Behe and Hunt: Irreducible Complexity and Numerology

Here are two to focus on,

  1. The first is the Irreducible Complexity argument, which for some definitions does appear to be falsified. There are several distinct claims being mashed up into one, and the only one capable of definitely untangling this is Behe himself. I want to know honestly the current status of this argument is from Behe’s point of view, and form DI, because at least some forms of the argument appear directly falsified. Which Irreducible Complexity Argument?

  2. This paper ( is put forward as example of CSI (ala Dembski) and is essential numerology. @evograd has linked to some clear rebuttals of this, and it has not been picked up by any of the leaders. Though, we still see it positively referenced at ENV: Do they stand by this paper? Do they agree it is an example of CSI? If not, when will they point out and clarify its problems?

I’d start with those two examples.

@colewd, that might be true, but it is also true that the two of you sorting it out is not likely going to be productive. The goal of this forum is to understand and be understood, not litigate other people’s arguments. If you want to see those arguments to make progress, let’s see some of the principles clarify these points.


I will start with your first comment on irreducibly complexity. I have looked at what you have and emailed my understanding to Mike and expect a response in the next couple of days.

This claim was targeted at the Darwinian mechanism of random mutation and natural selection and proposed as a challenge to that mechanism.

I do believe that you do have the definition of an irreducibly complex system right.

I will respond when he gets back to me.

On number 2 I think the more interesting is the work that Giuseppe Puccio is posting at UD on indirect measurements of functional information as he is using empirical observations of DNA and Protein sequences. Their have been many back and forth discussions between UD and TSZ where Joe Felsenstein was very involved and wrote and op challenging Puccio’s work.


I will look forward to hearing from Behe. He is a nice guy, and I have no ill will to him. Remember also that @art had been wanting him to engage on T-urf23 too, which appears to be a very complex enzyme arising out of non-coding DNA.

I’ve covered in that past and will again. I even have a preprint on it in biorxiv. The real issue, however, is…

This case is a clearly bad argument, or at least it seems that way. If it isn’t they might try and defend it.

If they can’t retract or defend this paper after that glowing review in ENV, how can we trust they will ever retract a bad argument?

1 Like

Just to expand on this, from @evograd:

This one is interesting. It purports to show a “wow signal” in the standard genetic code – the code that determines which codons are translated into which amino acid. The authors claim that they have detected a signal (mathematical pattern) in the genetic code that can’t be explained naturally, and therefore points to the genetic code being intelligently designed by ET. There are a myriad of problems with this, the key one being that the analysis boils down to numerology. PZ Myers wrote a blog post on this paper shortly after it was published, which I would recommend everyone read. I’ll be paraphrasing him and some other commentators. Numerology is the practice of juggling arbitrary numbers around in an arbitrary fashion in order to find some of kind interesting patterns, which are then proclaimed to be somehow meaningful.

If you ignore the snark, PZ Meyer’s blog very aptly explains how this study is essentially numerology:

Here’s the short summary of what they do: they jigger the identities of the amino acids coded for by each codon into a number, a nucleon sum . What is that, you might ask? It’s determined by adding up the number of protons and neutrons in the amino acid, which is simply the mass number of the compound. Further, you can distinguish the amino acid into it’s R group, and the atoms that make up the peptide chain proper, which he calls the B group, for standard block. The mass number of the B group is always 74, except for proline, so he transfers a hydrogen from the R group to the proline B group to bring it up to 74, and by the way, did you notice that 74 is two times 37, which is a prime number? Now if you take all the three-digit decimals with identical digits (111, 222, 333…999), and sum their digits (111=3, 222=6, 333=9, etc.) you get the quotient of the number divided by… 37 !!!1!!

Are you impressed yet? This is simply numerology, juggling highly derived quantities that have little to do with functional properties of the molecules to come up with arbitrary numerical relationships, and then claiming that they’re somehow significant.

This, also, is listed as one of the ID papers published in peer-reviewed journals. Unfortunately for their larger story, they identify this numerology paper with Demski’s CSI metric.


3 posts were split to a new topic: Who is Responsible for Public Ideas?

Mike got back to me on both cases. Regarding irreducible complexity he agrees that the definition that I lifted from your reference from Bio Logos is correct.

He said to me he is not attempting to prove a negative “so evolution cannot work” is not his claim. Quoting him:

He also said he is going to expand his claims of IC in his next book that I think is due in February of next year.


The other problem is that scientific theories shouldn’t be based on negative arguments. I don’t know of any theory in science that is solely supported by the argument that another mechanism can’t produce the observations. Negative arguments are nearly impossible to support because they boil down to an argument from ignorance.


Thanks for the update, @colewd.

I think that’s the worst case scenario I alluded to earlier: “It’s unlikely evolution could work.” or “It’s undemonstrated evolution could work.” Choose the specific cases carefully and it’s either a defeated notion or an undefeatable notion.

Perhaps Michael won’t call it IC in his next book, so we can avoid future confusion.


@colewd why is he arguing against a Darwinian pathway when we think non-Darwinian pathways are important too?


This argument is 23 years old and it sounds like he is soon to update it. Regarding Hunts paper he read Rhodes et al 1995 and Mackenzie et all 1994 confirming @Art’s claim that a new gene can be created by genetic recombination of existing genetic information. Here is his direct response.

1 Like

It’s only undefeatable if the Darwinian mechanism cannot be demonstrated to generate an irreducibly complex biological structure like the bacterial flagellum.

If a flagellum was produced in the lab with lots of bacterial replication and knockout of the DNA repair mechanisms then his claim is toast if you can show a mutational path to the functional flagellum.

Hi Bill,

Thanks for forwarding this comment by Behe. i will get back to this tonight, but for now, I am wondering - what do you think of Behe’s statement? Do you think it is convincing, or does justice to my essay?

Just wondering…

Again, thanks for this.


Note 1:
No. It’s being set up as a continuum: That there are some IC systems (IC version #1) that are accessible to evolutionary mechanisms and others that aren’t. If a possible explanation for system ‘X’ is made, one can always move on to system ‘Y’ not being explained. I don’t find the bacterial flagellum a particularly enlightening test case.

I believe that for most systems, the requisite signal or remaining evidence will make it impossible to trace historical pathways in any great detail. Signals decay with time. The bacterial flagellum is ancient and highly evolved. Personally, I’m surprised that there seems to be some links to secretory systems. If I wanted to test the limits of a hypothesis about ‘evolvability’, that is not a system I would consider tractable or likely to achieve conclusive evidence. In contrast, the blood clotting cascades found in mammals are younger and there seems to be more discernible connections, particularly with replication of proteases in the cascade. But even there, Behe asked whether the models could predicted whether changes in the components could modeled with terms including blood clotting times, clotting thickness & permeability, and not clotting under inappropriate conditions. Note: This is a level of detail we don’t even have for extant models of clotting cascades, so it’s unlikely any model of the evolution of the blood clotting cascade will come up to Behe’s level of request for quite some time. Color and scent receptors are also systems where enough data could be retained to evaluate ICness and evolvability, but again, possibly not likely at the level of detail Behe might request.

In actual science and determining the applicability of a hypothesis, researchers tend to investigate the simplest cases with the most likelihood of returning a positive results. So, what is the most recently emerged system for which we are likely to have sufficient data to discern evolvability? It’s not the bacterial flagellum.

Note 2:
It’s a queer asymmetry of logic that defeat of Behe’s IC hypothesis requires proof of evolutionary pathways. Ultimately it reduces to: “I might be right as long as you don’t prove me wrong”. That’s the problem of setting up a hypothesis as the negation of another hypothesis rather than as a positive argument. For example, the evolutionary hypothesis leads us to suspect is that connections of linear descent may be found between past and present organisms. What is the positive formulation of the IC hypothesis? Can it be formulated or evaluated independent of whether an evolutionary mechanism can be found?


What you, Behe, and other ID proponents need to show is that intelligent design was responsible for the bacteria flagellum instead of constantly arguing against evolution.


First is the good news is that Behe affirmed your hypothesis which gives us common ground to work with. I also agree that the 4-helix bundle is a challenge to explain through random mutation.

His last comment is speculative and I think as a minimum open for discussion. He claimed the comment as a possibility and not a conclusion.

I agree with this criticism but evolution can support this by walking back on its unsupported claims.

1 Like

Here is my take, very briefly:

— No Darwinian evolution of the supposedly new gene was shown. It was only discovered in that strain of maize

Maize is not subject to Darwinian evolution? This remark makes no sense at all.

— The ability of a random sequence of DNA to form a 4-helix bundle when transcribed beggars belief. It would not be expected to happen in many, many tries. That alone makes it seem very likely that the sequence was not random.

As you said, Bill, Behe grants that T-urf13 is pretty remarkable.

— The authors of the review write: “Although novel when discovered, it is now apparent that many other CMS genes have arisen by recombinational events involving other mitochondrial-encoded genes. In fact, mitochondrial DNA rearrangements are a significant force in changing the genome organization and in causing mutations in the mitochondrial genomes of higher plants (Mackenzie et al., 1994).” A distinct possibility implied by this, it seems to me, is that plant mitochondrial DNA is in fact built to provide easily accessible functional variability. In other words, intelligently provided, pre-existing information was tapped for this protein.

Doesn’t this sound sort of circular? Anyways, how would one determine this, without assuming the conclusion Behe wishes to draw?

Hopefully, Behe can take a bit more time and offer some more commentary about my essay.

One more comment - that “the 4-helix bundle is a challenge to explain through random mutation” is only a valid argument if one grants that protein functionality is rare in sequence space. It is pretty clear that this is not the case.


Raises a really interesting point. Consider @colewd’s falsification study…

Let’s imagine this happened. I’m pretty sure Behe would respond the same way…

After all, why not just suppose that God directly reached into the experiment to make a flagellum? Why suppose that the non-coding DNA was designed to evolve into a flagellum?

It seems that the argument is merely from incredulity, based on a strong assumption that complex proteins can’t evolve. If they do, it “beggars belief.” Honestly, it really seems to beg the question instead.

I think Mike stated an untested hypothesis so yes if he made a positive claim its circular but I don’t think thats what he did.

No, I think he would call his hypothesis falsified if he could tie the changes to random mutations.

Rare is a relative comment. Do you still stand by the range of “rareness” in your paper criticizing ID’s use of Axe’s work?

How exactly could that be done? @art tied the changes required for T-urf13 to random mutations, and Behe dismissed this as imossible to believe because it falsified his hypothesis. If not the way how @art did so, how could we by any means tie the changes to random mutations?