Irreducible Complexity and TalkOrigins

I just wanted to come back on this. Behe, when he coined IC, never intended it to be used on the gross level - it was meant to address complex systems with several well-matching interacting parts at the molecular level. In his book he explains the eye at the macroscopic level is not IC as you have explained, but the molecular mechanism of phototransduction (the conversion of photons to chemical/electric signals) can be an example of IC. For phototransduction to occur, the light absorbing component 11-cis-retinal changes to all-trans-retinal and dissociates from opsin; the now transformed molecule interacts with a G protein (transducin) which sets of an amplifying cascade ultimately converting cGMP to GMP and hence closing cGMP gated channels. This molecular process is irreducibly complex as all the many parts are required, the parts are well-matched (i.e. they wouldn't have effects on other molecules) and they interact. Behe argues that it is these systems, complex at the molecular level, which includes the blood clotting cascade, signalling cascades, bacterial flagellum and cillia, which cannot be created by undirected darwinian evolution. I'm not much of a scientist but from the academic journals I have seen Behe does contribute. His idea has been attacked and he has responded. For example, he has written in response to Shanks and Joplin (who equated IC to redundant complexity) at the famous jstor and he has other articles to his name on that online collection of journals.

Darwinism works by the accumulation of gradual benficial changes in offsprings guided by natural selection. Behe differentiates between an indirect and a direct Darwinian model for creating an IC system. The direct model is where a simple functioning system gradually becomes more complicated and an indirect model is where parts with various other functions come together and create an IC system. The direct route to an IC system is by definition not possible as an IC system requires all the parts for it to function so could not have had a physical (as opposed to conceptual) simpler functional precursor. As for the indirect route, Behe concedes it is possible but highly implausible depending on the number of "parts" and the extent of how well-matched the components are (he makes a distinction between simple interaction - SI - , which have unspecified interactions, and IC). He further argues the more these systems are found the less likely an indirect Darwinian route could account for IC systems. I personally believe Behe gives a satisfactory reason as to why Darwinism cannot account for IC systems at the molecular level.As to non-Darwinian mechanisms of creating an IC system, Behe suggests only two have been proposed: symbiosis (Margulis) and complexity theory (Kauffman); however neither can account for IC systems as Behe explains. Direction is required because to create an IC system an "end-point" must be envisaged. Darwinian evolution is blind, unintelligent and undirected and therefore does not see ahead (Dawkins describes this brilliantly in his Blind Watchmaker); an IC system however requires direction and therefore Darwinism cannot account for it.The solution Behe proposes to Darwinian and non-Darwinian explanations for IC systems is design, and he argues design is the best explanation. There is no need to know who or what the intelligent designer is, but that one (or many) necessarily exist for IC systems to have come about is the argument. Francis Crick quite seriously said aliens came to earth and planted spores to seed the earth. Religious folk would naturally invoke God as the intelligent designer. It's not difficult to imagine how an intelligent designer would direct the genome to create an IC system. Behe offers examples in his book (pg. 199 - 205)Edited by Suroof, : No reason given.

You seem to have misunderstood the implications of an IC system. The following are two passages from Behe's book which I believe addresses your concerns:"We must also consider the role of the laws of nature. The laws of nature can organise matter - for example, water flow can build up silt sufficiently to dam a portion of a river, forcing it to change course. The most relevant laws are those of biological reproduction, mutation and natural selection. If a biological structure can be explained in terms of natural laws, then we cannot conclude that it was designed. Throughout this book, however, I have shown why many biochemical systems cannot be built up by natural selection working on mutations: no direct gradual route exists to these IC systems and the laws of chemistry work strongly agaisnt the undirected development of the biochemical systems that make molecules such as AMP. Alternatives to gradualism [darwinism] that work through unintelligent causes, such as symbiosis and complexity theory cannot (and do not even try to) explain the fundamental biochemical machines of life. If natural laws peculiar to life cannot explain a biological system, then the criteria for concluding design become the same as for inanimate systems. There is not magic point of IC at which Darwinism is logically impossible. But the hurdles for gradualism become higher and higher as structures are more complex, more interdependent.""The designing that is currently going on in biochemistry laboratories throughout the world - the activity that is required to plan a new plasminogen that can be cleaved by thrombin, or a cow that gives growth hormone in its milk, or a bacteria that secretes human insulin - is analagous to the designing that preceded the blood-clotting system. The laboratory work of graduate students peicing together bits of genes in a deliberate effort to make something new is analagous to the work that was done to cause the first cilium."As for the existence of IC systems, the cilium and the blood clotting cascade, are as good examples as any. The above description of phototransduction is also an example of IC.Edited by Suroof, : No reason given.

Retinal is not part of the protein. It is a derivative of vitamin A that is bound to opsin, a G-protein coupled receptor (GPCR). The opsin is necessary as its conformational change as a result of the attachment or detachment of the light sensitive molecule determines whether it can activate the G protein transducin. The components of the systems must be specific, and in this case they are: transducin will activate phosphodiesterase via its alpha subunit, phosphodiesterase converts cGMP to 5'GMP, and cGMP no longer maintains cGMP gated channels open. By "change structural features" you mean change a few amino acids on the opsin molecule or the phosphodiesterase enzyme, it probably won't have much effect (although you'd be surprised how much change even one amino acid can cause: consider sickle cell anemia where the sixth amino acid on the beta chain changes from valine to glutamate). However, this would not disprove the highly specified nature of the parts involved, since even then the system would be IC. That all the components are necessary - the rhodopsin (retinal + opsin), transducin, GMP molecules and cGMP gated channels (incidentally other regulatory parts are required like guanylyl cyclase converting GTP to cGMP, opsin kinase and arrestin for reversing opsin state, and GTP and GDP in reasonable concentrations are naturally required for a G protein to work). Because it is the same parts in those signalling cascades that act on cGMP. Guanylyl cyclase produces cGMP from GTP and phosphodiesterase enzymes converts it to 5'GMP. Phosphodiesterases are themselves subject to activation and inhibition by several compounds and drugs, but the relative non-specificity of this one component does not undermine the specificity of the system (as yet, the parts before and after are specific and there are no other means, except externally, of influencing PDE). I'm not sure about varying organisms but there are different opsin molecules within the human retina (three cone opsins and one rod) which gives us the spectral preferences of the cones and rods. Despite this, the opsin molecules are not highly variable but this small degree of variability results in preferences for different wavelengths, and this does not detract from its specificty. There are other known opsin molecules in other regions but they are not involved in vision. I suppose you are referring to dolphins, whales, puffer fish etc. However Behe does say the blood clotting cascade is irreducible only after the intrisinc and extrinsic pathways meet (i.e. at the basic level of regulation, thrombin, fibrin, plasmin, TPA and probably a little beyond that, it is irreducibly complex; Doolittle attempted to show that mice did perfectly fine without fibrinogen and plasminogen but he misread the paper he was citing: http://www.arn.org/...mb_indefenseofbloodclottingcascade.htm)Edited by Suroof, : No reason given.

Let us look at one IC system - the blood clotting cascade. In his book, Behe describes quite extensively the cascade and the consequent regulation. Despite describing the entire cascade he says "Leaving aside the the system before the fork in the pathway, where some details are less well known, the blood clotting system fits the definition of IC". The component parts are fibrinogen, prothrombin, stuart factor, accelerin, plasminogen and tPA, and this is leaving aside the control of stuart factor by the two pathways and tPA itself. [a breif sketch: after a wound certain factors are activated and via a cascade activate stuart factor; prothrombin is inactive and requires activation by being converted to thrombin; however thrombin is available in very small amounts; small amounts of thrombin activates accelerin and together with the now activated stuart factor, accelerin, activates more prothrombin; thrombin activates fibrinogen creating clots; fibrin (active form of fibrinogen) are bound together by FSF and eventually after healing has taken place FSF are broken down by plasmin the activated form of plasminogen - activated by tPA]The question is: could this have evolved in a Darwinian fashion? Yes it does seem that way because people generally overlook Behe's above statement. For example, Doolittle in response to Behe wrote "In particular, we noted that fish should not have both Hageman Factor and prekallekrin, two of the factors described in Behe's outline of blood clotting in his book." (Home - Boston Review) But the problem is Behe did not say Hageman Factor and prekallekrin formed part of the irreducibilty of the IC system. It is in that same article (in the last paragraph) where Doolittle makes the mistake of misreading a paper and assumes mice are healthy with plasminogen and fibrinogen knocked out. Behe refers to this paper as well as a similar paper by Doolittle in the same year (2003) in his Afterword "Ten Years Later" (2006). The two papers, as Behe wrote, "compares sequences...sequence comparisons can give an interesting overall picture of what proteins came first and second, and of who is related to whom, but can't even begin to tell what is the driving process". The two papers make the assumption Darwinian evolution had taken place in creating the clotting cascade, and the comparisons are used to highlight it but not to show a step-by-step pathway of how it took place.In an article defending the ICness of the blood clotting cascade, Behe said a better definition of IC would be: the number of non-advantageous unselected steps before the selected step are too many to occur at random (http://www.arn.org/...mb_indefenseofbloodclottingcascade.htm). For blood clotting to evolve, an explanation must give a selective advantage of each step, and not just a sequence comparison of the various proteins. The reason this can't be done (by Darwinian evolution) is because it works gradually, and therefore all the proteins I mentioned above (which have to appear simultaneously) could not have come about except by several unselected steps.Edited by Suroof, : expandEdited by Suroof, : No reason given.Edited by Suroof, : No reason given.

The question is: what functional intermediates could there be? As Behe wrote there are other simpler ways of stopping blood flow from a wound like platelet aggregation and constriction of blood vessels but the clotting cascade could not have arisen from these for obvious reasons. The simplest blood clotting system imaginable might be just a single protein that randomly aggregated when the organism was cut. However this has several problems: no triggering of clotting factor, non-directional, unregulated - all of which would result in low perfusion and ultimately death. So, what exactly is the functional intermediate?"One could imagine a blood clotting system that was somewhat simpler than the real one - where, say, Stuart factor, after activation by the rest of the cascade, directly cuts fibrinogen to form fibrin, bypassing thrombin. Leaving aside for the moment issues of control and timing of clot formation, upon reflection we can quickly see that even such a slightly simplified system cannot change gradually into the more complex, intact system. If a new protein were inserted into the thrombin less system it would either turn the system on immediately - resulting in rapid death - or it would do nothing, and so have no reason to be selected. Because of the nature of a cascade a new protein would immediately have to be regulated..." (Behe) This is not an argument from ignorance. The idea of IC and ultimately design has a positive and negative element. The negative argument is that such interactive systems resist explanation by the tiny steps that a Darwinian path would be expected to take. The positive argument is that their parts appear arranged to serve a purpose. Richard Dawkins in his Blind Watchmaker makes a clear case for design, that biological systems clearly look designed (of course, he goes on to say Darwinian natural selection can account for design). Hence, the positive argument is in a nutshell: 1. we infer design whenever parts appear arranged to accomplish a function, 2. the strength of the inference is quantitative and depends on the evidence; the more parts and the more intricate and sophisticated the function the stronger is our conclusion of design 3. aspects of life overpower us with the appearance of design (Dawkins) and 4. since we have no other convincing explanation for the strong appearance of design, Darwinian pretensions notwithstanding, then we are rationally justified in concluding that parts of life were indeed purposely designed by an intelligent agent (Darwin's Black Box, pg. 265)Edited by Suroof, : No reason given.

That was not the argument. The argument was: we have no OTHER convincing explanation, and the explanation of an intelligent agent is the best explanation of IC available. Hence, there is a negative element that Darwinian evolution cannot account for these systems, but there is also the positive element that a purposeful arragnement of parts to serve a function (which varies in the clarity of design - i.e. some systems are more clearly designed than others) requires an intelligent designer.Edited by Suroof, : No reason given.

I don't know much about evolutionary algorithms and how they create design or complexity in biology but the ID proponent, mathematician Dembski, appears to contest the idea that EA can account for IC systems (which he says is biological SC): http://www.antievolution.org/...19990913_explaining_csi.html

The system that evolved (disulphide bond) was akin to IC but as Behe explains IC systems vary in their complexity, and for some, design is easily inferred, and for others it is not so easily inferred. As for the quantitative point at which systems become irreducibly-complex-beyond-Darwinian-explanation, Behe extends on this in his The Edge of Evolution which I haven't yet read or purchased but intend to - I guess an expansion on the probability argument (for Ringo) is also contained in that work.

I was addressing your question, not turning my back to Behe. He mentions SC in biology and says IC systems are SC and EA cannot account for them. Nobody disputes this. But because Darwinian evolution may have some hand in creating some level of "design", it is the ID argument that at the highest level of complexity, at the edge, EA does not have the capacity to produce this level of complexity - and which requires an intelligent agent.

Yes Dawkins does say the point of the program was to distinguish between one-step and cumulative processes and he admits it is misleading as Darwinian evolution has no end goal. However, if its purpose was only to delinate single step and cumulative selection, the program was mostly redundant, and it shows no real evolutionary algorithm exists to explain complexity. Edited by Suroof, : english

Even assuming the intelligent agent cannot be explained, it doesn't rule it out as the best explanation. For example somebody who views Mt Rushmore for the first time even if he/she doesn't know the designer he/she will immediately correctly assume it was designed (despite the problem of explaining the designer).ID goes as far as saying there is a designer; as for identifying and explaining the designer that is the job of metaphysics, philosophy, theology - it isn't really science (not the conservative naturalistic take on science anyway) unless you assume the intelligent designer is natural (aliens).

No they haven't - the blood clotting cascade, the cilium, phototransduction (http://www.arn.org/docs/behe/mb_idfrombiochemistry.htm) and many others Yes that is the question - and I have yet to read Behe's latest book which attempts to address that question. I suppose we could start by asking how Darwinian evolution could account for systems way beyond the edge like the blood clotting cascade or the cilium Well the default is intelligence, and the claim that there are alternatives must be proven:"But this raises the obvious question, whether there might not be a
fundamental connection between intelligence or design on the one hand and
specified complexity on the other. In fact there is. There's only one known
source for producing actual specified complexity, and that's intelligence.
In every case where we know the causal history responsible for an instance
of specified complexity, an intelligent agent was involved. Most human
artifacts, from Shakespearean sonnets to Drer woodcuts to Cray
supercomputers, are specified and complex. For a signal from outer space to
convince astronomers that extraterrestrial life is real, it too will have
to be complex and specified, thus indicating that the extraterrestrial is
not only alive but also intelligent (hence the search for extraterrestrial
intelligence-SETI).Thus, to claim that laws, even radically new ones, can produce specified
complexity is in my view to commit a category mistake. It is to attribute
to laws something they are intrinsically incapable of delivering-indeed,
all our evidence points to intelligence as the sole source for specified
complexity. Even so, in arguing that evolutionary algorithms cannot
generate specified complexity and in noting that specified complexity is
reliably correlated with intelligence, I have not refuted Darwinism or
denied the capacity of evolutionary algorithms to solve interesting
problems. In the case of Darwinism, what I have established is that the
Darwinian mechanism cannot generate actual specified complexity. What I
have not established is that living things exhibit actual specified
complexity. That is a separate question."Dembski explains why EA cannot generate SC - do not ask me to explain it! - http://www.leaderu.com/...es/dembski/docs/bd-algorithms.htmlEdited by Suroof, : Include supplementary site

A good review of this is : http://www.designinference.com/...2003.08.Encyc_of_Relig.htm.We can detect design - for example SETI looks for patterns in radio signals received. As that essay shows if signals were received with beats and stops corresponding to the prime numbers scientists would conclude intelligence. Dembski goes on to explain why we infer this: firstly because the signal doesn't HAVE to be that way - it isn't necessay, it is contingent; secondly because the signal is long, therefore the probability of getting any known sequence is low - it is complex; and thirdly we can relate to the signals through an independent information/ known patter in this case prime numbers - it is specified.If aliens did observe Mt. Rushmore they would conclude intelligence because the complex arrangement corresponds to an independent known pattern (the human face).IC systems also exhibit this feature of SC as arriving at IC through Darwinian evolution (indirectly) is amazingly small (therefore the system is complex) and the system complies to an independent function beyond the capacity of the parts themselves (therefore the system is specified). Dembski goes into a bit more detail here: http://www.designinference.com/....Irred_Compl_Revisited.pdf

Sorry, what I meant to say was: Dawkins program isn't an example of an actual EA and actual EAs explaining SC doesn't exist.