A Decade Spanning Single Exchange
by GutsWesley Elsberry has a post up that he touts whenever something negative is posted about me on the internet, I wanted to examine some of Elsberry's claims. He writes:
I pointed out that the mammalian middle ear ossicular chain is an IC system providing an impedance-matching function, and that the impedance-matching goes away if you remove any of the parts. Nelson tried to deny that this qualified as IC, at least in part because the fossil record is clear that the system evolved.
Not surprisingly, that is completely false.
In the forum I had made it as clear as I possibly could that the theory that this transition had no possible intermediates was a purely creationist claim with not-so-much importance to the concept of irreducible complexity, due to the differences in concepts that exist for the evolution of molecular machines vs. morphological characters.
It's not that I agree with everything I have written on the internet (people change). However, I have no idea where Elsberry gets the notion that my argument that the middle ear bone system is not IC was due to the fossil record being clear that the system evolved. In fact, Behe was very clear in his book that IC systems, can evolve, through circuitous routes, something that Elsberry, to this day, is quite obviously ignorant of. So that was never the issue. Nonetheless, I was arguing in the thread Wesley links to, against PvM, who claimed quite explicitely that hearing itself was the function of the higher middle ear system. And PvM was ( and is) quite wrong.
My argument was very simple, the middle ear bone system, with respect to hearing, does not require every component and therefore it is not irreducibly complex. I've said this several times within the same post.
Elsberry , in the same post, kept harping on impedance matching:
My point was not that impedance-matching in the middle ear is
*necessary* to any amount of hearing, but rather that trying
to dismiss the impedance-matching function on the basis that
hearing itself is not completely eliminated is a digression.
However, it is not a "digression", as the same argument is employed against the irreducible complexity of the eubacterial flagellum, and has been for years. That the flagellum can lose various parts and still function as a type III secretory system has been used by anti-ID activists for over a decade, especially at the recent ID trial. And rightly so , not a digression in this context, because it is relevant when thinking about how it might have developed. Thus, I was employing the same reasoning for the middle ear system.
Given that the eardrum and the three middle ear bones are mechanically coupled in a serial fashion, impedance matching will be compromised if any part of the path is disrupted. However the hearing loss is not infinite due to interruption. This is because when sound reaches the oval and round windows it can still excite the cochlea leading to the perception of sound. With this direct acoustic route, the hearing sensitivity is compromised due to the impedance mismatch. In such a case a respective animal (no matter if lizard or mammal) would then be able to 'hear' if only vibrations in the ground or very deep sound.
It might be, however, that low frequencies, for example received by the lungs, would have alternate, non-tympanic pathways. This is certainly the case in frogs, where frequencies below 3-400 Hz apparently are received by 'extratympanic' pathways. This may be similar to bone-conduction, i.e. sound-induced vibrations of the skull. There are many species of 'earless' frogs that have secondarily lost their middle ear, and they have reduced sensitivity at higher frequencies compared to 'normal' frogs, but essentially unchanged sensitivity at lower frequencies.
In lizards, the situation seems to be the same. Since the ear is a pressure-gradient ear, the tympanic response at low frequencies is very small, that part of the low frequency sensitivity is extratympanic. However, at higher freqs hearing clearly depends on the eardrum and its impedance matching.
Regardless, one other point needs to be discussed, and this is with regards to fundamentals. Evolutionary changes around the ear are usually attributed to developmental 'regulatory' changes. In an evolutionary context, to understand how change in development leads to change in character states that can be selected for (ie, evolve) we need 1) to understand how development works, how developmental events actually yield character states, 2) be clear on how we identify a character state, and 3) formulate an ordering mechanism to describe the correlations of 1 and 2. Historically, developmental changes leading to morphological changes were thought to occur through changes in the timing or rates of a developmental event. More recently, workers in the field have considered changes in the relative positioning of a developmental event.
I would expect ear-bone evolution to involve developmental regulatory changes. Any change in the place or time of development of a structure is regulatory in nature. Changes in proteins are less likely to underlie such changes because most regulatory proteins are constrained due to their pleiotropic roles in the development of different structures. This is not to say, of course, that such changes are irrelevant, morphological evolution can also be affected by protein sequence changes. Thus, protein sequence change in a trans-regulatory molecule, (eg coding changes in a transcription factor) would affect downstream regulation of downstream genes. Gene duplication and divergence allows both regulatory changes in a copy, and changes in coding sequence to some modified activity of one of the copies, with the other maintaining its former activity. But certainly more important when it comes to molecular machines, which is the focus of Behe's first book.
November 27th, 2008 at 7:11 am
I hope you don't mind if we start slowly. You say:
I assume I'm misunderstand something in this sentence. If Behe's point wasn't that IC systems cannot evolve, thus creating a need for an explanation involving intelligence, what was his point? Are you using "evolve" to refer to something other than what the theory of evolution describes, perhaps? I'm afraid any other discussion will be confused unless we clear this up from the start.
Comment by don provan — November 27, 2008 @ 7:11 am
November 27th, 2008 at 7:18 am
I think you're making a very good point here: this is the same argument used to suggest the flagellum isn't IC. Does that mean you agree that the flagellum is not IC?
Comment by don provan — November 27, 2008 @ 7:18 am
November 27th, 2008 at 9:30 pm
Don wrote:
Behe's argument has always been one of improbability of molecular machines, his argument is that although circuitous routes are possible, the more complex an IC system, the less likely a circuitous route can accomplish the feat.
Don wrote:
The flagellum is IC as a motility organelle, if you remove one part, the machine doesn't work. But as a type III secretion system, it does work without various parts. If the eardrum is no longer physically connected to the inner ear, i.e. loss of a single ossicle, the impedance matching effectively reduces to zero (in fact "poor impedance matching" is just an oxymoron), but hearing does not require all parts of the system. In my opinion, this is apples and oranges though.
Comment by Guts — November 27, 2008 @ 9:30 pm
November 28th, 2008 at 1:29 am
Can IC systems evolve or not?
So it does work, it just doesn't work in some special way you think is important? Can you be more specific about how working as a type III secretion system is not good enough, but working as hearing in an entirely different way is? It seems as if we're making inferences based on the physical function without any logical foundation for thinking the exact relation of the physical functions has any bearing.
Comment by don provan — November 28, 2008 @ 1:29 am
November 28th, 2008 at 1:35 am
Don wrote:
I thought I had made that clear, the answer is according to Behe, it depends on the IC system. PCP degradation is IC but easily evolvable, the eukaryotic cilia not so easy.
Don:
I can't make any sense out of this. Not in "some special way", simply motility. Bacteria become non-motile if you remove a component from it's core. Hearing works if you remove one of the ossicles. Please be more specific with your questions because I don't understand them.
Comment by Guts — November 28, 2008 @ 1:35 am
November 28th, 2008 at 11:26 am
The problem? I thought that was the whole point.
Comment by The Pixie Again — November 28, 2008 @ 11:26 am
November 28th, 2008 at 12:50 pm
I thought this was settled when Rock pointed out that an irreducible minimum needed for present function was a concept he (a researcher) was accustomed to applying. Do we have to continually re-post the same information?
Comment by Bradford — November 28, 2008 @ 12:50 pm
November 28th, 2008 at 2:55 pm
Wow! An IC post with ONLY 8 comments!
My observations in no particular order:
1. IC is a well-defined concept, and to the best of my knowledge, no one (other than opponents) have changed the definition.
2. Further, I propose that the concept of IC is NOT controversial, but rather Behe's use of it.
3. Behe said it was impossible for direct evolution to come up with IC systems, but only improbable (low-end possible on the Explanatory Continuum rating) for indirect routes to come up with IC system(s).
This is probably just a reiteration of what was said above, but hey, I've got real work to do! See you during the next model analysis!
Comment by JJS P.Eng. — November 28, 2008 @ 2:55 pm
November 28th, 2008 at 5:31 pm
Well, I don't know about continually, but I've never heard about it. It might be easier to just post a link whenever it comes up instead of bothering to complain about it not being common knowledge.
I'd be more sympathetic if I could think of any relation whatsoever between "irreducible complexity" and "irreducible minimum" beyond the fact that they both use the term "irreducible".
Comment by don provan — November 28, 2008 @ 5:31 pm
November 29th, 2008 at 8:29 am
Hi JJS P.Eng
Okay, so here is Behe's definition from Darwin's Black Box, p39
That has to be the definition all ID proponent use, right, because no one on the ID side has changed the definition. Yes?
Which leaves us wondering: So what? Improbably things happen. Just roll a die. The chances of you getting that number is one in six, it was an improbable event – but it still happened. The evolution of the bacterial flagellum by an indirect route was improbable, but it looks like it still happened.
Behe's IC concept offers no indication that evolution did not happened. And indeed, Behe apparently believes it did happen. So does this help ID? Not one bit.
Bradford
Again, this concept offers no indication that evolution did not happened. So does this help ID? Again, not one bit.
So as far as I can tell the whole IC idea is irrelevant to ID! And yet IDists keep promoting it. Why is that? My suspicition is that they hope the less science-savvy will get fooled by the technobabble, and come away thinking that IC proves evolution did not happen, but I would be interested to hear otherwise from the IDists.
Comment by The Pixie Again — November 29, 2008 @ 8:29 am
November 29th, 2008 at 11:32 am
Behe vs. Behe.
One problem is that the two definitions give different answers for the same structure. If we consider the ossicles of the middle ear as having an impedance-matching function, then removing any one of them causes the system to lose its function. That means the system is IC by the first definition. However, we know the transitions that led to the bones in the reptilian jaw being coopted for hearing, so by the second definition, the system is not IC (degree zero). But if we didn't know, we might have a false positive!
Another problem is the delineation of function and system, which are not always discrete entities, but often overlap with other functions and systems. The boundaries of the function and system under consideration are often arbitrarily defined, i.e. observer-dependent.
Still another problem is when the definition slips during a discussion. This type of conflation is pervasive in ID circuits. Sometimes when I ask for clarification, I am told that I should know because IC is a well-defined concept.
Comment by Zachriel — November 29, 2008 @ 11:32 am
November 29th, 2008 at 3:21 pm
I really love that page from ISCID, and I'm really proud of them for keeping it. The reason it's so perfect is because it puts the fallaciousness of the IC argument in a nutshell: in Darwin's Black Box, Behe started with what he was claiming was an observable characteristic that showed evolution couldn't explain the structures. But then the subsequent definition he gave is just the opposite, defining IC based on the ability to explain the structures through evolutionary processes without any regard for their observable properties. Armed with this tool, an ID proponent can "disprove" evolution by simply shifting between the two conflicting definitions without reference to anything about the original structure.
Comment by don provan — November 29, 2008 @ 3:21 pm
November 29th, 2008 at 5:09 pm
Look at these comments by The Pixie Again and Bradford. They're quoting me, but my post is not there! Strange.
Comment by Zachriel — November 29, 2008 @ 5:09 pm
November 30th, 2008 at 6:19 pm
Hello there Pixie. In your Behe quote, I believe you provided both the definition and a claimed consequence (according to Behe) of IC systems. The definition of IC is the first sentence:
The consequence of the IC definition (according to Behe) followed:
In Chapter 8 of The Design Matrix, Mike Gene quotes a 2000 journal paper from Thornhill and Ussery that support the above claim.
I don't know. You tell me. Enlighten me please.
Until Mike Gene took it for a test drive. (see The Design Matrix or read this review at EE for more details)
Comment by JJS P.Eng. — November 30, 2008 @ 6:19 pm
November 30th, 2008 at 6:28 pm
Good day Zachriel. Let's take a closer look at all three definitions stated in the link you provided:
Dembski seems to have elaborated more on implied details (well-matched, mutually interacting, nonarbitrarily individuated parts, irreducible core) which seem extraneous. It should be noted that Behe himself elaborated on these details in Darwin's Black Box.
The so-called second definition, IMHO, seems to be a different item altogther. The first defines what an IC system looks like. The "second" describes the what the evolutionary pathway to an IC system is (according to Behe). A path or road is a completely separate item from the destination with IC being the destination and evolution being the road.
Comment by JJS P.Eng. — November 30, 2008 @ 6:28 pm
November 30th, 2008 at 6:52 pm
If the "degree of irreducible complexity" is zero, then it is quite apparent the object under consideration would be considered *not* irreducibly complex. As described in my previous comment, this can lead to ambiguity, false positives and conflation.
Comment by Zachriel — November 30, 2008 @ 6:52 pm
November 30th, 2008 at 7:55 pm
Zachriel:
I think it is obvious (to me at least) that IC would have a "floor" value for "degree of IC", and that zero is not that floor. IMO, the floor value is two components.
Also implied in the definition is the more components that make up the core of the IC system, the greater the complexity. I'd be interested if this is a linear relationship (doubtful) or exponential (more likely).
I say doubtful for the linear relationship since most relationships are not purely linear. They may start out linear, but then either become exponential or multi-linear.
Comment by JJS P.Eng. — November 30, 2008 @ 7:55 pm
November 30th, 2008 at 9:52 pm
In other words, you would agree that a degree zero of irreducible complexity is *not* irreducibly complex. And, if we don't know a plausible pathway, one might erroneously conclude a higher degree than is warranted—simply due to ignorance.
Then there is the common conflation.
Comment by Zachriel — November 30, 2008 @ 9:52 pm
December 1st, 2008 at 12:49 am
Zachriel said:
Actually, I implied that there is no such thing as zero-degree of IC. There is a lower bound to IC and zero is not it. Something is either IC or it is not.
No, identifying an IC system comes first, then one can investigate whether co-option or some other mechanism can assemble it. One must identify the destination before the path can be completely known.
Comment by JJS P.Eng. — December 1, 2008 @ 12:49 am
December 1st, 2008 at 1:31 am
Zachriel wrote:
The problem here is that you're specifying a different function, hearing. If you stick with impedance matching, there are unselectable steps because it requires each and every part.
Zachriel:
Thats true with most things in Biology. Not really a "problem". Function however can be very, I would even say often is, specific, for example, the epsilon subunit inhibits ATPase activity if F1 is not attached to Fo. In order to say so definitively one would have to assess each and every biological function currently known.
Comment by Guts — December 1, 2008 @ 1:31 am
December 1st, 2008 at 9:02 am
If something has zero degree of irreducible complexity, is it irreducibly complex? Your next comment may make it clear.
Okay. That is how I would normally construe it. We observe a system with a function. We determine whether removing any part causes it to lose its current function. If so, it is Irreducibly Complex.
Of course, Irreducible Complexity is a typical result of evolutionary processes through a number of mechanisms, such as functional drift and cooption. So, an Irreducibly Complex system can have a Degree of Irreducible Complexity of zero. (It's obvious why there would be confusion.)
Behe's problem is the second half of his statement. It's not a powerful challenge. It's not a challenge at all!
Comment by Zachriel — December 1, 2008 @ 9:02 am
December 1st, 2008 at 9:03 am
…
Comment by Zachriel — December 1, 2008 @ 9:03 am
December 1st, 2008 at 9:07 am
Acoustical impedance didn't occur suddenly during the transition from reptilian jaw bones to mammalian ear bones.
Comment by Zachriel — December 1, 2008 @ 9:07 am
December 2nd, 2008 at 11:26 am
The function of the ossicles (ear bones) in the mammalian middle ear is to efficiently transmit sound vibrations from the tympanic membrane to the inner ear. They are, in Dembski's wording, "a set of well-matched, mutually interacting, nonarbitrarily individuated parts such that each part in the set is indispensable to maintaining" the basic function. If we remove any of these three ossicles, the function fails. The mammalian middle ear is reasonably said to meet the definition of irreducibly complex.
What Behe is trying to claim is that evolution works by putting parts together, and that we should be able to reverse the process and remove parts one at a time. It seems that the parts have to be assembled in toto, by analogy to human design.
In the case of the mammalian middle ear, that's not what happened. And it is instructive of many other similar evolutionary transitions. All three components were already there, inherited from the common ancestor, and already fit together. But they performed a different function. The three parts all gradually morphed into their new configurations, slowing gaining a new function and losing the old.
Early in the 20th century, studies of developing vertebrate embryos showed that the same structures that led to jaw bones in reptiles led to ear bones in mammals. It wasn't until many decades later that a variety of fossils organisms with transitional features had been found; including this recent discovery, Yanoconodon allini. So we have a case where studies in living embryonic development predict the content of 125 million year old strata. Rocks. This is why the Theory of Evolution is considered such a strongly supported scientific theory: it makes important, interrelated predictions in many different fields of study.
Comment by Zachriel — December 2, 2008 @ 11:26 am
December 3rd, 2008 at 1:04 am
Zachriel wrote:
Actually thats not correct. There is no "transition" from poor impedance matching to better impedance matching. As I mentioned, "impedance matching" refers to an "equality" of two impedances. So the concept "poor impedance matching" is kind of an oxymoron.
Zachriel wrote:
No. Behe had already mentioned that circuitous routes can occur , but there is no "losing the old" here. One of the problems the middle ear has to overcome is the mismatch in impedance between air and the cochlea, but impedance matching by the middle ear results in only a 30 dB increase in sound energy transmitted to the inner ear. But the ear is sensitive to a range of 100 dB.
However as I said the situation is somewhat different for morphological characters than for protein based molecular machines.
Comment by Guts — December 3, 2008 @ 1:04 am
December 3rd, 2008 at 8:59 am
Nor did I say that. Impedance matching of the tympanic middle ear occurred very early in amniote evolution. That's the whole point.
Mammalian ossicles no long function as jaw bones.
It's much the same. As described by Muller (1918), we have an existing complex function A adding a helper B. Through a process of evolution, the components become optimized and A becomes dependent on B, and B may then lose its original function. This is how a number of cascades evolved, including the blood clotting cascade.
The reason I often mention the mammalian middle ear as an example is that it allows the general reader to understand why Behe has constructed a fallacious argument. We have a clear example of cooption and mutual evolutionary accomodation of the components that have become dependent on one another.
But as the origin of many molecular structures is not known or not known with much certainty or detail (they evolved early in the history of life, left no fossils, and the far vast majority of lineages are extinct), this does allow ID to argue in the Gaps.
Comment by Zachriel — December 3, 2008 @ 8:59 am
December 3rd, 2008 at 9:27 am
As Austringer notes, an increase in sensitivity of 30 db is hardly insignificant. Also, the degree of impedance matching depends on sound frequency. As for selection, the pathways are incremental, and sensitive hearing is of clear benefit to both predator and prey organisms.
Comment by Zachriel — December 3, 2008 @ 9:27 am
December 3rd, 2008 at 9:31 am
Müller & Tsuji, Impedance-Matching Hearing in Paleozoic Reptiles: Evidence of Advanced Sensory Perception at an Early Stage of Amniote Evolution, PLoS ONE 2007.
Comment by Zachriel — December 3, 2008 @ 9:31 am
December 4th, 2008 at 2:54 am
Zachriel:
You said:
That it appeared is not at issue.
Zachriel:
The evolution of the middle ear was clearly dependant on improved sound conduction, which did move gradually and was never lost with the acquisition (obviously) of impedance matching. Even in Yanoconodon we see that hearing occured through bone conduction, and gradually improved as it evolved.
Zachriel:
You're talking about a geneticist and then jump to a molecular process, so I'm not sure why you said it's "much the same", it isn't. Most regulatory proteins are constrained, so morphological evolution is realized by such. Regulatory changes played a role in changes to remodeling of features such as ear bones. So, clearly regulatory changes can affect timing, location, and level of expression of genes. These can produce big results, and I think are likely crucial in morphological evolution in addition to structural changes (even if it turns out that the latter is more important), especially when change is occurring rapidly or among reasonably closely related forms that share a body plan that is modified in various ways.
There are a great many 'character states' that the cells and tissues which will instruct the formation of the middle ear can be described as having. As for the ear region then it is likely that overlap of a vareity of mechanisms have been involved in evolutionary changes in this region.
Zachriel:
Behe already mentioned that circuitous routes are possible.
Zachriel wrote:
Where did I say it was insignficant?
Comment by Guts — December 4, 2008 @ 2:54 am