay... I hope you've reached this far :) I know what you're thinking: absurd! Upon first encountering the argument, most people think it has to be obviously false. However, reflect on it for a while: trying to explain what is wrong with it makes for some heavy thinking. The argument is devilish; impossible to prove, but seemingly as hard to refute. There have been numerous attempts to shoot it down since its conception nearly two decades ago; but for every paper trying to refute the argument there have been an approximately equal number of papers refuting these refutations.

I originally encountered the concept of this so-called "Carter Prophecy" in the novel 'Time' by Stephen Baxter, from where I've adapted important parts of my explanation of it, sometimes almost verbatim (the novel is quite good, by the way -- I recommend it to any of you science fiction buffs out there). Having since read more about the Doomsday Argument on the web, I don't know exactly what to make of it. Yes, it's still absurd, but on the other hand, it makes perfect sense.
Previous post Next post
Up