Join 3,558 readers in helping fund MetaFilter (Hide)


Researchers calculate that life began before Earth existed
April 18, 2013 9:29 AM   Subscribe

Geneticists have proposed that if the evolution of life follows Moore's Law, then it predates the existence of planet Earth.
posted by Confess, Fletch (92 comments total) 9 users marked this as a favorite

 
That might be the biggest "if" I've ever seen.
posted by 2bucksplus at 9:33 AM on April 18, 2013 [27 favorites]


Why the heck would evolution follow Moore's Law? As an evolutionary scientist, that idea makes absolutely no sense to me.
posted by Scientist at 9:33 AM on April 18, 2013 [41 favorites]


Oh my god, von Daniken was right!
posted by Pope Guilty at 9:37 AM on April 18, 2013 [5 favorites]


MIT Technology Review's arXiv blog has a slightly less credulous take on the paper.

Also, from the FPP article:

"The two researchers acknowledge their ideas are more of a 'thought exercise' than a theory proposal, but at the same time suggest their calculations ought to be taken seriously. "
posted by griphus at 9:37 AM on April 18, 2013 [1 favorite]


Life began before life began!
posted by KokuRyu at 9:38 AM on April 18, 2013 [1 favorite]


Hmmm....
posted by Artw at 9:39 AM on April 18, 2013


Life: It's just a cereal.
posted by jonmc at 9:39 AM on April 18, 2013 [2 favorites]


Life began before life began!

Who's using magical thinking now!
posted by OmieWise at 9:39 AM on April 18, 2013 [1 favorite]


This one's going to be being trotted out as "proof" either that evolution is bogus or that life comes from another planet or that God made us all in heaven before he made the world for decades to come. Sigh.
posted by yoink at 9:40 AM on April 18, 2013 [4 favorites]


Therefore by modus tollens, evolution of life does not follow Moore's Law.
posted by graymouser at 9:41 AM on April 18, 2013 [11 favorites]


Following Moore's law, the amount beer in my fridge should double very week.
posted by PareidoliaticBoy at 9:41 AM on April 18, 2013 [42 favorites]


I think it's going to be used more from the "prometheus was a documentary" angle.
posted by 2bucksplus at 9:41 AM on April 18, 2013


Wow, if you use a formula for growth from 50 years ago to measure a completely unrelated phenomena that predates the existence of our species, it provides an interesting result!

If Earth's atmosphere worked like my eyelids, it'd be night time right now! OH CRAP I CANT' SEE3
posted by Uther Bentrazor at 9:43 AM on April 18, 2013 [19 favorites]


This one's going to be being trotted out as "proof" either that evolution is bogus or that life comes from another planet or that God made us all in heaven before he made the world for decades to come.
Yep, expect to see this chart with the caption "Proof of irreducible complexity" at the next event sponsored by the Discovery Institute.
posted by Creosote at 9:43 AM on April 18, 2013


hmm... I don't think any natural process follows a straight geometric curve. They tend to be highly sigmoid.

Even Moore's law is ultimately sigmoid: it depends on what you are trying to optimize for. We've largely hit the limit on transistors per unit area, because that's not what we want to optimize for. Instead, we want operations per watt, which we are improving at a geometric rate.
posted by Freen at 9:44 AM on April 18, 2013 [2 favorites]


Clearly, life was seeded on the young Earth by unearthly intelligence and we are destined to manifest The Singularity and rejoin our predecessors among the stars. After all, the number of transistors crammed onto a chip eternally and necessarily doubles every 18 months, and that is the core metric of how life and the universe function.

All this has happened before and will happen again.

[plays Watchtower on the sitar like a boss]

posted by pleurodirous at 9:44 AM on April 18, 2013 [20 favorites]


Has anyone other than physicists commented on this? Because physicists do not have the greatest track record in other fields....

(Yes, the original authors are not physicists, but I never get tired of bringing up the story.)
posted by GenjiandProust at 9:45 AM on April 18, 2013 [1 favorite]


This follows their earlier work proposing: "If wishes were horses, beggars would ride."
posted by demiurge at 9:45 AM on April 18, 2013 [5 favorites]


I, for one, have no problem with thought experiments.
posted by stbalbach at 9:46 AM on April 18, 2013 [2 favorites]


How is this reported?

Moore's Law predicts life originated billions of years before Earth

A new study co-authored by a geneticist at the National Institutes of Health concludes that life originated elsewhere in the Universe around 9.8 billion years ago – roughly five-billion years before the Earth was even formed. But how did the study arrive at this conclusion, and does it make any sense?


The answer might suprise you!
posted by Artw at 9:47 AM on April 18, 2013 [5 favorites]


Then why aren't you helping the turtle, stbalbach?
posted by griphus at 9:47 AM on April 18, 2013 [9 favorites]


Does Moore's Law even tell us anything useful about the origin of computing?
posted by Artw at 9:48 AM on April 18, 2013 [5 favorites]


Geneticists have proposed that if the evolution of life follows Moore's Law

Let me go ahead and stop you right there...
posted by NotMyselfRightNow at 9:49 AM on April 18, 2013 [1 favorite]


That would only be true if life, like microchips or Doritos flavors, was created by an intelligent designer.

That chart is a perfect example of manipulating data to fit the hypothesis, what does that dot that says 'fish' mean? Do we have the sequence of fish from 500 Myr? Or are they saying that fish from today have that genome size and that fish appeared apx 500Myr?

The scale on that vertical axis is a log(10) function, so there's that too..
posted by kzin602 at 9:50 AM on April 18, 2013 [2 favorites]


What well known popular laws doesn't evolution abide by? I mean, simply by producing Hitler it conformed to Murphy's, Godwin's and Sod's Laws.
posted by yoink at 9:51 AM on April 18, 2013 [4 favorites]


I, for one, have no problem with thought experiments.

I am now imagining what will happen if Moore's Law is applied to Zeno's Paradox. We will never get anywhere faster and faster!
posted by GenjiandProust at 9:52 AM on April 18, 2013 [19 favorites]


Does Moore's Law even tell us anything useful about the origin of computing?

Moore's Law is applied to Zeno's Paradox

Hang on I think we just proved we're all living in a giant *EXHALE* computer.
posted by griphus at 9:54 AM on April 18, 2013 [1 favorite]


I love it when MetaFilter comments confirm my "Wait, what? What? Actually, what the fuck?" reaction to things. Makes me feel smart.
posted by MCMikeNamara at 9:54 AM on April 18, 2013 [4 favorites]


If evolution follows Murhpies Law there Earth would be full of mistakes and accidents.
posted by Samuel Farrow at 9:55 AM on April 18, 2013 [4 favorites]


from link: “Of course there are other possibilities to explain what happened, as the two acknowledge—life could have evolved following Moore's Law during certain periods but not at others...”

"They've done studies, you know. 60% of the time, it works every time."
posted by koeselitz at 9:55 AM on April 18, 2013 [2 favorites]


I have to say, though, that I'm surprised at the incredulity here. Moore's Law clearly does apply to evolution. No matter how far back you go, in every single life form that's ever existed (that we know about, anyway), the number of transistors on any integrated circuit has consistently doubled every two years. Hell, if you want anecdotal evidence, I've been doubling my number of transistors every two years since I was born, too, but I guess I'm just one guy.
posted by koeselitz at 10:01 AM on April 18, 2013 [22 favorites]


I've been doubling my number of transistors every two years since I was born, too, but I guess I'm just one guy.

Hey man, whatever turns you on.

Or off.
posted by yoink at 10:03 AM on April 18, 2013 [6 favorites]


Or Murphy Brown's law; an independent professional woman is sure to stir up controversy.
posted by 2bucksplus at 10:03 AM on April 18, 2013 [3 favorites]


Am I crazy? Did they do this extrapolation based on five data points? There are only two data points for the critical first three billion years of Earth's existence, and their curve undershoots the first and overshoots the second? Did they even try to fit other models?
posted by PercussivePaul at 10:06 AM on April 18, 2013 [2 favorites]


Don't tell me, let me guess... the lead author on this paper is Anne Elk, isn't it?
posted by gmh at 10:07 AM on April 18, 2013 [5 favorites]


An Elk...ahhhh!
posted by Confess, Fletch at 10:07 AM on April 18, 2013


I know everyone else is arguing against this, too, but no one has commented on that graph. It takes various clades of organisms, then plots their "functional genome size" against their estimated time of origin to make this prediction. That has the implicit assumption that they can arrange old to young clades in terms of this increasing complexity, which is estimated by functional genome size. This is so absurd! It's arbitrary; whichever clades you choose will totally dictate the slope of that line. You could get a flat line by showing that many recently derived bacterial lineages have the same functional genome size as ancient lineages, or a line with negative slope (predicting that genome size was infinite before the formation of earth!) if you contrasted the ancient bacterial lineages with newly derived endosymbionts (which have not existed for nearly as long as other bacteria, and can have super tiny genomes). So it's totally subjective - if you choose metazoans as the most derived clade, you get the slope they have; if you choose among many other equally valid options, you get a totally different prediction. So it's not just that this is wrong because it's silly to assume an exponential curve like Moore's Law applies to evolution, it's also wrong because, given that assumption, only one arbitrarily chosen set of data will arrive at this conclusion.
posted by Buckt at 10:07 AM on April 18, 2013 [15 favorites]


Um. It's sort of hard to believe, but I don't see the original study linked anywhere ... ?!

Here it is: Life before Earth (PDF)

Some interesting reading in there, at least to me.

Yep, expect to see this chart with the caption "Proof of irreducible complexity" at the next event sponsored by the Discovery Institute.

Yes, I'm sure the Discovery Institute is very interested in promoting new research about evolution. SNARF!

I have to say, though, that I'm surprised at the incredulity here.

For popular posts, the first 20-30 comments are often pure backlash. It usually takes about 50% of the way through the discussion to reach lucidity.
posted by mrgrimm at 10:10 AM on April 18, 2013 [4 favorites]


Back in my misspent youth as a biologist, there were too many open questions about what genome size means, how to measure it, and how to deal with weird outliers to make ideas about quantitative relationships between eukaryotes and prokaryotes stick. There's also the sticky problem that we really don't have a clear view of what prokaryote genomes looked like in the Archaic period. Another problem is that can we really compare genome sizes of modern prokaryotes and eukaryotes given that they have very different processes of replication?
posted by CBrachyrhynchos at 10:11 AM on April 18, 2013


"The two researchers acknowledge their ideas are more of a 'thought exercise' than a theory proposal, but at the same time suggest their calculations ought to be taken seriously. "
posted by Navelgazer at 10:11 AM on April 18, 2013 [2 favorites]


MIT Technology Review's arXiv blog has a slightly less credulous take on the paper.

"There’s no question that this is a controversial idea that will ruffle more than a few feathers amongst evolutionary theorists.

But it is also provocative, interesting and exciting. All the more reason to debate it in detail."
posted by mrgrimm at 10:12 AM on April 18, 2013


I think evolution follows a variation of Parkinson's Law:
Life expands so as to fill the time available for its completion.
posted by Kabanos at 10:12 AM on April 18, 2013


griphus: Also, from the FPP article:

"The two researchers acknowledge their ideas are more of a 'thought exercise' than a theory proposal, but at the same time suggest their calculations ought to be taken seriously as proof that they are attention-whores, unworthy of employment by any reputable university science department on earth. "
FTFTFA.
Artw: Does Moore's Law even tell us anything useful about the origin of computing?
Also a magnificent point...
posted by IAmBroom at 10:17 AM on April 18, 2013 [2 favorites]


I think evolution follows a variation of Parkinson's Law:

Or perhaps the Peter Principle--life rises to its level of incompetence.
posted by yoink at 10:17 AM on April 18, 2013 [2 favorites]


Those interested in mathematic criticism of the theory of evolution may want to take a look at Mathematical Challenges to the Neo-Darwinian Interpretation of Evolution.
posted by No Robots at 10:25 AM on April 18, 2013


I love how there are three metazoans, but only one prokaryote and one 'eukaryote'. Yeah, let's just ignore the plants, the fungi, and the rest of the enormous diversity within the eukaryotic protists and prokaryotes . Oh, and one data point can cover both the Archaea and Bacteria, right?

For a quick indicator of how stupid this methodology is, the human genome is thought to have 3.2 billion base pairs. The genome size of a fern is about 10 billion base pairs. Within fish alone, the size of the genome can vary from 0.3 billion base pairs in pufferfish to over a hundred billion base pairs in lungfish - and yeah, that's not functional size, but I'm pretty sure that varies significantly too.
posted by Mitrovarr at 10:25 AM on April 18, 2013 [4 favorites]


Reasons to break out the table salt without doing a PhD:

1. The paper is in the physics.gen-ph archive, which is a haven of (at best) far-fetched theories (see for yourself). The authors are geneticists, and this is a quantitative piece of research, so why is it not in the q-bio arxiv? Was it reclassified by the arXiv mods?

2. The paper appears to have been uploaded in Word format. That's fine, of course, but it offers a lower barrier to submission than LaTeX and instantly makes me wonder why. Of course, in their field Word may be a standard format for manuscript submission but, if so, why submit to a physics section of arXiv?

3. What exactly is the purpose of including Figure 2 (this)? I mean, what exactly does it add to the scientific value of the publication? You don't see cosmology papers with that sort of publicity image used as a figure. Maybe in a talk or seminar, but emphatically not in a paper.

4. One of the authors has published papers on "cosmology" in open-access journals of questionable veracity.

But I'm just a physicist, so I will defer to geneticists to criticise the science.
posted by pockupstep at 10:26 AM on April 18, 2013 [8 favorites]


Just wanted to pipe in to say that the figure shown in the link may serve to reinforce incorrect assumptions about evolutionary thought. Please remember that taxa at the tips of phylogenetic trees are all extant (living) and not a left to right list of "what has evolved from what". In other words, a "great chain of being" does not exist. While this figure is not a phylogeny, it plays right into the popular conception of a "great chain of being".
posted by cnanderson at 10:26 AM on April 18, 2013 [5 favorites]


I have just calculated that if I assume no meteor hit the Earth 65 million years ago, we are up to our eyeballs in velociraptors right now.
posted by Holy Zarquon's Singing Fish at 10:28 AM on April 18, 2013 [5 favorites]


Seems like this is the old and mostly discredited idea that evolution has an arrow of (increasing) complexity taken as an assumption within the specific context of genetic complexity. Like these geneticists are well aware of what genetic complexity is within the context of molecular biology, but not so aware of complexity within the context of contemporary evolutionary theory, so they without considering their presumptions just assumed that genetic complexity has always on average been increasing throughout evolution.

But I'm no biologist or evolutionary scientist, so I could be completely off-base.

Also what cnanderson wrote.
posted by Ivan Fyodorovich at 10:32 AM on April 18, 2013 [2 favorites]


griphus: Hang on I think we just proved we're all living in a giant *EXHALE* computer.

...that's currently trying to figure out what you get when you multiply 6 by 9?
posted by Greg_Ace at 10:37 AM on April 18, 2013


So life is stuck at 3 GHz?
posted by RobotVoodooPower at 10:39 AM on April 18, 2013


That's fine, of course, but it offers a lower barrier to submission than LaTeX

So you're suggesting that they're practicing unsafe scientific intercourse?
posted by yoink at 10:40 AM on April 18, 2013


Assuming that Moore's Law does apply to biological complexity, this would suggest that life began somewhere other than on Earth and migrated here.

posted by Splunge at 10:46 AM on April 18, 2013 [3 favorites]


If the evolution of life follows Moore's Law...

...then Spock will be alive in time for the next sequel!
posted by mazola at 10:50 AM on April 18, 2013


So life is stuck at 3 GHz?

It is quite amazing how CPU speed just kept going up and up until it hit that wall eh. Not really Moore's law though.
posted by Chuckles at 10:55 AM on April 18, 2013


[HOYLE WAS RIGHT]
posted by yerfatma at 11:12 AM on April 18, 2013 [2 favorites]


...then Spock will be alive in time for the next sequel!

True dat.
posted by spock at 11:17 AM on April 18, 2013 [1 favorite]


This makes sense because nucleotide base pairs have been getting smaller (and thus packed closer together) at a steady rate for the past 9 billion years or so.
posted by straight at 11:19 AM on April 18, 2013


Does Moore's Law even tell us anything useful about the origin of computing?

Well, if you extrapolate this back to 1 transistor, you wind up within a year of the invention of the silicon transistor. Which would be pretty astounding, except that's not when we started computing things electronically - ENIAC ran with vacuum tubes 10+ years before there were any silicon transistors.
posted by aubilenon at 11:21 AM on April 18, 2013 [2 favorites]


If I were a carpenter, and you were a lady...
posted by bongo_x at 11:26 AM on April 18, 2013


2. The paper appears to have been uploaded in Word format. That's fine, of course, but it offers a lower barrier to submission than LaTeX and instantly makes me wonder why. Of course, in their field Word may be a standard format for manuscript submission but, if so, why submit to a physics section of arXiv?

Most (if not all) biology-focused journals take Word documents as submissions. Even Bioinformatics, a computationally-focused life sciences journal, takes in Word docs. I'm not a physicist so I can't answer your second question, though I am aware that LaTeX is the standard there.
posted by Blazecock Pileon at 11:32 AM on April 18, 2013


If evolution follows Murhpies Law there Earth would be full of mistakes and accidents.

Oh, I like where you're going with this.

If evolution followed Boyle's Law, then lifeforms naturally spread around the available space in an even distribution, and if you reduce that space they all get warmer.

If evolution followed Sturgeon's Law then 90% of all species are crap. Hm, considering how many mutations are weeded out by natural selection over millions of years there might be something to that, actually 90% seems low.

If evolution followed Godwin's Law then HITLER ENDED THE WORLD.

I'm wondering if this isn't all somehow an example of Poe's Law: Without a winking smiley or other blatant display of humor, it is impossible to create a parody of Fundamentalism that SOMEONE won't mistake for the real thing.
posted by JHarris at 11:34 AM on April 18, 2013 [7 favorites]


And the law to really worry about here is the Crazification Principle, because no matter how much we tell them this is bunk about 27% of people who hear it will latch onto it as being SOMEHOW COSMICALLY IMPORTANT.
posted by JHarris at 11:45 AM on April 18, 2013 [1 favorite]


Does Moore's Law even tell us anything useful about the origin of computing?

Moore's Law tells us that in Charles Darwin's time, transistors were five miles on a side and computers took three million years to execute a single instruction. Before the late 1600s, transistors were larger than the observable universe!
posted by hattifattener at 12:05 PM on April 18, 2013 [18 favorites]


Punctuated equilibrium. The rate of evolution is non-linear.
posted by CheeseDigestsAll at 12:19 PM on April 18, 2013


non-exponential you mean
posted by aubilenon at 12:21 PM on April 18, 2013


While the ENCODE project has uncovered many functions for 80% of the noncoding DNA in humans (Pennisi, 2012), it has not yet addressed the C-value paradox. Thus we stick to the suggestion to measure genetic complexity by the length of functional and non-redundant DNA sequence rather than by total DNA length (Adami, et al., 2000; Sharov, 2006).

Interesting ideas. I might be biased, but I think I would question this as perhaps a weak point of the paper. The part of the ENCODE project I work on has shown that a significant portion of noncoding DNA (in humans) provides structural and sequence-based hints to how the coding portions are regulated, or turned on and off. Current work being done on comparing mouse and human regulatory elements shows that some repetitive elements (e.g., transposons) in orthologous regions are conserved to some extent and correlate with chromatin structure, which determines how different portions of the DNA open up to control specific genes.

We're in the early days of learning how the noncoding portions of genomes contribute to gene regulation, let alone how they have evolved, so leaving out those regions altogether makes me question the accuracy of the (to my mind, perhaps too-simplified) measurements of complexity the authors use to apply an exponential trend, which also perhaps use too few data points to reliably extrapolate out billions of years. We don't even know if our life chemistry is the only possible one — other chemistries could have different properties entirely.

Further, even if that trend of informational complexity works as the authors describe, then given the age of the universe there would be the Fermi paradox to contend with, where some other life form in the universe would have had an exponential "head start" on our form of life, for which we should have been able to see some evidence.
posted by Blazecock Pileon at 12:32 PM on April 18, 2013


How Pixar used Moore's Law to determine the production schedule for the first Toy Story movie.
posted by straight at 12:34 PM on April 18, 2013 [1 favorite]


IF evolution follows Godwin's law THEN at some point humanity would produce Adolf H-OH MY GOD YOU GUYS
posted by nathancaswell at 12:36 PM on April 18, 2013 [1 favorite]


If evolution followed Moore's Parkinsons's Peter's Zeno's Murphy's Boyle's Sturgeon's Godwin's Poe's Cole's Law, life on earth would be a side-order at Frische's.
 
posted by Herodios at 12:37 PM on April 18, 2013 [1 favorite]


This means the number of turtles at each level doubles as you go down forever, right?
posted by notme at 12:59 PM on April 18, 2013 [1 favorite]


Non-science: I was happy in Eden. Something happened in another part of the garden, and suddenly it started raining and I was in Oregon.
posted by Cranberry at 1:01 PM on April 18, 2013


This means the number of turtles at each level doubles as you go down forever, right?

For maximum stability, you want four turtles holding up each one higher up: one for each foot. But people come up a lot of with complex schemes to reduce that number. Systems like interlocking turtles or magnetic shells do allow a reduction in the branching factor. While this does keep initial costs down, reliability suffers as a result. To make matters worse, these schemes also make it much more difficult (and therefore costly) to replace a faulty turtle. So in the long run it's often cheaper just to use the extra turtles in the first place!
posted by aubilenon at 1:13 PM on April 18, 2013 [8 favorites]


Life began before life began!

Who's using magical thinking now!


This one's going to be being trotted out as "proof" either that evolution is bogus or that life comes from another planet or that God made us all in heaven before he made the world for decades to come. Sigh.


I'm no scientist so I get that people who know better are rubbishing the paper. But is the very idea that life is older than earth ridiculous in itself? Who knows what planets were smashed to smithereens in the billions of years before ours came along? Does it not kind of make sense that life (of some kind) has existed as long as the Universe has? No? Anyone?
posted by billiebee at 1:47 PM on April 18, 2013


Does it not kind of make sense that life (of some kind) has existed as long as the Universe has? No? Anyone?

It's perfectly possible that there is life in the universe older than planet Earth. What is hugely improbable (though not, it must be admitted, impossible) is that life on earth originated extra-terrestrially and subsequently colonized this planet.
posted by yoink at 1:54 PM on April 18, 2013 [1 favorite]


The idea isn't insane, but it does tend to be latched onto by fringe "scientists." And the idea of using a "law" that's really a rule-of-thumb for the progress of computing power since the 1940s to prove Chariots of the Gods is in fact bonkers.
posted by Holy Zarquon's Singing Fish at 1:55 PM on April 18, 2013


billiebee: But is the very idea that life is older than earth ridiculous in itself? Who knows what planets were smashed to smithereens in the billions of years before ours came along? Does it not kind of make sense that life (of some kind) has existed as long as the Universe has? No? Anyone?

It's not completely impossible, but it's extremely improbable. For life to colonize different solar systems, it would have to survive some kind of incident capable of launching it into space (a huge impact, probably) and then survive floating for millions of years in the interstellar void until it survived impact onto a suitable body. All of those are pretty unlikely, in particular survival in interstellar space. That would imply that life could survive frozen solid for tens of millions of years or more, accumulating radiation damage it couldn't possibly fix (since it would have to be frozen and would be unlikely to be metabolically active.)
posted by Mitrovarr at 2:00 PM on April 18, 2013


That would imply that life could survive frozen solid for tens of millions of years or more, accumulating radiation damage it couldn't possibly fix

Not only that, the functional genome size was increasing at an exponential rate that whole time.
posted by aubilenon at 2:20 PM on April 18, 2013 [2 favorites]


But stars are "alive " aren't they? I assume they're considered to be, if there's such a concept as "dying" stars. So life is extra-terrestrial, no? It's just the type of life found on Earth that people think can't have existed earlier. Is that it?
posted by billiebee at 2:32 PM on April 18, 2013


Moore's Law tells us that in Charles Darwin's time, transistors were five miles on a side and computers took three million years to execute a single instruction. Before the late 1600s, transistors were larger than the observable universe!

Which implies that the universe is only a few hundred years old, rather than a few billion.

Suck on that, science!
posted by It's Never Lurgi at 3:27 PM on April 18, 2013 [1 favorite]


It's not impossible for life to have originated before the formation of our solar system. The limits on this are defined by the lifespan of first-generation stars in our galaxy. Multiple lines of evidence suggest that the primordial gas of our galaxy was almost exclusively hydrogen with very tiny traces of helium and lithium, and you don't get much chemistry or planets out of that. There's a promising line of inquiry regarding exosolar organic chemistry, driven by the discovery of some surprisingly complex chemical signatures in places like planetary nebula. So it is a possibility.

However to explain my objection (as someone who dabbled in my misspent youth in monkeywrenching both prokaryotes and eukaryotes). The dominant taxonomy for life on Earth these days (at least from the 1980s to about a month ago), has three (or four) big groups:

eubacteria (almost all bacteria)
archaebacteria (a funky group of single-celled organisms separated by chemistry, not on the graph)
eukaryotes (animals, plants, fungi, and a bunch of single-celled stuff)

Viruses don't quite fit in anywhere, but they're parasitic and not especially relevant for this article.

The trend line depends on putting bacteria on the same trend as animals. And that (to me) is a big problem because bacteria and animals are different in many ways. (*) The structure of the DNA is different. The process for replicating and repairing DNA is different. The process for translating that DNA into proteins is different. Evolution has different tricks at the deepest level of genetic sequences across the two groups. It's like comparing current model bicycles and automobiles. Both are equally the products of advanced engineering and production technology, but with radically different constraints.

(*) Ignoring mitochondria and chloroplasts, who are bacterial symbiotes.
posted by CBrachyrhynchos at 4:30 PM on April 18, 2013 [3 favorites]


The insanity level of this paper doubles every 1.0 sections. By the time we've reached section 8, with gems like "What makes humans superior to mice is not the genome but mind.", the ideas are piling up like clowns in a Volkswagen. Then oxygen runs out and we're into stock markets and intelligent machines, as the "perishable mind" perishes.

Minus 1,000,000 points for calling bacteria "primitive organisms". Bacteria don't build up a trillion-cell shell around them for later disposal, bub, and they don't make mortgage payments either.

As for Figure 1, I love how "Eukaryotes" is a separate dot from worms, fish and mammals. "Better eukaryotes" is what they are, I guess.

Apart from that, I notice that the Figure 1 awesomeness is reproduced from one of the authors' earlier paper in Biology Direct, before that journal instituted their "reject obvious nonsense" policy. The great thing about Biology Direct is that you can read the peer reviews, which I highly recommend. You can feel the incredulity battling with the need for professionalism as you read the referee reports.
posted by aeolicus at 4:46 PM on April 18, 2013 [7 favorites]


It's a lot easier to explain the origin of life by pushing it somewhere "out there". In the end, "panspermia" is the same as "turtles all the way down". Luckily, serious professionals aren't stooping to such facile answers yet.
posted by Twang at 4:50 PM on April 18, 2013


"But stars are 'alive' aren't they? I assume they're considered to be, if there's such a concept as 'dying' stars. So life is extra-terrestrial, no? It's just the type of life found on Earth that people think can't have existed earlier. Is that it?"

If you're asking this seriously, because at the moment I'm honestly uncertain, then that's just an idiom and not technical, like saying that a fire "died".

It's a convenient idiom because stars have pretty well-defined progressions from when they're "born" (from either primordial, early universe hydrogen and helium clumping together under gravity, or hydrogen and helium and a few other elements that are remnants of a supernova similarly clumping together in a nebula or by the shock waves of another supernova) to when they "die" (variously, from just smoldering and cooling indefinitely, to shrinking down to a relatively hot mass of highly compressed "degenerate" matter that also slowly cools, to spectacular supernovae). Relative to how much astrophysics we don't know, stellar evolution is really very well-understood and regular. So you think in terms of stars being born and living and dying. But it's nothing like life.

See: Main Sequence.

As to your question, coincidentally one of the scientists who made some of the biggest contributions to stellar physics was Fred Hoyle who was notoriously a proponent of panspermia. That is, the idea that the earliest live that ever existed on Earth, early microbial life, actually somehow arrived at Earth from somewhere else, most likely in a cometary impact, or in a meteorite that originated as a piece of a neighboring planet (read: Mars) accelerated to escape velocity by some really big impact and came down here, or even more speculatively perhaps somehow just from interstellar space from nebulae or the like where it originally evolved.

All of these things are pretty unlikely as best we can guess and, as Twang wrote, it seems to be an unneeded answer to an invented question that doesn't actually accomplish anything because then we would just have to answer the question of how it originated over thataway instead of here. We don't really have a satisfying theory for the origination of life here; but it sure seems that elsewhere would be even more hard to explain, so what have we gained?

That said, I think the previous commenters have slightly overstated the fringeness of panspermia these days. It's still very heterodox and not much taken seriously, but back in Hoyle's day he was roundly mocked for it in pretty much precisely the same way that everyone was mocked for even supposing that life was likely to exist anywhere else in the universe, or even that there were other planets like Earth. Those were fringe ideas until the last couple of decades.

But as the previous presumed terran exceptionalism has been soundly destroyed by the rapidly accelerating discovery of extrasolar planets, and by what we've discovered on Mars (no life yet, but a lot of evidence that it had the raw materials for it), the part of the skepticism of panspermia that rested upon that exceptionalism has faded away. So now it's just, you know, unlikely and not really taken very seriously by anyone but not totally nutso like it was. Mars was in its most hospitable phase a billion years before the Earth even stopped being mostly lava. If life did evolve on Mars, then because of that very large head start and all that time passing (and especially during the earlier era of the solar system when there was more junk flying around), it's pretty conceivable that life could have originated on Mars and then been exported to Earth via an impact.

So, not kooky-nutty anymore, but still far outside the consensus. Personally, I think we're going to find elementary microbial life in numerous places in our solar system and then have to assume that rudimentary life is very common — which would change the analysis quite a bit, even taking into consideration what Mitrovarr wrote about the implausibiity of any sort of viability after such a journey. Then you'd get into a inverted situation, more like linguistics, where you'd have to justify life spontaneously originating over and over again when given the timescales involved migration from just a few or one origination arguably makes more sense.

I don't mean to be endorsing panspermia, only arguing that it's not quite a nutty as it once was thought to be. And, all in all, my objections to this paper are more about everything that's not the implicit panspermia stuff.
posted by Ivan Fyodorovich at 7:15 PM on April 18, 2013 [2 favorites]


After skimming the paper, I suspect it's a stealth intelligent design work to get something in print that ID advocates can point at and say, "look, the timelines don't add up, THEREFORE GOD!"
posted by CBrachyrhynchos at 8:51 PM on April 18, 2013 [4 favorites]


Or more appropriately, "See! There is a debate in print about the mathematical possibilities of evolution based on shoddy statistics possibly involving improperly collected samples! Why won't you let us teach the debate!"
posted by CBrachyrhynchos at 9:02 PM on April 18, 2013 [1 favorite]


Hey, why aren't the number of comments in this thread doubling every second? Stupid piece of shit Moore's Law! Why don't you work the way I want you to?
posted by duffell at 7:20 AM on April 19, 2013 [1 favorite]


Ivan Fyodorovich - I was asking seriously because I'm fascinated but clueless about this stuff! So thanks for the explanation.
posted by billiebee at 9:43 AM on April 19, 2013


Metafilter: "Wait, what? What? Actually, what the fuck?"

Never change.
posted by glasseyes at 6:34 PM on April 19, 2013 [1 favorite]


Massimo Pigliucci has written a critique of Sharov and Gordon's paper.
posted by crLLC at 11:11 AM on May 6, 2013


Yean, Pigliucci's critique of the core claim centers upon the two main criticisms discussed in this thread: that the authors' assumption of increasing complexity is inherently mistaken; and that even if it weren't, they've cherry-picked their data in a hugely distorting fashion.

I didn't realize that there was all that other dubious stuff piled on top, either. The paper is like what you get when you cross a couple of post-docs with a couple of stoned college sophomores.
posted by Ivan Fyodorovich at 5:35 PM on May 6, 2013


« Older Caitlin Freeman started out as a photographer, but...  |  31 percent of employers involv... Newer »


This thread has been archived and is closed to new comments