Jump to content
Objectivism Online Forum

Longevity Of Human Life Relative To Civilization

Rate this topic


DavidV

Recommended Posts

Consider the following thought experiment:

I am going to pose two scenarios regarding human civilization and make a claim regarding the likelihood of one of those scenarios. I would like you to consider whether my reasoning for making that claim is valid.

For purposes of this experiment, I define human civilization as "a state in which mortal human beings are physically and intellectually capable of considering this scenario" I'm going to semi-arbitrarily set the beginning of that date to 500BC (the Greek philosophers).

Here are the scenarios: (1) human civilization as we know it lasts for a length of time several orders of magnitude longer that it has already (i.e. 10,000 – 10 million years) (2) human civilization will end within 50 years.

Let's consider (1).

Take a length of time – say 100,000 years. Suppose that the average human lifespan is 100 years, (64 now) and the population averages at 10 billion (6 now).

Question: what is the probability that you will be born in the first 50 of that 100,000 years? If my math is right, it's (10b x 100)/(10b x 100,000) = 1/1000 or .1%.

If (1) is correct, we are in fact in that .1%, since the population passed 3 billion sometime after 1950, and there are less than 4 billion people that were alive from 2500BC to 1955.

Now consider (2): human civilization ends in 50 years. To approximate, 10 billion will have lived for 100 years (a generous 4 billion for everyone who lived 2500BC-1955). The probability is then (10b x 60)/(10b x 100) = 6/10 = 60%

Conclusion: There is a 60% of being alive now if human civilization will end in 50 years and a .1% chance of being alive now if human civilizations will last 100,000 years. Considering these statistics, it is overwhelmingly unlikely that human civilization will exist for much longer, and scenario (2) is correct. Even if it lasts till the year 3,000, your chances only improve to 3%.

There are two possibilities for scenario (2): either civilization becomes extinct, or we all become effectively immortal. Not literally immortal, but long enough that the percentages are in our favor – a few hundred million years, perhaps.

Again, this is a thought experiment, and does not mean that I am preparing my will for my non-existent progeny.

(Edit: changed title to be more neutral, 500BC date corrected)

Edited by GreedyCapitalist
Link to comment
Share on other sites

Let me try to summarize:

It’s highly unlikely that we would live at the very beginning of human civilization if our lives were a small percentage of the total duration of civilization. But since that is true, it is likely that our lives compose a significant percentage of the total duration of civilization, which means that either human civilization is very short, or our lives will be very, very long.

Link to comment
Share on other sites

Let me pose the same problem a different way, David..

Suppose I number 1 million balls and then pick a random one. The chance that I pick a particular ball is 1 in a million. Since this is so small a probability, should I assume there are actually only (say) 1000 balls?

Link to comment
Share on other sites

I considered this analogy. But I don’t think it applies because none of the balls are more significant than any other. But if the balls were numbered 1 to 1,000,000 and you got 1 or 1,000,000, wouldn’t you suspect that the selection wasn’t random?

Link to comment
Share on other sites

But if the balls were numbered 1 to 1,000,000 and you got 1 or 1,000,000, wouldn’t you suspect that the selection wasn’t random?

The problem with "suspecting" such a thing is that you would need evidence for that suspicion. Mathematically and logically, there is no less chance of picking ball #1,000,000 than there is of picking #419,457. I have to echo Jennifer in saying I don't understand the purpose of this thought experiment, as the duration of civilization has nothing to do with the probability of my existence at this time. There is no causal relationship being presented.

Link to comment
Share on other sites

If you got the ball numbered 555,555 that would be odd too. However, that does not imply that it was not random.

Well, ok. But history is not a list of numbered balls either. It is a progression, an evolution, and if the conditions are right, a continual improvement in man’s condition.

Now in fact, I had this idea because of certain empirical evidence. I have been studying social and technological progress across several fields and found a pattern of exponential acceleration across thousands, or hundreds of thousands of years. In the fields where progress can be numerically measured, that growth curve inverts in the 2020’s and approximates a vertical slope in the latter 21st century.

There is a lot more theory to all this, but I don’t think it is necessary to consider the point I make. We do know that technology is evolving, that we have the potential to destroy ourselves, and that if the past is any indication, the future will be very different from the present. We know that we are at the beginning of history, and if there is a logical explanation that resolves improbabilities into probabilities, that may be an indication of the nature of the changes to come.

Link to comment
Share on other sites

Suppose I number 1 million balls and then pick a random one. The chance that I pick a particular ball is 1 in a million. Since this is so small a probability, should I assume there are actually only (say) 1000 balls?

I just thought of a more accurate restatement of your analogy. In fact, we don’t know how many balls there are. It may be any number between 2500- and infinity. If I pick up a random ball and it’s 2050, it would be logical to conclude that the number is probably closer to two thousand than two million. In fact, it would be evidence against Noahs premise – that there is a causal relationship.

In science, it is common to discover a statistical improbability without knowing the underlying cause. The proper step is not to discard or ignore the evidence, but to determine whether the underlying causal relationship is significant.

Link to comment
Share on other sites

Conclusion: There is a 60% of being alive now if human civilization will end in 50 years and a .1% chance of being alive now if human civilizations will last 100,000 years.

I think I can clarify my objection, which is that you are using the concept of probability in a invalid way or at least in an improper context. Probabilities can tell us something meaningful when they arise out of facts of reality, or when they pertain to entities which have identities (and therefore causal relationships with other entities).

For example, the 50% probability that a flipped coin has of landing heads up arises from the fact that the coin has two possible states, heads and tails, and that these states are part of the coin's indentity. When the coin is flipped, it MUST result in one or the other state. Clearly, there is a causal relationship between the coin's being flipped and its resulting in a particular state 50% of the time.

Or, taking a more complex example, one might be able to compile statistics on a certain disease. Given enough data to match with a particular patient's symptoms and genetics etc, you might be able to reasonably assign that person a probability of X% of surviving the disease. In this case too, the probability involves a causal relationship between the disease and the health of the patient.

But in your example, you cite a "probability" of our being alive at a given time in history, as if there were an alternative, and then take this to imply that it is "unlikely" that civilaztion will continue much longer. I don't think probability applies here (in any meaningful sense) because there is no causal relationship between our existence at this time and the longevity of civilization. The one fact has no impact on the other. You could just as well compare the probabilities of winning the powerball and of Condi Rice getting elected in 2008.

Link to comment
Share on other sites

I think your probabilities are wrong.

Take a length of time – say 100,000 years. Suppose that the average human lifespan is 100 years, (64 now) and the population averages at 10 billion (6 now).

Question: what is the probability that you will be born in the first 50 of that 100,000 years?

This is impossible to calculate without information about the birth rates and death rates because the answer is indeterminate. For instance, we could have 5 billion people at the start (now), which drops to 1 million next year due to nuclear holocaust, before experiencing exponential growth and working its way up to 20 billion. In this case, your probability of being born within the first 50 years would be fairly low. Alternatively, we could have linear growth for all the way (unlikely), in which case your probability of being born in the first 50 years would be a lot higher.

Even with knowledge about the birth/death rates, this would still be a non-trivial probability calculation and may involve solving non-linear differential equations (I'm too tired just now to think properly about this).

Now consider (2): human civilization ends in 50 years. To approximate, 10 billion will have lived for 100 years (a generous 4 billion for everyone who lived 2500BC-1955). The probability is then (10b x 60)/(10b x 100) = 6/10 = 60%
I dont understand what youre calculating here. Edited by Hal
Link to comment
Share on other sites

Hmmmm, I think I see what youre saying. If X is the total lifespan of the human race, and P(Y) is the probability of you being born in the first Y years, then for any fixed Y, P(Y) increases as X becomes smaller.

If so, this is correct in a sense. Its like saying that if youre choosing a numbered ball out of a bag with B balls, then for any N<B, the probablity of drawing ball N increases as B decreases (theres more chance of drawing '47' when you have you 100 balls than when you have 1000). So the question is whether you can make a reasonable inference about how many balls are in the bag, given that you have drawn some number N.

Its an interesting question, and I'm not sure if theres an answer. If I drew ball #43, would I be more entitled to conclude that there are 100 balls in the bag than that there are 10000? Intuitively, it does seem like this should be the case. But lets do the math.

Let B be the probability of there being B balls in the bag, and N be the probability of me drawing ball N. We are interested in P(B|N) (the probability that there are B balls in the bag given that I draw ball N). Using Bayes Rule, we have

P(B|N) = P(N|B)P(B)/P(N)

But the problem here is that we dont know P(B) - we dont know the prior probability of there being B balls in the bag. Without this, we cant possibly get a figure for P(B|N), hence we arent entitled to conclude anything about the total number of balls in the bag, regardless of what we draw.

(edit: to clarify, P(B) is the probability of there being B balls in the bag, before we draw anything. Ie given that we only know there is a bag with some balls in it, what is the probability of there being B balls?)

Now, in real life when it comes to drawing balls, we actually DO have some idea what P(B) is. If someone randomly asks you draw balls out of a bag, its fairly unlikely that there are going to be 20 billion balls - what kind of lunatic would even own this many balls? And how big would a bag have to be to fit them all? Realistically, not many people are ever going to use more than a few thousand balls. So we can reasonably put an upper bound on B, say B < 10000 and then guess an appropriate prior distribution P(B) (we wouldnt really have to specify a definite upper bound, we could just choose a distribution where P[b>10000] was quite small). And the calculation we made using this distribution would probably justify our assumption that there are more likely to be 100 balls than 10000 balls given that we have drawn ball #47

But when it comes to esimating the lifespan of the human race, do we have any grounds for deciding what the prior probablity of the human race lasting for X years is? I dont think so - we dont know anything about any other advanced civilisations, so we have no real empirical grounds for saying how long a civilised species is likely to last. And without this sort of knowledge, we cant really justify any choice of prior. So no, I dont think you can justifiably infer anything about the lifespan of the human race from the fact that you are born within some particular period of time. But if we managed to somehow study alien civilisations and find out how long they lasted, this would change.

Its an interesting question anyway.

Edited by Hal
Link to comment
Share on other sites

Let me pose the same problem a different way, David..

Suppose I number 1 million balls and then pick a random one. The chance that I pick a particular ball is 1 in a million. Since this is so small a probability, should I assume there are actually only (say) 1000 balls?

This isnt the same problem, because here you know there are 1 million balls in the bag. Imagine you had no idea how many balls were in the bag, and you draw out the ball labeled #1. You now have to guess how many balls there were.

My argument here is that a reasonable guess would maybe be "between 2 and 10", but that this is only justified because we know in advance that people who make ball guessing games arent very likely to put 4 million balls in the bag. If we lived in a culture where most ball guessing games normally involved around 500 balls, then guessing '500' might be more justifable than guessing '10', even if you draw the ball labelled #1.

edit: here's a good example - you are told that in the national Turkish lottery last sunday, the first ball to be drawn out of the machine was labelled '#2'. Given this, how many balls do you think were in the machine? Here I would guess '50', because I know that the British lottery uses around 50 balls, and this strikes me as a normal number for a lottery game. In other words, I'm using the prior knowledge I have about lotteries in order to form a reasonable guess (this is the essence of Bayesian inference).

Edited by Hal
Link to comment
Share on other sites

There is a 60% of being alive now if human civilization will end in 50 years and a .1% chance of being alive now if human civilizations will last 100,000 years. Considering these statistics, it is overwhelmingly unlikely that human civilization will exist for much longer, and scenario (2) is correct. Even if it lasts till the year 3,000, your chances only improve to 3%.

Given that I have written this and you are reading it, I would say that there is a 100% chance of being alive now in either scenario.

Link to comment
Share on other sites

If I pick up a random ball and it's 2050, it would be logical to conclude that the number is probably closer to two thousand than two million.
True. However, you make a presumption of randomness even though your broader context (the observer is here, at this point, in a chain of unknown length) implies that the sample is not random. To accept it as random, you have to assume things that you have no reason to assume.

Suppose I were to sit on a street corner and observe how many black people went by. Suppose 90% of the people who went by were black, what conclusions could I draw? The example of looking at ourselves further complicates the sample: if a black man looks in the mirror, what is the probability that he will see the reflection of a black man?

...growth curve inverts in the 2020's and approximates a vertical slope in the latter 21st century...
No it is not vertical. It only looks that way because of the scale you choose on the graph. The part of the curve that you think is gradual could have appeared vertical to a person living 200 years ago, looking at it in his context, using a scale that he thought was reasonable.
Link to comment
Share on other sites

However, you make a presumption of randomness even though your broader context (the observer is here, at this point, in a chain of unknown length) implies that the sample is not random. To accept it as random, you have to assume things that you have no reason to assume.
I've been trying to figure out what the deal is here in this thread, and for me this plus David's earlier statement
In science, it is common to discover a statistical improbability without knowing the underlying cause. The proper step is not to discard or ignore the evidence, but to determine whether the underlying causal relationship is significant.
are the essential points. The validity of an inference from observed distribution compared to theoretical ideal distribution is usually taken to depend on randomness. This is a correct assumption as long as we understand what randomness is: it refers to causal factors which we are mostly clueless about. When we observe that there are too many people given what we know from simple combinatorics, then we have learned that there is some causal factor that we have to consider -- e.g. rate of reproduction may be influenced by a person's age or by technological factors that counterbalance disease and starvation which kill off people. The presumption of "randomness" is essential to becoming aware of previously unknown causal factors.
Link to comment
Share on other sites

Now in fact, I had this idea because of certain empirical evidence. I have been studying social and technological progress across several fields and found a pattern of exponential acceleration across thousands, or hundreds of thousands of years. In the fields where progress can be numerically measured, that growth curve inverts in the 2020’s and approximates a vertical slope in the latter 21st century.

There is a lot more theory to all this, but I don’t think it is necessary to consider the point I make. We do know that technology is evolving, that we have the potential to destroy ourselves, and that if the past is any indication, the future will be very different from the present. We know that we are at the beginning of history, and if there is a logical explanation that resolves improbabilities into probabilities, that may be an indication of the nature of the changes to come.

Isn't this the singularity theory? The idea that a confluence of scientific advancements will result in a fundamental change in what it means to be "human"

Link to comment
Share on other sites

Isn't this the singularity theory? The idea that a confluence of scientific advancements will result in a fundamental change in what it means to be "human"

I would define it differently, but yes. I am only focusing on a particular premise – that of accelerating change. However, I wanted to keep this particular thread limited to the specific point I brought up, which I think can be examined independently. If my reasoning is valid, especially the statistical correlation with the middle of the 21st century, it could be important evidence for that theory.

Link to comment
Share on other sites

No it is not vertical. It only looks that way because of the scale you choose on the graph. The part of the curve that you think is gradual could have appeared vertical to a person living 200 years ago, looking at it in his context, using a scale that he thought was reasonable.

What I mean is that an exponential growth curve has a “bend point” where the growth becomes much more apparent in comparison to linear growth. The rate is unchanged, but it becomes much more evident in the context of a human lifetime.

Link to comment
Share on other sites

But when it comes to esimating the lifespan of the human race, do we have any grounds for deciding what the prior probablity of the human race lasting for X years is? I dont think so - we dont know anything about any other advanced civilisations, so we have no real empirical grounds for saying how long a civilised species is likely to last. And without this sort of knowledge, we cant really justify any choice of prior.

Hal, I’m not sure whether I agree with you about the need to for prior probabilities, but I think we have several important datum points. From a doomsday perspective, we know that of 99% of the species that have existed are extinct, and that there is a non-trivial chance of a nuclear or biological holocaust wiping us out. From a positive perspective, we know that man is a very flexible creature, and our ability to control our environment is improving rapidly, and that the pace of change itself is accelerating.

Here is an analogy I just thought of:

We know that all previous human civilizations have collapsed sooner or later. But suppose you didn’t know that because you were living in the very first human civilization, Sumer, which lasted from 3500 to 2000 BC. If you were living in the year 1999BC, would it be reasonable to conclude that your civilization is unlikely to last beyond another 2000 years, at least with the current population levels?

I think that would be a reasonable conclusion, especially given the fact that the Babylonians, among other peoples are itching to raze Ur.

Likewise, we know that our civilization is relatively young, undergoing rapid change, and there are several possible threats and opportunities facing us as a species.

Link to comment
Share on other sites

Conclusion: There is a 60% of being alive now if human civilization will end in 50 years and a .1% chance of being alive now if human civilizations will last 100,000 years. Considering these statistics, it is overwhelmingly unlikely that human civilization will exist for much longer, and scenario (2) is correct. Even if it lasts till the year 3,000, your chances only improve to 3%.

(Edit: changed title to be more neutral)

David,

The calculations you made will be relevant only if something along the lines described below takes place -

Out of all the people who lived (and are ever to live) during the timeframe of human civilization, one person is randomly chosen.

Now if it is a given that human civilization will end in (or around) the year 2056 (i.e. in 50 years), then there is a probability of that randomly chosen person being alive in the year 2006 (the present) of (around) 60%.

On the 0ther hand, if it is a given that human civilization will last for more than 100,000 years, then the probability of finding that the individual who was randomly chosen exists in the year 2006 would be less than 0.1%.

Now say we make such a random choice. Out of all the people who have ever lived and are ever to live within human civilization, we choose one individual and it turned out that the person chosen is alive in the year 2006.

Now what are we to make of such an eventuality?

Should we conclude that civilization is very likely to end around the year 2056?

Intuitively, such a conclusion would seem reasonable.

The thing is, it is fairly obvious that it would be impossible to conduct such an experiment where one person out of all the people who ever lived (and are ever to live) within this civilization is chosen randomly. [One would have to be a supernatural and semi-divine being to do so.]

The error in your thought experiment is the implicit assumption that such an experiment had been performed and that you are the result of such an experiment. The reasoning seems to be something like –

If civilization were to last around 100,000 years, then the probability of me living within the very beginning of that civilization is extremely low. On the other hand if civilization were to end around the year 2056, then the probability of me being alive around the year 2006 is 60%.

Well it turns out that I am in fact alive in the year 2006. WOW! That probably means that one way or the other (either through extinction or by achieving immortality) this generation is likely to be the last generation of this civilization!

I think pointing at yourself as being one among all the people who ever lived (and are ever to live in the future) within this civilization would amount to an arbitrary choice rather than the result of a random experiment. Therefore the calculations that you made would not be relevant.

[Also, there were no Greek philosophers in the year 2500BC. I think Greek philosophy began to appear around 600BC. But that is not central to the thought experiment that you described.]

[Edited to correct a grammatical error.]

Edited by shakthig
Link to comment
Share on other sites

[Also, there were no Greek philosophers in the year 2500BC. I think Greek philosophy began to appear around 600BC. But that is not central to the thought experiment that you described.]

Oops, I meant to say 2500 years ago. (Socrates was born around 470BC)

I think pointing at yourself as being one among all the people who ever lived (and are ever to live in the future) within this civilization would amount to an arbitrary choice rather than the result of a random experiment. Therefore the calculations that you made would not be relevant.

There are two types of scientific experiments: controlled experiments and observational experiments. Much of our scientific data is based on observational rather than controlled experiments, such as astronomy, archeology, evolutionary biology, etc.

The word “arbitrary” means “without logical a tie to reality.” Clearly, the facts I state are observations of reality. The proper word is random – I don’t know (for a fact) the causes of my being born now, but it is a fact that I was. (Random means “the causal factors are unknown”) The scientific process essentially turns random observations into meaningful knowledge about the world.

Refer to my last post for an example similar to the one I make.

Link to comment
Share on other sites

The word “arbitrary” means “without logical a tie to reality.” Clearly, the facts I state are observations of reality. The proper word is random – I don’t know (for a fact) the causes of my being born now, but it is a fact that I was. (Random means “the causal factors are unknown”) The scientific process essentially turns random observations into meaningful knowledge about the world.

I used the word “arbitrary” to indicate the fact that the choice of the person whose birth of date would determine the likelihood of civilization ending soon depended on your discretion. [http://www.m-w.com/dictionary/arbitrary]

Why wonder about the date of your birth? Why not wonder why Einstein was born in the 19’Th century, why George Washington was born in the 18’Th century, why Christopher Columbus was born in the 15’Th century and why Alexander the Great was born around 2000 years ago etc. The timing of their birth days was as random as yours.

I get what you mean by “observational experiment” (Though I think that the word “experiment” would be redundant in this case. Wouldn’t it be simpler to say that most knowledge in fields such as astronomy and archeology depend on data gathered from observations?)

But it is important to keep in mind the context within which the observation is taking place. I think the issue gets complicated by the fact that you are both the observer and the subject of the observation.

You are the one making the observation and the one who interprets the results of the observation and you are making that observation in the year 2006. That is the context within which the observation is taking place. That being the case, the “probability” of you “observing” that you happen to be alive in the year 2006 is 100% regardless of when civilization is to end. The calculations you outline would not be relevant.

Those calculations will be relevant only for an experiment that would result in a “pointer” to one person out of all people who ever lived (and are ever to live in the future) within civilization.[And only when there is an equal probability of any individual being pointed to by the result of the experiment.]

Link to comment
Share on other sites

  • 3 weeks later...

"It is likely that unlikely things should occur"- carl sagan(?)

The problem I am seeing is that the people you refer to are not random samples.

If you could take every human ever born-pluck out 50 of them at random and find that they were all born before 2056, then it would have meaning, but what it seems like is that you are picking 50 people born in the 20th century and saying "what are the cahnces they were born in the 20th century?". That's like having numbered balls from 1-1000 that are visably labled, choosing 1-10 in order and then asking "what are the chances I picked 1-10?". My answer is that the chances are pretty damn good. But I'm a smart-ass. So what am I not understanding?

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...