Hacker News new | past | comments | ask | show | jobs | submit login
Life on Earth may have begun 300M years earlier than previously thought (theguardian.com)
89 points by wrongc0ntinent on Oct 19, 2015 | hide | past | favorite | 30 comments



> If confirmed, the discovery means life emerged a remarkably short time after the Earth was formed...

So at what point does the lag time between accretion and first evidence of life become too short for "native" abiogenesis to be plausible, pointing toward a transfer scenario instead?


Honestly, that might be a different order of magnitude when it comes to speed -- maybe if we have evidence of life within the first ~10,000 years instead of hundreds of millions. We've only been trying to replicate the origin of life for couple hundred years, tops, at an extremely small percentage of locations on earth where we'd notice if it happened. It might be extremely likely to happen when you try everywhere on earth for hundreds of thousands of years, and that's a rounding error compared to dates we are looking at.

A much better source for what you are talking about would be e.g. finding tardigrades or similar in the asteroid belt. That would show that life was part of the mix that went into the forming of the planets.


Yeah, I suppose once we've identified an actual mechanism or two of abiogenesis, we'll be in a better position to estimate expected wait times in likely environments. And if the wait times are as fast as you suggest, it simply may not be possible to date rocks/minerals accurately enough for the necessary comparison.


On the other hand, if life creating was so relatively easy it would be much more likely to find evidence for multiple independent instances of life starting on earth. My understanding is that, so far, all life on earth is part 'one web' not multiple webs.


tl;dr It's very likely that first-life creates a lock-out, quickly soaking up all resources that could be used for independent abiogenesis.

Let's say the probability of new life is roughly constant (and a relatively low probability). It should roughly follow a Poisson distribution -- in other words, using current knowledge, and assuming (1) life would be created noticeably differently given similar circumstances and (2) it didn't (molecule chirality indicates both of these, for instance) AND (3) the events are independent, which you need for the distribution -- then it'd be pretty weird for something following a Poisson distribution where:

Earth Exists: Time 0. We have physical things from the time period: Time 2 Event in the distribution: Life: Time 4 Then absolutely no new instances until time 45 (unit here is 100MM years)

This is all by means of formalizing your argument, which is largely correct. But it's important to note that it assumes that the events are independent, which I strongly, strongly suspect they are not. Life takes resources and uses them up, and will expand until it meets carrying capacity. So first-past-the-post-wins seems like a much better guess as to the relationship between origins-of-life.


As far as I understand, the idea is that friendly conditions to life could turn into life in the geological blink of an eye, so there isn't such a point.


Another idea is those friendly conditions only seem to have occurred once here on Earth - everything living is descended from that one Thing.

That's the kind of train of thought that keeps me up at night.


At astronomical timescales even Earth's life is just a blink of an eye. It's not at all unlikely that life on other planets happened, is happening or will happen somewhere out there. The odds are just pretty good that we'll never be able to find out because we not only happen to be in the wrong place to observe it, but also in the wrong time.

That's pretty depressing IMO.


> At astronomical timescales even Earth's life is just a blink of an eye.

Age of the universe: ~13.8 billion years

Life on earth: ~4.1 billion years

That's a decent chunk, not the blink of an eye.


Yeah, it would be really fascinating if it turned out that life rare is in space, but not so rare in time.


> Age of the universe: ~13.8 billion years

If the age of the Universe is 13.8 billion years, I wonder, what was happening 13.9 billion years ago? What about 13.8^2 = 190 billion years ago and even before that ?

Kind of mind bending, but time seems to be infinite (at least to us) so at that scale, the life of Earth is really a blink of an eye or less.


As far as we know, time began 13.8 billion years ago. Talking about events before that is not meaningful.


Well, time is "synonymous" with change. So talking about events before "change" began is indeed not meaningful. But it is still meaningful to ask if there was change before the point where most people think there was no change.


Oh, it's more complicated. Talking about time gets weird when masses and energies are in the relativistic realm.


> As far as we know

What we 'know' is given to us by the mathematical models which we invented to describe the observed reality.

Currently, we have different models to describe reality based on the scale of observation.... Which is interesting because as you zoom into the microcosmos there's a point were the laws of physics magically 'switch' from relativity to quantum mechanics.

As a programmer who is not a physicist, I can smell a hack right there. Which brings me to the inevitable conclusion that maybe our model of reality is not entirely correct.

So if I were to bet on weather time existed before our model's mathematical moment zero or that our mathematical model is imprecise, I would definitely choose the infinity of time.

Our mathematical model is for us or future generations to adjust to whatever we discover in the future.


> Currently, we have different models to describe reality based on the scale of observation.... Which is interesting because as you zoom into the microcosmos there's a point were the laws of physics magically 'switch' from relativity to quantum mechanics.

There is no hack and there is no switch. QM and relativity are just abstractions at different scales over the same underlying principles. We do have a problem of stiching them together, but that's just us not having figured out the relevant math yet. But you're not free to notice a problem in one part of the model and then discard the parts that match the reality pretty well. Any updated, better, unified model invented will not differ in its predictions from QM at scales and precision levels where QM is applicable, because our present models are based on empirical observation. Transition from Newtonian to relativistic models didn't make apples fall up.

"Time" is also something we define and is valid only within that definition. For now, "time" starts with the universe, so the question of what was before is meaningless and will be, until we find a plausible way to extend the concept beyond the beginning of the universe.

The map and territory analogy is a really good thing. The territory doesn't change in any way when you make a better map.


Try as they might, physicists haven't found any way to adjust it that allows for an infinite time while also being consistent with so many observed phenomena. Of course we could make up an ad hoc model that says "It just is" but the big bang model is more consistent with observation than anything else we have.

We don't have freedom to just change it. We first have to find a possible valid way to change it and nobody has. Also nobody's found a way to fuel a car with water, fly on magic carpets or travel back in time to meet our grandparents. It might turn out to be impossible. Would you bet on those things happening one day? As another commenter said, your intuition is no match for nature - it's based on human's very narrow experience of the world, our inability to see how limited the concepts we can imagine are and our arbitrary biases for the ideas we like at the expense of ones we don't understand.


> As a programmer who is not a physicist

Let me stop you right there...


>What we 'know' is given to us by the mathematical models which we invented to describe the observed reality

So, do you have to propose anything better?

Mere intuition "but time seems infinite" is not that.


> So, do you have to propose anything better?

If I did, would anyone listen to it here ?

The point being that, throughout history humans always had a dominant theory about the origin of Universe and it was considered the truth at the time.

The bing bang theory is one such theory which is currently accepted as 'truth' by the scientific majority, but it is not the last and not the only theory explaining the origins of the Universe.

The man himself said it:

"We may now be near the end of the search for the ultimate laws of nature." - Stephen Hawking


> If I did, would anyone listen to it here ?

Maybe not here, but people would read a good paper on Arxiv.


Asking "what happened before the big bang?" isn't necessarily any more meaningful than asking "what's beyond the edge of the screen in a Pac-Man game?" or "what's the predecessor of a Garden of Eden configuration?"


Earth didn't exist 13 billion years ago. By most calculations Earth is only 4.5 billion years old.


Sorry, did you mix something up? I'm not even mentioning the age of the earth anywhere.


Not necessarily. For all we know, abiogenesis may have occurred many times on the early Earth, it's just that the lineage that lead to extant organisms is the only one that survived.


Nick Lane makes the argument that abiogenesis of our kind of life itself has very specific conditions involving warm alkaline thermal vents (not the hot super smokers, but the kind) and iron rich water oceans.

But he also points out that the development of eukaryotes, which are a class apart from prokaryotes (1000x the size and probably 1000x as complex) is even more of a mystery/fluke. And, again, looks to have happened just once in 4 billion.

Some series of unlikely events.


Never, really. When you've got ten to the forty something atoms jiggling up against each other on picosecond timescales, the difference between a rare event happening every million years and one that happens every billion years is pretty small.


I guess that point will be reached at a timescale when life can be bootstrapped in a lab.


It really makes me wonder if life started multiple times independently.

I guess certain molecules essential to life are right or left-handed, and if life started several times independently, we should (statistically) see both types of handedness. So then the question is: why didn't life start multiple times independently?


I guess if certain particles essential to matter as we know it are either particles or anti-particles, and if matter coalesced several times independently, we should see both types of matter ;)

The main difference between particles/antiparticles and chirality, though, is life tends to extinguish other life in its competition for replication. Antiparticles just tend to annihilate 1:1.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: