Blog Archive

Wednesday, June 3, 2009

Probability that life could evolve from a single cell to us is zero

From pg. 153 of Who was Adam?

While paleoanthropologists struggle to discern evolutionary connections among the hominds in the fossil record, astrophysicists (attempting to quantify the likelihood that intelligent life exists in the universe beyond Earth) have identified another series of problems for human evolution. According to various calculations based on physical conditions, it is extremely improbable that modern humans evolved from baceteria through natural means, given the brief time window of Earth's habitability.

In one study, astrophysicists John Barrow, Brandon Carter, and Frank Tipler comment on the surprisingly large number of highly improbable steps in the supposed natural evolution of an intelligent species on Earth. Moreover, the number of such steps merely represents a lower limit; evolutionary biology has not yet advanced sufficiently to determine their actual number. Restricting the count to just the known problem steps (which are statistically independent) in the evolution of Homo sapien sapiens, the trio produced a probability figure for the emergence of humans from a suite of bacterial species in 10 billion years or less: 10^-24,000,000. (In other words, a decimal point 24 million places to the left of the 1.)

An independent calculation done by evolutionary biologist Francisco Ayala places the probablility for humans arising from single-celled organisms at 10^-10,000,000. As Ayala and others pointed out, animals on ancient Earth did not know they were supposed to evolve in such a way that human beings could later appear. Natural selection operates only during an animal's lifetime. It cannot select a portion of a genome with the intent of using that genome protion 1, 2, or 3 billion years later.

To put the calculated probabilities for humans arising from single-celled organisms into perspective, if every proton and neutron in the universe were a planet, and if each of these planets contained as many single-celled organisms as Earth does today (a trillion quadrillion single-celled organisms ) the probability that humans could have arisen once in the universe would be 10^-999,921, according to Ayala's calculations. According to Barrow, Carter, and Tipler's calculation the number would be 10^-23,999,921.

Such incredibly tiny probabilities warrant the conclusion that, from a natural perspective, no physical intelligent life should exist at all --anywhere in the universe.

Brandon Carter, "The Anthropic Principle and Its Implications for Biological Evolution," Philosophical Transactions of the Royal Astronomical Society A 370 (1983): 347-360
John D. Barrow and Frank J. Tipler, The Anthropic Cosmological Principle (New York: Oxford University Press, 1986), 510-573 and 557-566

Frank J. Tipler in "Intelligent Life in Cosmology," International Journal of Astrobiology 2 (2003): 142


Steven J. said...


First, best wishes for your interview.

Second, thank you for going to the trouble of typing all that out for me; I understand that it's a pain.

Third, though, this leaves me only marginally less confused.

Fourth, I apologize for the length of my reply.

Here, the independent calculation done by evolutionary biologist Francisco Ayala seems to be dealing with the probability of "humans" -- not "something like humans," not "technological intelligence" -- evolving. The odds against that, he calculates, are incredibly remote. But then, presumably, we're interested in the odds of any sort of technological intelligence (presumably, the "image of God" doesn't require hair, or three bones in the middle ear, or a vermiform appendix): regardless of number of limbs, or whether it has fur, feathers, scales, or something else entirely. If Ayala was really just calculating the odds of getting one particular species, his results do not address this broader question (whether any sort of physical intelligent life at all should exist anywhere in the universe).

Again, outside of articles written for RtB or citing RtB authors, searching for work by "John Barrow, Brandon Carter, and Frank Tipler" turn up lots of articles on the Anthropic Principle. Such articles concern the odds of having the particular natural laws we do in the universe, the odds of finding a planet suitable for life, etc. more than the odds of abiogenesis itself once such conditions are met, or the odds of evolving bacteria or humans once life exists.

For my own part, while I hesitate to take on such eminent cosmologists in their own field, I don't see how one can get the data needed to run meaningful calculations on such issues. Take the simple and horribly relevant questions: is the universe finite or infinite? is there one set of physical constants throughout the universe, or are their cosmic domains beyond this Hubble volume with different physical constants (i.e. is this a universe or multiverse?)? I'm pretty sure no one knows.

On the grounds of parsimony, one might prefer just assuming that the visible universe is what exists, but then, if one is invoking parsimony, is an infinite universe, or an infinite variety of space-time domains, less parsimonious than a Designer capable of fine-tuning entire universes, or some mysterious unknown cosmic force that requires the universe to evolve life? The answer is not obvious to me.

We have too small a sample of ecologies and types of life to be sure in just what sorts of environments life might be possible; if we ask "how likely are the conditions to which life on Earth is adapted," we're not necessarily asking the right questions (any more than, say, intelligent polar bears would be right to assume that bears can't exist in California because there are no ice floes there). Just because life as we know it can't exist and presumably couldn't evolve in most of the universe, I'm not sure that means that no possible life forms could do so. We don't know what sorts of life forms are possible, except for those we've actually observed, which presumably do not exhaust the possibilities.

I think -- I'm not sure, and I might be wrong -- that Ross's argument here is like looking at an arrow sticking out of the broad side of a barn, and calculating the odds of the arrow hitting that exact spot, and then assuming that there's no difference between those odds, and the odds of hitting the side of the barn at all (where the side of the barn represents life of any kind, or technological intelligence of any kind).

Mintz said...

Why is it that creationists have such a hard time understanding probability?

The probability that you or I will win the very next lottery draw (assuming we bought a ticket of course) is vanishingly small. The odds of winning are something like 1 in 14 billion for the UK lottery. Yet somebody wins nearly every week, and if we retrospectively look at the odds that this individual would have won in this particular week with this particular ticket, the odds are still about 1 in 14 billion.

The fact that the probability of something occurring in a single instance is tiny does not mean that it is impossible, and if the event is repeated often enough the chance approach certainty.

Continuing with the same analogy - The chances of the same lottery numbers coming up two weeks in a row is the same 1 in 14 billion quoted earlier, and its the same no matter how many times we repeat it. The chances of a winning draw bearing the exact same numbers as the previous week will always remain 1 in 14 billion. However, but if we make 14 billion draws (at 1 per week that would take approximately 270 million years) the chance approaches 1:1 that during those 14 billion years two consecutive weekly draws would have produced the same numbers.

Please learn some basic math.

Mintz said...

I actually need to correct my own maths in the previous post.

THe probability of winning the UK lotters is 1 in 14 million not billion, and consequently, the figure I should have subsequently used is 270,000 yrs.

Froggie said...

What are the odds of there being a God?

Hi Vera!

Hey, could you link to the article where the word for day in Genesis is discussed and thought to mean age rather than day?

See Ya!

verandoug said...

Sorry, once again, for dropping the ball here. I worked diligently to pass the ACLS. And did. I had to learn about 10 ECG anomalies to pass and what to do in an emergency. It was not easy.
More later,

verandoug said...


This is the link.

Hope that helps.

Can you give me some advice? I am been struggling getting a job. Everyone loves me. I pass the initials with flying colors. But then I go into these interviews and fail. Nobody is asking me spiritual questions like, "What do you do when you're between a rock and a hard place?" But they do ask ridiculous questions like, "Tell me how you would make a peanut butter and jelly sandwich in complete detail." It's like, how do you pass something like that? Do you know of a web site where they have some good advice on this kind of thing?


verandoug said...


One more question, can they get personal details about me from the Internet? Would you do that? Can my credit report, children's past experiences, and everything I've ever written on the Internet be available to a prospective employer as part of my background check? Can they find this blog easily?


Steve said...

Uh, Mintz, there has only been 10 to the 18th power seconds since the Universe began. Compare that to the 1 in 10 to the 24 millionth power event you need. Duh.