Celestia Forums

Something to consider - this was first proposed by Brandon Carter:

Imagine I have a house with 100 rooms, numbered 1-100.

I toss a fair coin, and if it lands heads, I create a person in each of the rooms 1-10. If it lands tails, I create a person in each of the rooms 1-100.

You wake up knowing this information, and find yourself in room 7. Did the coin land heads or tails... i.e. given that you are in one of rooms 1-10, what is the probability that only 10 people were created?

If the coin landed heads, then the probability that you are in one of rooms 1-10 is 100%. If the coin landed tails, then the probability that you are in one of the rooms 1-10 is 10%.

So given what you know, and your existence in room 7, you calculate the probability that the coin landed heads is about 91% - it is much more likely that only 10 people were created.

Now, suppose we number all the humans that have, do and will exist according to the order in which they are born - the first human is number 1, the second is number 2, etc. Present day humans have numbers ~60 billion.

Now suppose there are two futures for the human race - an imminent extinction, in which there are only, say, 100 billion humans ever, or a scenario where we expand into the galaxy, and the total number of humans is many trillions.

Assuming both of these futures have the same probability, the imminent extinction scenario is far more likely than the expansion, for the same reason as the person in room 7 would consider the situation of there being only 10 people created to be more likely, as opposed to the situation where 100 people are created.

chaos syndrome wrote:Now, suppose...

chaos syndrome wrote:Now suppose...

chaos syndrome wrote:Assuming both of these futures have the same probability...

You assume quite a lot, but anyway... The reason for mankind's future extinction will very likely not be considerations like this

chaos syndrome wrote:Assuming both of these futures have the same probability, the imminent extinction scenario is far more likely than the expansion, for the same reason as the person in room 7 would consider the situation of there being only 10 people created to be more likely, as opposed to the situation where 100 people are created.

Ummm, no.
This notion assumes several things that could break the premise:

A) It assumes that the number of people living simultaneously always will increase. I find that unlikely. Our culture might just as well decide that (pick any number) of people is ideal, and eventually settle there. If, for example, we end up with a society of one million persons with extremely long lives, the argument falls apart...

B) It assumes that human "souls" are a concrete, tangible resource, where each "soul" is an unique entity that can be numbered. There is no way to know that. We might all be different parts on one greater whole - which invalidates the entire thing. Likewise if reincarnation is real... Or if everything is pre-ordained by The Spaghetti Monster...

C) It assumes that the universe works like a lottery. It might not...

D) It assumes that there is a qualitative difference between yesterday and tomorrow. There might not be...

In short, the argument is based on a few human-centric ideas on how the universe works. I??m sure i could find more examples...

- rthorvald

A) It assumes that the number of people living simultaneously always will increase. I find that unlikely. Our culture might just as well decide that (pick any number) of people is ideal, and eventually settle there. If, for example, we end up with a society of one million persons with extremely long lives, the argument falls apart...

No... what matters is the total number of humans that have and will exist. Even if a population of X is achieved and maintained, births are still needed to compensate for any deaths which occur, and this increases the total.

B) It assumes that human "souls" are a concrete, tangible resource, where each "soul" is an unique entity that can be numbered. There is no way to know that. We might all be different parts on one greater whole - which invalidates the entire thing. Likewise if reincarnation is real... Or if everything is pre-ordained by The Spaghetti Monster...

- rthorvald

Assumptions are pointless. As I tell my students, when you assume, you make an ASS out of U and ME... (heh, that one is older than the dinosaurs.) The future of mankind, I believe, is governed by mankind (granted the Sol system doesn't suddenly meet up with an Interstellar Cloud). We can choose to teach to forward humanity (and cover our butts) or choose to sit on our butts. I teach that the more you know about your surroundings (earth and beyond), the better off we will be, showing that it is better to be proactive rather than waiting for the Flying Spaghetti Monster to touch you with his noodly appendage and make you a pirate. GI Joe was right, knowing IS half the battle. Make sure you teach what you know!!! It just might make a difference 100 or even 1000 years from now.

This sounds like the counter-intuitive probability statistics of life expectancy: as you get older, your life expectancy increases. In other words, if you make it to age A, then you are more likely to reach age B, B>A.

If I understand the way you've put the argument, Chaos, the '10% of rooms' and 'room 7' is like saying we in 2005 are at 7% of the way through humanity's expected maximum lifetime. The chances of us making it to 10% of humanity's expected maximum lifetime from the beginning of humanity's lifetime is 50%.

Applying these kinds of odds to an individual's life expectancy, say a human of 75 years' life expectancy at birth, is to say he has only a 50% chance of surviving beyond 7?? years age and seems very bleak.

Also, we're left with guessing humanity's expected maximum lifetime too because we have no real data for lifetime expectancy of humanities (we know of only one, and it ain't extinct yet), and that in turn leaves us with trying to figure out when humanity was 'born': 10,000 B.C.? 100,000 B.C.? If 100,000 B.C., and 2005 is the 7% mark, then humanity's expectancy is until 14,300,000 A.D. and we've got a 50% chance of making it to 431,000 A.D. Not bad. Fits in with H G Wells 'The Time Machine' too.

So, I think the construction of the argument is OK, because I think it's based on a constant probability of death/extinction throughout time. But why those particular numbers? Why 50% at 10%? Why not 99% at 10%? And why can someone say we are at 7% (room 7) right now?

rthorvald wrote:Or if everything is pre-ordained by The Spaghetti Monster...

Defiler! The Flying Spaghetti Monster to you, me hearty!

Spiff.

chaos syndrome wrote:what matters is the total number of humans that have and will exist. Even if a population of X is achieved and maintained, births are still needed to compensate for any deaths which occur, and this increases the total.

Yes, but it spreads out the number of rooms more evenly - and so the odds you live at the end times decrease, no matter which age you live in...

-rthorvald

Spaceman Spiff wrote:And why can someone say we are at 7% (room 7) right now?

It is years since i read this, but if i remember correctly, that is the percentage of all people that has ever lived that is alive right now (or then).

-rthorvald

rthorvald wrote:that is the percentage of all people that has ever lived that is alive right now (or then).

Oh! That is interesting. I heard some soundbite from some humanitarian advert claiming that most people who ever lived are alive now. I thought that implied a very narrow view of our past: 1) that pre-civilisation humans don't count (they were quite as intelligent as us) and 2) that they dismiss estimates that the human population has been around 500 million for most of humanity's history. So I'm glad to hear of a corroboration that my suspicions are correct!

Anyway, so the people alive today comprise the latest 7% of the population. How does that turn into: 'are at the 7th percentile of all humans who will ever exist?' I don't follow, Runar.

In general, if we consider the human population to be exponentially increasing until sudden extinction, then it's not difficult to realise that any randomly selected human will come most likely from near the end (like Philip Jose Farmer's 'Riverworld': most resurrectees came from the 20th century, none from the 21st). But, still, the fact that we have the highest population now doesn't tell us that we are right near the end, does it? It's like stock indices: a stock index is more often breaking records than falling, but that doesn't stop people saying, 'it has to fall now' while it goes on to double or triple before falling, then still start rising again...

Spiff.

The Carter Argument has one major wrong assumption, that there is no reason for the observation to occur at any particular point in history- that is to say that the action of self sampling is completely random.

However tht is not the case; the action of self sampling is not random, but can only occur at one point in the history of the human race- and that point is now.
If instead of just looking at the human race, you imagine a much wider set of alien intelligent races with the sort of population distribution over time described by Carter; they will have a range of populations from very small to very large compared to our current situation, but they will all have become aware of the Carter Argument at approximately the Same time in their history, that is to say during the early part of their technological development.
The discovery of statistical methodology by any alien race similar in outlook to humans (a particular subset of the presumed whole) is likely to coincide with the particular state of population described in the argument. In short, they will discover the Carter Doomsday Argument
at precisely the time when it appears to apply, not before, and not afterwards.

Much later in the history of that subset of intelligent races similar to humans the population will be much larger, and the Carter argument will be disproven.

I am basically saying that we are in a privileged position wuth regard to self-sampling of this nature; therefore it is not random, and arguments based on random self-sampling are wrong.