Wesson argues that even if the actual microbes are dead on arrival, the information they carry could allow life to rise from the charred remains, an idea he calls necropanspermia.
“The vast majority of organisms reach a new home in the Milky Way in a technically dead state,” Wesson wrote. “Resurrection may, however, be possible.”
The key lies in how much genetic information survives the trip, Wesson says. An organism’s genetic information is encoded in the sequence of nucleotides in their DNA. This information can be measured in bits in the same way as computer processes. Bacteria like E. coli, for example, carry about 6 million bits of information in their DNA.
Random chemical processes couldn’t produce enough information to run even a simple cell. Over 500 million years, random molecular shuffling would produce only 194 bits of information, Wesson says.
One possible way around this paradox is the idea that life on Earth was seeded by biological molecules that already had a large information content that survived the journey even though the molecules themselves were killed.
Wesson is a bit fuzzy on how that information would translate to new, healthy living things.
Today's vocabulary word: Necropanspermia.
All Life on Earth Could Have Come From Alien Zombies
24 Responses:
As soon as he says "random chemical processes couldn't produce enough information" we know he's out of his mind. Finishing the sentence with "over 500 million years, random molecular shuffling would produce only 194 bits of information" shows that he's not just crazy, he's really, really bad at math.
But the word necropanspermia is quite awesome, even if the dude is an innumerate loon.
An innumerate zombie-loving loon.
Because adding zombies always makes things more sciencey.
See below. He's not bad at math; he didn't do the math, nor does he claim to have. "He" is Lisa Grossman, the author of the Wired article. Her article is a popular gloss of an article by Paul Wesson, and Wesson cites an earlier article by Edward Argyle. Argyle is in fact quite good at math; Wesson uses his figures with little comment; and Grossman published it without comment.
Grossman is bad at something, but she's bad at being a critical reader. Fortunately for her, that's not her job.
He" is Lisa Grossman, the author of the Wired article.
Not quite. See this bit: "Over 500 million years, random molecular shuffling would produce only 194 bits of information, Wesson says."
Which is to say, she's quoting Wesson reciting that tidbit. It doesn't matter if it's unique to Wesson or if *he* got it somewhere else and uncritically repeated it, it's simply a fucking stupid statement.
I'm curious why you think that statement is stupid — and whether you mean Grossman's statemnt, Wesson's, or Argyle's. I haven't literally run Argyle's numbers by hand, but his (clearly-stated) assumptions and his equations are both reasonable.
I'm afraid I don't have access to the full paper and have no desire to pay $30 for it, so I can't see his assumptions or his equations.
I'm saying that any assumptions that lead you to conclude that 500 million years of *random* behaviour will result in only 194 unique 1/0 changes in "information" are either transparently bad assumptions, or you've got a really lousy definition of "bits of information". And you've also apparently got a WAY low number of tests, at that point.
I'm saying that the assumption that the behaviour is random is, itself, a transparently bad one. It's the "tornado in a junkyard assembles a 747" argument, and, having only the abstract and the first page, the only *smart* thing I know Argyle actually said was his conclusion that all his assumptions must be fucked up because his math seemed good-given-assumptions but his results didn't make sense.
Ok, it wasn't clear whether you had read the paper or not. Argyle is quite clear about what he's doing: getting an optimistic estimate of the complexity of a polypeptide that might arise randomly, without any evolutionary forces. If that polypeptide itself could reproduce itself, now you have a building block that could reasonably be at the mercy of evolution.
Argyle's model is a toy one: 1043 amino acid molecules floating around in the prebiotic ocean, combined in chains of average length 10 for 1042 chains. These break and recombine roughly every 10ms.
(1042 peptides) x (5 x 108 years) x (3 x 107 seconds/year) x (100 reactions/peptide/second) = 1.5 x 1060 possible results.
Once you remove the redundancy, you're down to 2.5 x 1058 different polypeptide molecules in that half-billion years. Taking the base-2 log of that number gives 194 bits, for a very optimistic estimate of the polypeptide complexity that might result in the soup.
It's not a totally crazy back-of-the-envelope calculation. And given what he's trying to prove, it's appropriate. His point is that you can't simply jump from a chemical brew to the genome of even the simplest single-cell organism. You need self-replicating molecules first. That's his point.
That actually makes much more sense. And the summary used in Wired *still* makes my teeth itch because it's phrased in such a misleading way - dammit, the entire point was that this was evidence that the process *was not* simply random, and that there was selection involved: self-replicators are favoured, then more efficient self-replicators are favoured, then self-replicators that build structures, then self-replicators that work together, and then you have a cell.
... or you've got a really lousy definition of "bits of information"
He has exactly what you need when dealing with information: a measure of what's signal, and what's noise. There are far more than 194 bits of total information in any chemical system. Much of that is what we in the business like to call "heat". He has a very, very narrow definition of what he's looking for: the nucleic acid sequence information that would be needed to encode polypeptides.
That 194-bits number is entirely out of context.
Another option — the one proposed by Edward Argyle, the man who came up with that 194-bits number — is evolution. In Argyle's words, "Darwinian evolution."
I should be clear: I've now read the Wired articl, the section of Wesson's paper where he invokes that number, and the relevant parts of Argyle's 1977 paper, "Chance and the Origin of Life", in which he derives the number.
Wired is off the deep end: they suggest that the 194-bit limit requires some sort of panspermia. Wesson is more conservative, if still a little disingenuous; he himself states that panspermia is one option:
Argyle himself is entirely reasonable. He derives the number of 194 bits in section 4 of his paper, and then goes on, in sections 6 through 8, to demonstrate how you get around that limit. He demonstrates that real, living organisms like e. Coli generate information much faster than this; he describes how they go about it; and he concludes that the missing piece is actually a picture of what life looked like in between amino chains and e. Coli.
But most importantly: I have to congratulate Wired for providing an honest-to-god LINK to the scientific article. That's the first time I've seen any lay publication do that.
Argyle's paper suffers from being 30 years old - his assumptions regarding the nature of "random" and his definition of "life" are.... out of date. But you're right, he has one stupid statement ("only 194 bits!") and proceeds from there to conclude that this same statement must be wrong.
Panspermia (either necrotic or radiant types) would be neat, but it doesn't actually solve the problem of how life got started. I've never understood why it gets so much attention. Okay, I do, LIFE FROM SPACE, but still.
it beats the answer "xenu" ;-)
1) It gives you a lot more time and planets to work with, so if you believe that life is so complicated that it's unlikely to have arisen in the time available to it on this planet, it panspermia gives you the exponents you need.
2) Vulcan-Human hybrids.
I didn't quite follow all of latemodel's arguments of who was right about what, but I kinda got the rough math of what might happen to polypeptides in a primordial soup.
So "time and planets" got me thinking in reverse: anyone done the rough math for panspermia (of more than 194 bits) across time and space of the universe? Is this universe even old enough for a unit chance of earth getting, um, seeded? Actually the limiting factor might be the age of the earth, or perhaps even the window when seeding was posited, say a 1 billion year window 2-3 billion years ago. Also, where do those seeds come from, how did they form, and how did they cross space?
I think the answers to all these questions will resemble something like the primordial oceans of Earth.
I think the Drake Equation is already complex enough and depressing enough and you are just making it worse.
Thanks, dude.
Good news, everyone! According to Claudio Maccone, civilizations probably average around 2,000 light years apart at this point, while habitable planets are likely around 60 light years apart on average.
And on a personal note, I don't care nearly as much about where life came from as where it's headed.
Ok. But isn't this all just punting? I mean, even if zombie aliens _were_ present at biogenesis, they themselves had an origin. And wouldn't their existence suffer from the same improbabilities?
It's turtles all the way down.
194-bit turtles.
Just when I thought the term "panspermia" couldn't get more lurid.