Abiogenesis and Arguments From Ignorance


I used to be much impressed with Dr. Douglas Axe’s infamous “10^77” argument. If one accepted that figure, as well as the propositions that the first organism must have been a cell, and that a minimal cell requires 250 different proteins, then it’s easy to show that life is vanishingly improbable. However, I now regard all of the above as gratuitous assertions at best, and I have been persuaded that the 1 in 10^77 figure is bad science. Hence I would now say that the boot is on the other foot: it’s incumbent on design proponents to show that abiogenesis is astronomically unlikely.

Let me reiterate that my change of opinion has nothing to do with theology. For me, it’s the math that counts. Cheers.


I disagree that Dougs numbers are bad math as it depends on the application but for arguments sake let’s use the more conservative numbers of Szostak of 1/10^11.

Now build me a living cell from random chance. The first issue is minimum capability for sustaining life and self replication so you need several say 400 well functioning genes for this.

Now take Szostak’s numbers and go from this to a Eukaryotic cell. This is maybe an even more difficult problem for random mutation and natural selection.

The problem here is to get successful self replication the chickens and eggs all need to be available at once. Can you sustain life without rapid amino acid and ATP production? Can you sustain life without DNA repair?

This is a monster challenge so I think the default position is it is highly improbable for a random mechanism to have accomplished this.

I agree with you that this is not a theological issue but it is a very large problem for science.

Perhaps that’s true but the math is sufficiently ambiguous here. If we allow for Gods action (which science does not) I think the most rational conclusion is rational and rigorous agnosticism.

1 Like

That’s the wrong way of looking at it.

We know that there is randomness involved in snowflake formation. We can see that from the large variety of snowflake shapes. Yet nobody would describe that as “building a snowflake from random chance.”

Hi @colewd,

Here’s the problem if you accept Szostak’s 1 in 10^11 figure, and your estimate of 400 genes for a minimal cell. At first blush, you might think: 1 in (10^11)^400 is 1 in 10^4,400, which is well below Dembski’s Universal Probability Bound. But during the history of the primordial Earth, it is estimated that there may have been up to 10^43 amino acid sequences explored (to the nearest order of magnitude), according to Dryden, Thomson and White (2008). If 1 in 10^11 yields something functional, then we have 10^32 functional molecules floating around on the primordial Earth. Getting 400 of these to hook up in a way that supports life and self-replication doesn’t sound so impossible now, does it?

As a Christian, you are familiar with the idea of God’s providence. How does God achieve it? He definitely does not always do it ‘supernaturally’, overriding natural law that he has established. Scripture is replete with examples, as is recorded history in the lives of multiple hundreds of Christians since, and in mine. I am reminded of one involving George Müeller, founder of many orphanages in England, when they sat down at breakfast to empty plates, not knowing where the food was going to come from, and gave thanks anyway. God provided by design. So probabilities are irrelevant, but you can still believe in design, as do I.

1 Like

Hi Bill,

A couple of notes to help you better understand the criticisms that bounce around here and elsewhere.

First, it is not necessarily Axe’s math that is bad (although his approach to analyzing his data is bereft of what most would consider to be a suitable statistical component), but that his experiment in and of itself does not allow for the conclusions (and the numbers) Axe reaches.

Second, as far as Szostak’s value of 1/10^11, remember that this is the ratio of functional to all sequences. It is not a probability. To get a probability, one needs to consider the number of sequences in a population (pool, setting whatever). For example, in a liter of prebiotic solution that has a molar concentration of peptides of 5 picomoles (that is 5x10^-12 moles, by most peoples’ estimations a pretty low concentration) per liter, the probability of finding a functional sequence would be essentially 1. Which would also be the probability of finding 400 different functional sequences, given Szostak’s ratio.

Let us know if this is not clear.


The forum software told me that there were 6 unread posts in this thread.

So I clicked on it, to read those posts. And 5 of the 6 posts were hidden (flagged by the community).

Note that I’m not objecting to this. I was able to read those 5 flagged posts. And, in all honesty, they did not add anything useful to the discussion.

Perhaps we can do better than this.


How do you get them to hook up if there’s no enzymes or DNA synthesizers to hook them up with?

by design, naturally, not breaking any natural laws, but with supernatural timing and placing. That cannot be proven by science, but events and circumstances are infused with meaning, again, by design, and that could certainly be true in the biological realm, as well. “There is a difference between a set of circumstances that is improbable, and a set of circumstances that is improbable and also performs a function.” @terrellclemmons

1 Like

Try getting two of them to hook up. Let’s say that the chance is 1 in 10 per 1 second per 1cm^3, and this chance scales linearly with the number of molecules that need to work together. And furthermore it all works without any enzymes. Let’s be generous.
Then the volume of Earths oceans are 1.332 billion cubic kilometers which is 1.332e+24 cm^3
The chance of 400 of them hooking up is 1 in 1e+400
Divide the two and you get a chance of 1 in 1e+376
Divide with the number of seconds in a year 3.154e+7
And you get a chance of 1 in 3,17e+365 that’s 365 zeroes
Divide that with the number of planets in the observable universe 10^24 (give or take)
And you would need to wait around 3,17e+341 years to see the process you imagined would come to fruition and produce life. To put it in context that’s 2.3e+331 times the current age of the universe.

If abiogenesis is possible, it HAD to be more efficient than that.

Sounds like an argument from pure ignorance.

Or he got tried of dinosaurs.

That’s not the way macromolecules work in concert.

1 Like

actually the chance of finding a functional gene (with a complex function) is near zero. this is because we need a functional code to be exist in the first place before evolution\abiogenesis can work. genes for wings for instance are clearly the product of design. so we need to calculate not only the chance to find it in the sequence space but also the chance to find it at all.

  1. How do you get macromoleclueless.
  2. How do you get the right macromolecules.
  3. How do you get them to work in concert.

That’s been tried and found wanting. :sunglasses:

Nope, you were not mistaken. That was his clear and frequently, stridently emphasized point. His message was not only that they haven’t got a clue, but also, some people are lying to you when they try to make you think they do have a clue.

Those are my paraphrases, but that was his message. Loud. And. Clear.

1 Like

Yea, he kept the birds.

1 Like

I love the Freudian slip. I tip my cap your way…

Simple answer - chemistry. I believe others have given many pointers that elaborate, and I am too lazy to add anything here other than to thank the other contributors.

As I just explained above, in a situation with even a minimal amount of chemistry going on, if macromolecules are being made, then even dilute solutions will have all of the “right” ones.

Most polypeptides will have inherent tendencies to bind (if weakly) with others. RNAs even moreso. The limitations here would be binding constants, which may range over many orders of magnitude, probably towards the parts of the scale that would not favor efficient binding. But the occasional high-affinity interaction would not be so affected, as it were.

These are all general considerations, mentioned to point out that there are no inherent, general, or universal aspects of chemistry that may be invoked to rule out, a priori, a chemical route to the origin of life. If one wants to argue thusly, one must first know what the path was, and which steps in particular (I repeat - known, confirmed, accepted steps) are precluded by the rules of chemistry.

1 Like