More On T-urf13 – A Response To Arthur Hunt And Others
|March 3, 2011||Posted by Jonathan M under Intelligent Design|
A week ago I blogged about Arthur Hunt’s failure to refute Michael Behe on his concepts of irreducible complexity and the edge of evolution. My article quickly ignited into a heated debate which honed in on a host of different issues. Within just 48 hours of publishing the piece, more than 70 responses had ensued (as of now, there is more than 170!). Among those who commented was none other than Arthur Hunt himself, who raised a few criticisms of his own. This is my response to these criticisms.
I have distilled the objections down to three headings. These are:
- The challenge I (and others) pose to those of a neo-Darwinian persuasion to provide a detailed and testable account of the origin of biochemical systems is an exercise in the pot calling the kettle black. After all, have ID proponents articulated a testable theory which explicates how a biochemical system might have arisen by virtue of intelligent causality?
- Arthur Hunt claims that the processes of random recombination are demonstrably sufficient to account for the T-urf13 gene, and that there is evidence to support that this is actually what happened.
- Arthur Hunt provides a link to a blog where he critiques the Axe (2004) paper which I cited regarding the prohibitive rarity of functional protein folds with respect to the vast sea of combinatorial sequence space.
Let’s address each of these in turn.
1. Is The Pot Calling The Kettle Black?
This is an objection I hear rather frequently: While it may be true that those of a neo-Darwinian persuasion have utterly failed to provide a detailed step-by-step account of any complex biochemical system (thus demonstrating the feasibility of their model), ID proponents have also failed to provide such detailed accounts of how the design came to be implemented in the living system. It seems to me that this objection, however, is missing a subtle point: A rigorous and testable scientific theory of design detection is quite different from a rigorous and testable scientific theory of design implementation. I am happy to freely confess with regards the latter that, though one or two interesting ideas have been advanced (see, for instance this pro-ID paper by Michael Sherman — not to be confused with Michael Shermer — in the journal, Cell Cycle.), we are currently lacking any form of testable historic narrative which outlines in testable fashion how, say, the bacterial flagellum came into existence by virtue of intelligent causality. What we do have, however, is a rigorous and testable scientific theory of how design might be detected. The ability to detect the products of intelligent design is not contingent upon the ability to discern how that design came to be implemented. One may justifiably infer that, let’s say, a complex computer system is best explained by intelligent causation without knowing anything about the identity of the designer or how it came to be assembled.
Moreover, it is still unclear how mind is able to interact with matter, even in the context of human minds (I lean in the direction of substance dualism with respect to my concept of brain and mind). We make design inferences every day, and it is not clear at all that one requires independent knowledge of the designer or the agent’s modus operandi in order to justify that inference.
Intelligent design is consistent with some concepts of mechanism with regards the history of life (such as certain variants of front-loading). But it is by no means committed to such scenarios. Creation of independent taxons de novo is one possibility, and one concept which I am personally sympathetic towards.
2. Recombination – A Sufficient Causal Explanation?
Arthur Hunt asks:
[A]re you suggesting that the mechanisms known to be involved in homologous and/or non-homologous recombination (the processes that pretty clearly gave rise to the T-urf13 gene) do not obey the rules of chemistry? This is the only way I can make sense of your argument.
Stated another way, you seem to be claiming that T-urf13 must be beyond the reach of “Darwinian” processes, that in this case must involve recombination. Since T-urf13 obviously exists, and since we know when, where, and how (at least in general terms) it arose, then your argument seems to boil down to an assertion that recombinational mechanisms are either insufficient (obviously wrong) or guided in some (unstated) way. If you are trying to say something else, a bit of clarification would be appreciated.
Again, what is being discussed is the causal sufficiency of random (that is, unguided) processes in accounting for the de novo gene (in this case, T-urf13): Did ample probabilistic resources exist in order to justify attributing such an event to chance? It seems fairly clear to me that they did not. Let me re-iterate again: A demonstration of molecular homology or common ancestral derivation is not, in and of itself, a causal explanation. It is not that the processes of homologous and non-homologous recombination do not obey the laws of chemistry. Rather, we are positing that agency is an indispensible factor in accounting for novel functional information – just as a combination of agency and chemistry is indispensible in producing this blog post. Moreover, the transmembrane system described by Hunt is substantially simpler in composition than almost all of the systems raised by Behe (see, for instance, my detailed discussion of the complexity of the flagellar apparatus here).
The Uncommon Descent writer and commenter “PaV” made note of one or two things which I had neglected in my original piece. Here are his unedited remarks (which can be found at comment #101 in my previous post):
(1) The method of inheritance of mitochondria is not the same as that of nuclear DNA—the benchmark of neo-Darwinism.
(2) The idea of three CCC’s is hypothetical, and not more. The “third” CCC that Hunt proposes—a binding site between a toxin and the gated ion-channel—can just as easily, and more plausibly, be explained by the toxin ‘evolving’ a binding site for the ion-channel.
(3) The kind of “recombination” that takes place in (plant) mitochondria is not your normal Mendelian recombination. Hunt eludes to this when he says at : “Jonathan, are you suggesting that the mechanisms known to be involved in homologous and/or non-homologous recombination (the processes that pretty clearly gave rise to the T-urf13 gene) do not obey the rules of chemistry?”
Notice his use of “rules of chemistry” and not, e.g., the “rules of biology”. This is because in plant mitochondria, ‘sub-circles’ of mitochondrial DNA can accumulate through ‘intra-molecular’ recombination. There apparently is some kind of machinery that allows portions of the original ‘circular’ genome of the mitochondria to take parts of the original mitochondrial genome and fashion other circles. This machinery (notice that this terms presupposes some kind of inter-purposiveness) obeys not genetic, but chemical rules, with the result that a huge diversity of “recombinations” can be cobbled together. As someone pointed out in Hunt’s blog back when T-urf 13 was being discussed, there are great similarities between the diversity bought about in anti-body production and that of plant mitochondrial recombination. This, quite obviously, falls outside of normal “Darwinian” mechanisms. It seems a little bit disingenious that Art is now correctly referring to this quite different type of recombination as following chemical rules, yet insist that the CCC’s he finds here dispute Behe’s claims that “Darwinian” mechanisms can’t produce much more than 2 CCC’s of complexity.
(4) Let’s just be aware that Behe uses White’s number of 1 in 10^20 in EoE, a number that represents not theoretical figures of probability, but actual in vivo probabilites; i.e., this is what is found in the lab. As to ‘theoretical figures’, the number should be 1 in 10^16, and, hence, 3 CCC’s would represent, theoretically, 1 in 10^48 improbability, under the UPB used by most scientists. I add this simply for the sake of clarity.
(5) In a recent paper, a “de novo” gene was being touted. Guess what? It turns out that a portion of a “non-coding” gene and its flanking element was involved in the manufacturing of this “de novo” gene. This is exactly what we find in T-urf 13. Hence, when I used the term “machinery” in (3) above, indeed, this seems to be a maneuver that living cells have at their disposal, thus warranting a search for this new mechanism and not the false claim of truly “de novo” genes. As in the case with T-urf-13, the “new” gene is nothing more than the demolishing of another gene: i.e., the critical portion of the “de novo” gene represented no more than a portion of another gene. (The transcribed portion of T-urf 13 that provides this ‘amazing’ gated ion-channel, is only a very small part of the “de novo” gene.) Again, this is consistent with Behe’s latest article.
(6) There are two “nuclear restorers” that can restore the plant to ‘male fertility’ from the sterile condition found in the Texas maize from which T-urf 13 is derived. Interestingly, and provocatively, when the “nuclear restorers” work their magic, guess what? T-urf 13 is no longer found: evidence, again, that a “mechanism”, and “machinery” is at play.
Getting back to the original post here, Jonathan quite correctly demonstrates that Darwinian mechanisms are being assumed to be at work with the manufacture of URF-13 protein, and, yet, from all indications, whatever is happening to the maize, has very little to do with true Darwinian mechanisms. I would hope Arthur Hunt might acknowledge this.
Let’s just finish here by pointing out again that T-urf 13 involves a kind of degradation of maize. In the case of the Texas maize–hence the T—the T-urf 13 was located by researchers because it was there that the toxin that decimated the corn grown in Texas in the late 60′s attached itself. So the “manufacturing” of this “de novo” gene proved to make the maize less fit. This is in keeping with Behe’s latest findings.
3. Is Douglas Axe Wrong?
Axe’s views about the nature of the functional space of protein sequences are plainly wrong. I try to illustrate this here (I also explain why the ID party line with respect to the 2004 JMB paper is incorrect.) Given this, the objections that rely on the alleged fantastic isolation of functionality in sequence space are pretty irrelevant. (This is, of course, one of the points of the T-urf13 example. It shows quite clearly that functionality, even irreducible complexity, is not the stupendous impossibility that is claimed by Behe and others.)
Hunt thus directed me to a blog (which I have happened across before) he had posted in 2007 critiquing Douglas Axe’s 2004 JMB paper. A thorough critique of this article is likely to shape up to be fairly lengthy and time-consuming. I am also not a specialist in protein bioscience and it has been a while since I read Axe’s 2004 JMB paper in any great detail. But I do think that Hunt raises a number of points in his blog which need to be appropriately and adequately addressed. I thus intend to write about this in more detail, and in greater length, when time permits. For the time being, let me point out that Axe’s conclusions regarding the rarity of functional protein catalytic domains in sequence space is not an isolated result. In addition to the Keefe and Szostak (2001) paper which I mentioned in my previous post, a similar result was obtained by Taylor et al. in their 2001 PNAS paper. This paper examined the AroQ-type chorismate mutase, and arrived at a similarly low prevalence (giving a value of 1 in 10^24 for the 93 amino acid enzyme, but, when adjusted to reflect a residue of the same length as the 150-amino-acid section analysed from Beta-lactamase, yields a result of 1 in 10^53). Yet another paper by Sauer and Reidhaar-Olson (1990) reported on “the high level of degeneracy in the information that specifies a particular protein fold,” which it gives as 1 in 10^63. I could easily continue in the same vein for some time. I also strongly encourage Arthur Hunt and others to read Douglas Axe’s excellent review article in Bio-complexity which covers this topic in more detail. Axe also has contributed an excellent chapter to the recently-published The Nature of Nature — Examining The Role of Naturalism in Science (a book which I highly recommend), which explains the relevant concepts at an accessible level for non-experts.
[UPDATE: Further discussion of Axe’s 2004 JMB result, in the context of Hunt’s essay, is now available here.]