[Link] Cosmological Infancy
A post by Nick Land who some of you are probably already following either on his blog Outside In or at Urban Future.
There is a ‘problem’ that has been nagging at me for a long time – which is that there hasn’t been a long time. It’s Saturday, with no one around, or getting drunk, or something, so I’ll run it past you. Cosmology seems oddly childish.
An analogy might help. Among all the reasons for super-sophisticated atheistic materialists to deride Abrahamic creationists, the most arithmetically impressive is the whole James Ussher 4004 BC thing. The argument is familiar to everyone: 6,027 years — Ha!
Creationism is a topic for another time. The point for now is just: 13.7 billion years – Ha! Perhaps this cosmological consensus estimate for the age of the universe is true. I’m certainly not going to pit my carefully-rationed expertise in cosmo-physics against it. But it’s a stupidly short amount of time. If this is reality, the joke’s on us. Between Ussher’s mid-17th century estimate and (say) Hawking’s late 20th century one, the difference is just six orders of magnitude. It’s scarcely worth getting out of bed for. Or the crib.
For anyone steeped in Hindu Cosmology – which locates us 1.56 x 10^14 years into the current Age of Brahma – or Lovecraftian metaphysics, with its vaguer but abysmally extended eons, the quantity of elapsed cosmic time, according to the common understanding of our present scientific establishment, is cause for claustrophobia. Looking backward, we are sealed in a small room, with the wall of the original singularity pressed right up against us. (Looking forward, things are quite different, and we will get to that.)
There are at least three ways in which the bizarre youthfulness of the universe might be imagined:
1. Consider first the disconcerting lack of proportion between space and time. The universe contains roughly 100 billion galaxies, each a swirl of 100 billion stars. That makes Sol one of 10^22 stars in the cosmos, but it has lasted for something like a third of the life of the universe. Decompose the solar system and the discrepancy only becomes more extreme. The sun accounts for 99.86% of the system’s mass, and the gas giants incorporate 99% of the remainder, yet the age of the earth is only fractionally less than that of the sun. Earth is a cosmic time hog. In space it is next to nothing, but in time it extends back through a substantial proportion of the Stelliferous Era, so close to the origin of the universe that it is belongs to the very earliest generations of planetary bodies. Beyond it stretch incomprehensible immensities, but before it there is next to nothing.
2. Compared to the intensity of time (backward) extension is of vanishing insignificance. The unit of Planck time – corresponding to the passage of a photon across a Planck length — is about 5.4 x 10^-44 seconds. If there is a true instant, that is it. A year consists of less the 3.2 x 10^7 seconds, so cosmological consensus estimates that there have been approximately 432 339 120 000 000 000 seconds since the Big Bang, which for our purposes can be satisfactorily rounded to 4.3 x 10^17. The difference between a second and the age of the universe is smaller that that between a second and a Planck Time tick by nearly 27 orders of magnitude. In other words, if a Planck Time-sensitive questioner asked “When did the Big Bang happen?” and you answered “Just now” — in clock time — you’d be almost exactly right. If you had been asked to identify a particular star from among the entire stellar population of the universe, and you picked it out correctly, your accuracy would still be hazier by 5 orders of magnitude. Quite obviously, there haven’t been enough seconds since the Big Bang to add up to a serious number – less than one for every 10,000 stars in the universe.
3. Isotropy gets violated by time orientation like a Detroit muni-bond investor. In a universe dominated by dark energy – like ours – expansion lasts forever. The Stelliferous Era is predicted to last for roughly 100 trillion years, which is over 7,000 times the present age of the universe. Even the most pessimistic interpretation of the Anthropic Principle, therefore, places us only a fractional distance from the beginning of time. The Degenerate Era, post-dating star-formation, then extends out to 10^40 years, by the end of which time all baryonic matter will have decayed, and even the most radically advanced forms of cosmic intelligence will have found existence becoming seriously challenging. Black holes then dominate out to 10^60 years, after which the Dark Era begins, lasting a long time. (Decimal exponents become unwieldy for these magnitudes, making more elaborate modes of arithmetical notation expedient. We need not pursue it further.) The take-away: the principle of Isotropy holds that we should not find ourselves anywhere special in the universe, and yet we do – right at the beginning. More implausibly still, we are located at the very beginning of an infinity (although anthropic selection might crop this down to merely preposterous improbability).
Intuitively, this is all horribly wrong, although intuitions have no credible authority, and certainly provide no grounds for contesting rigorously assembled scientific narratives. Possibly — I should concede most probably — time is simply ridiculous, not to say profoundly insulting. We find ourselves glued to the very edge of the Big Bang, as close to neo-natal as it is arithmetically possible to be.
That’s odd, isn’t it?
Star formation rates in the universe solve this problem.
See my old post on our position in space and time and how typical it likely actually is.
Partially quoted below:
More reading on my part makes me say that rather than being the ONLY place with high-metallicity stars, spirals are just by far the most common site for them. But the point still stands.
The stelliferous era refers to the time in which stars live, not the time in which they are forming. The longest-lived tiny stars will still be here a hundred trillion years from now, but there probably will have been almost no NEW stars formed for almost the entirety of their lifetimes.
If life either forms within a few gigayears of the formation of a star or not at all (something I find likely given the history of life on Earth and the state of current research on the origins of life, in fact it’s probably that it happens IMMEDIATELY after the formation of a star or not at all) then we should indeed expect to find ourselves near the start of the universe, when stars are still being born.
EDIT: You also just shouldn’t expect isotropy in time, there is no reason for it and indeed reasons against it. The universe is expanding. High-density, very interesting stuff happens near the start (after just enough expansion has occurred to provide some room between actual entropy and possible entropy in which structure can exist) and low-density, boring stuff happens the further in time you go forward.
That helps, but you also need an assumption that civilisations won’t expand and colonise the universe. If there are (or will be) such colonisers, we’re still atypical, regardless of star formation rates.
Why would anyone expect our intuitions to be accurate for universe creation events?
Some people believe that living today, assuming the growth of humankind, is an anthropic evidence that the end is near (otherwise we would be much more likely born later). If we apply the same thinking to the universe as a whole, and assume that if a species masters interstellar travel they can grow even more, it is an anthropic evidence that we will soon destroy the whole universe. Or at least that every species in our position either goes extinct or destroys the universe.
To try an optimistic interpretation, it is also possible that we will soon create a Friendly AI, which will discover that our extrapolated volition prefers preserving the existing minds to creating new ones, and since the energy of the universe is great but finite, creating new sapient beings will be forbidden to maximize the utility of the existing ones. Then the antropically most likely position is to be born a few decades before the Friendly AI is built.
(Anyone who takes this seriously, remember the Wizard’s First Rule—people will believe a lie because they want to believe it’s true, or because they are afraid it might be true. I provided both options here.)
Doesn’t the anthropic principle provide some difficulty for the latter solution as well—why should we find ourselves at the very beginning of such preposterously long lifespans?
*Our* extrapolated volitions might turn out to prefer immortality over reproduction, but it would be reasonable to guess that over all possible intelligent species, this would be relatively rare: Living things live to reproduce, literally. They’re all about reproducing. They like surviving too but it’s never the root goal.
So even if our CEV did turn out to favor immortality over reproduction, we would still find ourselves somewhere improbable, and we would still have to wonder, why?
I can imagine a sort of compromise CEV… Say we accept that immortality and reproduction are mutually exclusive. What if we ended up choosing a softer, less tragic kind of mortality where post-organic modes of communication allow all knowledge to be passed from parent to child, all projects continued. Might that be the norm instead?
Inflationary multiverse is essentially infinite. But as you take a slice through (a part of) the multiverse, there is way more young universes. The proportion of universes of given age is inversely (exponentially, as in memoryless distribution) proportional to the age. This resolves the doomsday paradox (because our universe is very young relative to its lifespan). http://youtu.be/qbwcrEfQDHU?t=32m10s
Another argument to similar effect would be to consider a measure over possible indices. Indices pointing into old times would be less probable—by needing more bits to encode—than indices pointing to young times.
Our universe might be very old on this picture (relative to the measure), so the conclusion regarding Fermi paradox is to update towards the “great filter in the past” hypothesis. (It’s more probable to be the first observer-philosopher having these considerations in one’s corner of a universe.)
See also http://www.youtube.com/watch?v=jhnKBKZvb_U
I don’t see a need for isotropy of the Universe. We may very well live on the edge of the World.
But experimentally, the Universe doesn’t look anisotropic on large scales.
Oh, yes, it looks! Big Bang only 14 billion years away, the end many trillion or so years away.
A very asymmetric situation, indeed.
I meant isotropic in space, not in time.
It’s not a need but a probability issue. We may indeed live on the edge of the World, it’s just surprising because the edge of the world is much smaller than the rest of it.
Repeat after me: scientific method does not deal with unique events or entities.
Such as the universe. You can’t repeat an “observation” that is, by definition, the sum of all observations. Unless we can get out of the universe and observe different universes and apply statistics to them, this whole talk about anthropic principle and such is nonsense.
I don’t think this is true in any meaningful sense. It routinely happens in astronomy or geology that we have a single instance of something and study it closely. For instance, I believe Oklo is the only site known to have had natural nuclear chain reactions. And nobody thinks studying the planet Venus is unscientific because there’s only one such planet.
You might say “ah, but we have many observations of Venus.” Well, we have many observations of the universe, too!
If a unique entity or event can be analyzed using the same theories and tools as other entities, it’s fair game for science. And the people who do anthropics are trying to apply statistics, which is a pretty general tool. You might think their methods are unscientific, but it’s not clear to me why the topic is objectionable.
Let me think of an example...
Imagine a universe comprised entirely of moving white balls. We study them and have a model that can predict their movements.
Suddenly, a black ball appears. We have never seen one before. But we observe it and find that it obeys the same laws of motion. So we conclude that the black ball is not unique in a sense that it is a BALL. We observed a lot of balls before.
But we never observed a BLACK ball before and if it remains unique, there is nothing science can tell us about its blackness.
The same thing applies to Venus. If we would invent a discipline called Venusology that only studies Venuses as a whole, as defined by a set of planets that are exactly similar to Venus, than that discipline won’t be able to do anything unless more than one Venus is observed. Until than, Venus would remain a unique phenomenon. But that’s not how Venus is actually studied.
When we study the orbital motion of Venus, it is not unique—just one of the objects that influenced to gravity. When we study the chemical composition of Venus’ atmosphere, it’s not unique—the elements it’s composed of are not unique, and there are other planets with atmospheres. When we study the geological features on Venus, they are not unique as well.
At that point you should realize that we don’t study Venus as a phenomenon—we study different aspects of Venus that we can apply the scientific method to. And the reason we can apply it is because the aspects are not unique.
The same things applies to the universe. We don’t have many observations of the universe, we have many observations of different parts of the universe. But if we want to study the whole universe as a phenomenon, then it becomes just one observation. And you can’t do anything scientific about it unless you observe another universe. (At which point you would have a new concept, the Multiverse, which again would be unique).
What caused you to invoke “Science!” in the sense you did here at all? It seems to hint a confusion about epistemology and a status move. It is easy to come up with scenarios where we are tasked with making predictions about things where science fails and do quite well. See this article by Eliezer.
I disagree. You know how scientific method works, right? You make an observation, create a model, make another observation and see how well your model fits. With a unique phenomena you only have one observation.
The whole anthropic principle thing just annoys me because it seems to stem from a confusion. People can certainly imagine a lot of universes without people. So they compare it to the real world and say it is fine-tuned. But this is a misapplication of scientific method. What you imagine is not an observation. You can’t invent different laws of physics and pretend they exist somewhere.
I mean, look at the text you posted. Nick Land thinks the idea that we are in the beginning of time is odd. Why? Does he have a window into another universe which isn’t odd?
How much of the sequences have you read?
Please forgive me for answering a question with a question, but have you heard about arguments from authority? They are bad arguments. ;-)
I was asking to see on what level to engage.
And if I told you I read 53.7% of the sequences (for example), would that help you gauge the level? Can’t you just write “you don’t understand Concept X, noob”? It would still be rude, but at least informative.
I wasn’t expecting you to reply with a percentage, more the names of sequences you’ve read. Inferential distances matter and my time is limited. I will now make the basic reply.