Evolution of the bacterial flagellum

Dr. Ian Musgrave

Pace Michael Behe and William Dembski, bacterial flagella are not irreducibly complex. Here is Ian Musgrave’s clear explanation, “Evolution of the Bacterial Flagellum.”

Dr. Musgrave points out, “The specification of “ an outboard motor,” which provided the IC [irreducibly complex] system description of motor, shaft, and propeller, is a flawed human analogy to the actual flagellar system.” He also points out that Behe’s own definition of irreducibly complex systems excludes flagella, whose original function was not motility but secretion. Dembski, by building on Behe’s flawed description, in my opinion goes further astray into “Why bumblebees can’t fly” territory.

There’s much more! Read the article.


Irreducible no more

It looks as if Nick Matzke’s hypothesis is correct, and the base of a bacterial flagellum was re-purposed from a structure used in the immune system. In fact, scientists are ready to recognize that the flagellum base plate is a variation of the type III secretion system. Nick wrote

Finally, if I were doing a revision, I would update the terminology along the lines suggested in Desvaux et al. 2006 (“Type III secretion: what’s in a name?” Trends in Microbiology 14(4), 157-160, April 2006 – DOI). As they point out, the terminological distinction between “flagellum” and “type 3 secretion system” is dubious and artificial, and it is more true to acknowledge that flagella have a type III secretion system. Therefore, there are two known groups of type III secretion systems, flagellar and nonflagellar, abbreviated F-T3SS and NF-T3SS.

There is much more to be said about recent research and its implications for flagellum evolution. For the near future I intend to post my thoughts on this in the new flagellum evolution section [UPDATE: fixed the link] of the Panda’s Thumb blog.

Rhetorical tactics: the Dembski Dodge

Wesley ElsberryWesley R. Elsberry, in discussing possible responses to factual evidence, mentioned a several avoidance tactics. This is from a long discussion thread in antievolution.org, where Wesley summarizes the patterns of common arguments. I’ve extracted them from their discussion thread and highlighted them here: how IDists avoid responding to real-world evidence.

One of the tactics is a hallmark of William Dembski’s responses about evolution, so I’ve decided to call it the Dembski Dodge.

The one I want to talk about is described below.

Non-Evidentiary Responses

The other category of approach is to ignore, so far as possible, any mention or discussion of actual fossil evidence… There are many routes to achieving this end. The simplest is non-response. The challenged person may decide that not saying anything further is the best option…. Yet another strategy is to discuss theoretical issues as if theory did away with the need to actually look at the empirical data.

funny picturesAnd there you have it. That’s the entire point of Dembski’s argument: construct a mathematical will-o-the-wisp and point at it as though it were the evidence we vainly seek.

My brother used to tease young ladies by eliciting various random facts about them, such as their eye colour, height, dog’s name, home town, favorite food, best subject, number of siblings, and so on, then multiplying the probablility of all those things being true, and producing a mathematical proof that the sweet young things in question were so improbable that they might simply disappear at any time!

And that, in a nutshell, is Dembski’s approach to the facts of biology. However, Dembski cooks the books in his favor by insisting that all the calculated events must have happened simultaneously instead of accumulating over a period of time.

Forty-seven ways to produce heritable genetic change

Allen MacNeill has taken of the challenge of falsifying the objection to evolution usually framed, “How can random mutation” produce enough variation for evolution?” or “It’s only random mutation and natural selection.” In debate, random mutation is often assumed to be the substitution of one single amino acid for another, in other words, a single point mutation. Allen says,

Allen MacNeillI promised a list of the real sources of variation that provide the raw material for evolutionary change. It’s taken me a while, but here it is. This list includes “random mutation,’ of course, but also 46 other sources of variation in either the genotypes or phenotypes of living organisms. Note that the list is not necessarily exhaustive, nor are any of the entries in the list necessarily limited to the level of structure or function under which they are listed. On the contrary, this is clearly a list of the minimum sources of variation between individuals in populations. A comprehensive list would almost certainly include hundreds (and possibly thousands) of more detailed processes. Also, the list includes processes that change either genotypes or phenotypes or both, but does not include processes that are combinations of other processes in the list, again implying that a comprehensive listing would be much longer and more detailed.

Here’s his list of forty-six other ways to produce genetic change.

Commenter -DG adds,

Not so much an addition as perhaps a minor correction with the deletions. Not all deletions necessarily result in a frameshift, although of course this would be the most common for deletions of any multiple not of three. But it is certainly possible for 3 (or some multiple of 3) nucleotides to be inserted or deleted at the same time resulting in insertion/deletions of the protein primary sequence. This seems to be especially prevalent in loop regions of the protein three-dimensional structure and may be one of the mechanisms by which new protein domains occasionally arise.

Commenter SPARC adds,

You may add exon shuffling. It belongs to the gene structure section (insertions/deletions) but in most cases the reading frame isn’t changed. I would further add exonization of transposable elements usage of alternate promoters, alternative splicing, multiple polyA signals and trinucleotide repeat expansions.

Commenter Art says,

It would be very nice if you could fold a whole ‘nother universe of genetic and regulatory mechanisms into your list. For the sake of completeness, and because the ID movement (typified by Behe’s recent dismissal of these core mechanisms) cannot deal with the concept.

I speak, of course, of the regulation of gene expression at the level of RNA and protein breakdown. Not only are they central to life (it’s doubtful that multicellular life could exist without the negative regulatory mechanisms afforded by these processes), they are inherently “accessible” to evolutionary modification. This is because, in the ID vernacular, they involve low information modes of recognition and action.

Keywords for a revised list: microRNA, siRNA, exosome, ubiquitin, cullin, E3 ligase, proteasome, SUMO.

Commenter -DG replies,

Great additions SPARC, since I work on protein evolution I really should have remembered to add exon shuffling. Along the same lines of alternative splicing we also have RNA editing. It isn’t carried out in many known systems but its an interesting system as well.

TalkOrigins Letter of the Month

For the January, 2004 TalkOrigins Letter of the Month, the runner-up is “A Dialogue with Dembski?”

It has been clear since at least Dembski’s book “Intelligent Design: the Bridge between Science and Theology” that all of Dembski’s verbiage about “specified complexity” was essentially pointless, because at the key point, when specified complexity is applied to biology, Dembski relies on Behe’s irreducible complexity to exclude the gradual buildup of “specified information” to reach “complex specified information,” i.e. “specified complexity.”Thus Dembski’s argument has always reduced to Behe’s argument, and I never saw much point in dealing with anything else — although many others, bless their hearts, have delved into DembskiLand rather deeply. But it remains the case that if Behe is wrong then Dembski’s arguments collapse as well (well, except for his emergency backup positions like “Well, even if evolution can produce IC that just means that specified complexity was frontloaded into creation from the beginning, or is introduced undetectably at the quantum level. Or something.”)

The great thing now is that Dembski sees the problem on some level and is trying to shore up IC. This leads to all kinds of entertaining obfuscation, along the lines of:

Dembski: “Specified complexity cannot be produced by evolution.”

Evo: “Well duh, you made ‘cannot be produced by natural processes’ part of the definition of specified complexity. Your statement is thus a meaningless tautology.”

Dembski: “Irreducible complexity cannot be produced by evolution, and this means that IC systems exhibit SC.”

Evo: “You are ignoring change-of-function, an important evolutionary process that has been cited as the explanation for systems with ‘irreducible complexity’ ever since Darwin himself.”

Dembski: “OK, well I guess evolution can hypothetically produce IC after all, but there’s no evidence that cooption actually happens in natural settings, the only good evidence of this is in technological evolution.”

Evo: “Here’s a bunch of examples of change-of-function in natural systems. A bunch of these have even resulted in IC toxin degradation systems in historical times.”

%d bloggers like this: