The AP Model and Shannon Theory Show the Incompleteness of Darwin’s ToE

self | January 26, 2009 | Jean F. Drew

[betty boop]

The AP Model and Shannon Theory Show the Incompleteness of Darwin’s ToE

By Jean F. Drew

“The commonly cited case for intelligent design appeals to: (a) the irreducible complexity of (b) some aspects of life. But complex arguments invite complex refutations (valid or otherwise), and the claim that only some aspects of life are irreducibly complex implies that others are not, and so the average person remains unconvinced. Here I use another principle—autopoiesis (self-making)—to show that all aspects of life lie beyond the reach of naturalistic explanations. Autopoiesis provides a compelling case for intelligent design in three stages: (i) autopoiesis is universal in all living things, which makes it a pre-requisite for life, not an end product of natural selection; (ii) the inversely-causal, information-driven, structured hierarchy of autopoiesis is not reducible to the laws of physics and chemistry; and (iii) there is an unbridgeable abyss between the dirty, mass-action chemistry of the natural environmental and the perfectly-pure, single-molecule precision of biochemistry.”

So begins Alex Williams’ seminal article, Life’s Irreducible Structure — Autopoiesis, Part 1. In the article, Williams seeks to show that all living organisms are constituted by a five-level structured hierarchy that cannot be wholly accounted for in terms of naturalistic explanation. Rather, Williams’ model places primary emphasis on the successful transmission and communication of relevant biological information.

Note here that, so far, science has not identified any naturalistic source for “information” within the universe, biological or otherwise. And yet it appears that living organisms remain living only so long as they are “successfully communicating” information. When that process stops, the organism dies; i.e., becomes subject to the second law of thermodynamics — the consequences of which the now-deceased organism had managed to optimally distance itself from while alive.

Evidently Williams finds Michael Behe’s irreducible complexity arguments insufficiently general to explain biological complexity and organization, and so seeks a different explanation to generically characterize the living organism. Yet his proposed autopoietic model — of the “self-making,” i.e., self-maintaining or self-organizing and therefore living system — itself happens to be irreducibly complex. That is to say, on Williams’ model, any biological organism from microbe to man must be understood as a complete, functioning “whole,” transcending in the most profound way any definition of a particular organism as the “mere” sum of its constituting “material” parts.

Further, the idea of the “whole” must come prior to an understanding of the nature and function of the constituting parts. Williams terms this idea of the “whole” as inversely causal meta-information; as such, it is what determines the relations and organization of all the parts that constitute that “whole” of the living organism — a biological system in nature.

Just one further word before we turn to Williams’ autopoietic model. To begin with the supposition of “wholeness” flies in the face of methodological naturalism, the currently favored model of scientific investigation, and arguably the heart of Darwinist evolutionary theory. For methodological naturalism is classical and mechanistic (i.e., “Newtonian”) in its basic presuppositions: Among other things, it requires that all causation be “local.” Given this requirement, it makes sense to regard the “whole is merely the sum of its parts” as a valid statement — those parts being given to human understanding as the objects of direct observation of material events. The presumption here is that, given enough time, the piling up of the parts (i.e., of the “material events”) will eventually give you the description of the whole. Meanwhile, we all should just be patient. For centuries if need be, as a collaborator once suggested to me (in regard to abiogenesis. See more below).

Yet subsequent to classical physics came the astonishing revelations of relativity and quantum theory, both of which point to “non-local” causation. The transmission of information across widely spatially-separated regions (from the point of view of the biological organism as an extended body in time) so as to have causative effect in the emergence of biological life and its functions is decidedly a “non-local” phenomenon. Indeed, non-local causation seems ubiquitous, all-pervasive in the living state of biological organisms, as we shall see in what follows.

Williams’ Autopoietic Model
Williams lays out the five-level, autopoietic hierarchy specifying the living system this way (parenthetical notes added):

(i) components with perfectly pure composition (i.e., pure elements)
(ii) components with highly specific structure (i.e., molecules)
(iii) components that are functionally integrated (i.e., components work cooperatively toward achieving a purpose or goal)
(iv) comprehensively regulated information-driven processes (DNA, RNA)
(v) inversely-causal meta-informational strategies for individual and species survival (we’ll get to this in a minute)

Pictorially, the model lays out like this:

Figure 1 summarizes the five-level, hierarchical specification of any living organism, microbe to man. But how do we get a handle on what this hierarchy actually means?

An interesting way to look at the problem, it seems to me, is to look at the available potential “information content” of each of the five “levels” or “manifolds” of the hierarchy.

You’ll note that Figure 1 depicts an ascending arrow on the left labeled “complexity.” For our present purposes, we’ll define this as “algorithmic complexity,” understood as a function that maximally yields “information content.” If we can find complexity measures to plug into the model, we might gain additional insight thereby.

Fortunately, algorithmic complexity measures have been obtained for certain levels of the hierarchy; i.e., Level (i) and Levels (iv) and possibly Level (v). For the latter two, the measures were taken with respect to the living human being. Figure 1 can thus be expanded as follows:

Notes to Figure 2:

¹ Gregory Chaitin: “My paper on physics was never published, only as an IBM report. In it I took: Newton’s laws, Maxwell’s laws, the Schrödinger equation, and Einstein’s field equations for curved spacetime near a black hole, and solved them numerically, giving ‘motion-picture’ solutions. The programs, which were written in an obsolete computer programming language APL2 at roughly the level of Mathematica, were all about half a page long, which is amazingly simple.”

On this basis, Chaitin has pointed out that the complexity we observe in living systems cannot be accounted for on the basis of the chemical and physical laws alone, owing to the paucity of their information content.

² George Gilder: “In each of the some 300 trillion cells in every human body, the words of life churn almost flawlessly through our flesh and nervous system at a speed that utterly dwarfs the data rates of all the world’s supercomputers. For example, just to assemble some 500 amino-acid units into each of the trillions of complex hemoglobin molecules that transfer oxygen from the lungs to bodily tissues takes a total of some 250 peta operations per second. (The word “peta” refers to the number ten to the 15th power — so this tiny process requires 250 x 10¹⁵ operations.)

A Word about Abiogenesis
Darwin’s evolutionary theory does not deal with the origin of life. It takes life for granted, and then asks how it speciates. Moreover, the theory does not elaborate a description of the constitution of the individual living organism, such as Williams’ irreducibly complex/autopoietic (“IC/AP”) model proposes.

It’s important to recognize that neither Darwin’s theory, nor Williams’ model, deals with the origin of life. It seems to me that evolution theory and ID are not necessarily mutually-exclusive. One deals with the species level, the other the biological structure of living individuals, the “building blocks” of species, as it were.

Yet there is tremendous hostility towards intelligent design on the part of many orthodox evolutionary biologists, which has gotten so bad in recent times that the more doctrinaire Darwinists have run to the courts for “protection” of their cherished beliefs (and interests personal and institutional), insisting that ID “is not science.” Judges are not scientists; in general they are ill-equipped to make judgments “on the merits” of scientific controversies. Yet they render judgments all the same, with profound implications for how science is to be taught. I fail to see how this redounds to the benefit of scientific progress.

If science is defined as materialist and naturalist in its fundamental presuppositions — the currently-favored methodological naturalism — then ID does not meet the test of “what is science?” For it does not restrict itself to the material, the physical, but extends its model to information science, which is immaterial. The problem for Darwinists seems to be that there is no known source of biological information within Nature as classically understood (i.e., as fundamentally Newtonian — materialist and mechanistic in three dimensions).

The problem of abiogenesis goes straight to the heart of this issue. Abiogensis is a hypothesis holding that life spontaneously arises from inert, non-living matter under as-yet unknown conditions. Although evolution theory does not deal with the problem of the origin of life, many evolutionary biologists are intrigued by the problem, and want to deal with it in a manner consistent with Darwinian methods; i.e., the presuppositions of methodological naturalism, boosted by random mutation and natural selection. That is, to assume that life “emerges” from the “bottom-up” — from resources available at Levels (i) and (ii) of the IC/AP model.

There have been numerous experiments, most of which have taken the form of laboratory simulations of “lightning strikes” on a properly prepared chemical “soup” (e.g., Urey, Miller, et al.). At least one such experiment managed to produce amino acids — fundamental building blocks of life (at the (ii) level of Williams’ hierarchy). But amino acids are not life. On Williams’ model, to be “life,” they’d need to have achieved at least the threshold of Level (iii).

For it is the presence of “functionally-integrated components” that makes life possible, that sustains the living organism in its very first “duty”: That it will, along the entire extension of its complete biological make-up (whether simple or highly complex), globally organize its component systems in such a way as to maximally maintain the total organism’s “distance” from thermodynamic entropy. An “organism” that couldn’t do that wouldn’t last as an “organism” for very long.

And so in order for the materialist interpretation of abiogenesis to be true, the “chemical soup” experimental model would have to demonstrate how inorganic matter manages to “exempt” itself from one of the two most fundamental laws of Nature: the second law of thermodynamics.

From cells on up through species, all biological organisms — by virtue of their participation in Levels (i) and (ii) — are subject to the second law right from creation. Indeed, they are subject to it throughout their life spans. A friend points out that the second law is a big arguing point for Macroevolutionists, who try to argue that the second law is irrelevent, i.e., doesn’t apply to living systems, because “it only applies to closed systems and not to open ones.” Thus they say that living systems in nature are “open” systems. But what this line of reasoning does not tell us is what such systems are “open” to.

And yet we know that every cell is subject to the second law — simply by needing to fuel itself, it subjects itself to the effects of entropy, otherwise known as heat death. And although it can and does stave off such effects for a while, doing so requires the cell or species constantly to deal with maintaining distance from entropy in all its living functional components, organized globally. Entropy plays a big part in all life — from cells to completed species.

When the successful communication of meta-information begins to slow down and break down, cells and species then begin to succumb to the effects of entropy, to which they have been subjected all their entire life. This is because they can no longer combat, or stay ahead of the “entropy curve,” due to inefficient communication processes and, thus, degradation of the maintenance procedures communicated to the cells via the meta-information system that is specific to each particular biological entity and to each particular species. After all, any species description is necessarily an informed description.

Yet another origin-of-life approach — the Wimmer abiogenesis experiment — is highly instructive. He managed to “create” a polio virus. He did so by introducing RNA information into a “cell-free juice,” and the polio virus spontaneously resulted.

Wimmer used actual DNA to synthesize polio RNA based on information about the polio virus RNA which is widely available, even on the internet. The RNA information was truly “pulled” from the DNA, which “resides” at the next-higher level. He could not synthesize RNA directly; he first had to synthesize the DNA from the raw information and then synthesize the polio RNA from the synthetic DNA.

Yet RNA information, like all information, is immaterial. In terms of the Williams’ hierarchy, clearly Wimmer had obtained an organism functioning at about Level (iii) — because it had sufficient information to propel it to that level, as “pulled” by the information available at the next-higher level where DNA information “resides” — Level (iv).

Unlike biological organisms expressing all five levels of the Williams model, the polio virus, though fully autonomous as an information processor (leading to its “successful communication” in Wimmer’s laboratory), somehow still doesn’t have everything it needs to be fully “autonomous” as a living being. A virus, for instance, is dependent on a living host in order to execute its own life program. As such, it is a sort of “quasi-life.” Shannon Information Theory helps us to clarify such distinctions.

Before we turn to Shannon, it’s worth mentioning that, according to H. H. Pattee and Luis Rocha, the issue of autonomy (and semiosis — the language and the ability to encode/decode messages) is a huge stumbling block to abiogenesis theory. For that kind of complexity to emerge by self-organizing theory, in the RNA world, the organism would have to involuntarily toggle back and forth between non-autonomous and autonomous modes, first to gather, and then to make use of information content as an autonomous living entity. The question then becomes: What tells it how and when to “toggle?” Further, it appears the source of the information content that can toggle non-life into life remains undisclosed.

Shannon Information Theory
The DNA of any individual life form is exactly the same whether the organism is dead or alive. And we know this, for DNA is widely used and proved reliable in forensic tests of decedents in criminal courts of law. And so we propose:

Information is that which distinguishes life from non-life/death.

Information, paraphrased as “successful communication,” is the reduction of uncertainty (Shannon entropy) in a receiver or molecular machine in going from a before state to an after state. It is the action which facilitates any successfully completed communication. Thus Shannon’s model describes the universal “mechanism” of communication. That is, it distinguishes between the “content” of a message and its “conduit”: The model is indifferent to the actual message being communicated, which could be anything, from “Don’t forget to put your boots on today — it’s snowing,” to Shakespeare’s Hamlet. The value or meaning of the message being transmitted has no bearing on the Shannon model, which is the same for all messages whatever. Pictorially, the Shannon communication conduit looks like this:

Information is further defined by its independence from physical determination:

“I came to see that the computer offers an insuperable obstacle to Darwinian materialism. In a computer, as information theory shows, the content is manifestly independent of its material substrate. No possible knowledge of a computer’s materials can yield any information whatsoever about the actual content of its computations. In the usual hierarchy of causation, they reflect the software or ‘source code’ used to program the device; and, like the design of the computer itself, the software is contrived by human intelligence.

“The failure of purely physical theories to describe or explain information reflects Shannon’s concept of entropy and his measure of ‘news.’ Information is defined by its independence from physical determination: If it is determined, it is predictable and thus by definition not information. Yet Darwinian science seemed to be reducing all nature to material causes.” — George Gilder, “Evolution and Me,” National Review, July 17, 2006, p. 29f.

Referring to the Shannon diagram above, we can interpret the various elements of the model in terms of biological utility, as follows:

Note the head, “noise.” Biologically speaking, with respect to the fully-integrated, five-leveled biological organism, “noise” in the channel might be introduced by certain biological “enigmas,” which broadly satisfy the requirements of Williams’ model and, thus, are living organisms. Shannon Information Theory describes such “enigmas” as follows:

Bacteria — typified by autonomous successful communication; bacteria are single-cell organisms. Because they are autonomous entities, communications follow the normal flow in Shannon theory — source, message, encoder/transmitter, channel, decoder/receiver. The bacteria’s messages are not “broadcast” to other nearby bacteria but are autonomous to the single-cell organism.

Bacterial Spores — typified by autonomous successful communication. Bacterial spores, such as anthrax, are like other bacteria except they can settle into a dormant state. Dormant bacterial spores begin regular successful communication under the Shannon model once an “interrupt” has occurred, for instance the presence of food. Anthrax, for instance, may lay dormant for years until breathed into a victim’s lungs, whereupon it actively begins its successful albeit destructive (to its host) communication, which often leads to the death of its host; i.e., the bacterium’s “food source.”

Mycoplasmas — typified as an autonomous bacterial model parasite successfully communicating. Mycoplasmas are akin to bacteria except they lack an outer membrane and so often attach to other cells, whereby they may cause such events as, for instance, the disease pneumonia. In the Shannon model, mycoplasmas are considered “autonomous” in that the communications are often restricted to the mycoplasma itself; e.g., self-reproduction. But because they also act like a parasite, they might alter the host’s properties and thus result in malfunctions in the autonomous communication of the host by, for instance, interfering with the channel.

Mimivirus — typified as an autonomous virus model parasite successfully communicating. Mimiviruses are gigantic viruses. They are viruses because they are parasites to their host, relying on the host for protein engineering. But the mimiviruses (unlike regular viruses) apparently do not need to be a parasite, and thus they are “autonomous” with regard to the Shannon model. But like the mycoplasmas, the presence of mimiviruses can alter properties of the host and thereby result in malfunctions in the autonomous communications of the host by, for instance, interfering with the channel.

Viroids — typified as non-autonomous virus-like noise/mutation contributing to successful/failed communication. Viroids have no protein coat. They are single strands of RNA that lack the protein coat of regular viruses. They are noise in the channel under the Shannon model; i.e., messages only that are not communicated autonomously within the viroids themselves. They can also be seen as “broadcast” messages, because viroids may cause their own message (RNA) to be introduced into the host.

Viruses — typified as non-autonomous virus noise/mutation contributing to successful/failed communication. Viruses feed genetic data to the host. They are strands of DNA or RNA that have a protein coat. Viruses are parasites to the host, relying on the host for communication; e.g., reproduction. In the Shannon model, viruses are either noise or broadcasts that are not autonomous in the virus and appear as noise messages to the host. It is possible that, unlike the polio virus which is destructive, there may be some viruses (and viroids) whose messages cause a beneficial adaptation in the host.

Prions — typified as non-autonomous protein noise/mutation contributing to successful/failed communication (protein crystallization). Prions are protein molecules that have neither DNA nor RNA. Currently, prions are the suspected cause of bovine spongiform encephalopathy — Mad Cow Disease. In the Shannon model, prions would be incoherent in the channel because they have no discernable message; that is, neither DNA nor RNA. Thus the prion would lead to channel or decoding malfunctions.

So far there is no known origin for information (successful communication) in space/time. This should be visualized as activity represented by the arrows on the above illustration. Possible origins include a universal vacuum field, harmonics, geometry.

Shannon’s mathematical theory of communications applied to molecular biology shows genuine promise of having some significant implications for the theory of natural selection in explaining the rise of information (successful communication), autonomy, and semiosis (language, encoding/decoding). — S. Venable, J. Drew, “Shannon Information and Complex Systems Theory,” Don’t Let Science Get You Down, Timothy, Lulu Press, 2006, p. 207f.

It seems worthwhile to note here that, under Shannon’s model, the thermodynamic “tab” is paid when the “molecular machine” goes from the before state to the after state. At that moment, it dissipates heat into the surroundings. Level (v) meta-information successfully communicated to the organism provides it with strategies to counter and compensate for local thermodynamic effects. Ultimately, when the organism reaches a state in which it is no longer successfully communicating, the entropy tab must be paid by ordinary means. And so eventually, the living organism dies.

Putting Williams’ IC/AP Model into Context
So far, the autopoietic model — though it provides an excellent description of the information flows necessary to establish and maintain an organism in a “living state” — seems to be a bit of an abstraction. Indeed, in order to be fully understood, the model needs to be placed into the context in which it occurs — that is, in Nature.

Each living entity as described by the model is a part and participant in a far greater “whole.” Niels Bohr put it this way: “A scientific analysis of parts cannot disclose the actual character of a living organism because that organism exists only in relation to the whole of biological life.” Including the species-specific meta-information unique to any particular species, which also controls and dictates how the entire biological system works as a “whole”; i.e., at the global level. And arguably, not only in relation to the entirety of biological life, but to the physical forces of nature, to inorganic entities, and to other biological beings, including the “enigmas” described above, which appear to be a sort of “quasi-life.” For even though they may be autonomous communicators, some of these “quasi-life” examples suggest an organic state that is somehow not “sufficiently informed” to stand on its own; i.e., they exemplify a state that needs to latch onto a fully-functioning biological entity in order to complete their own “program” for life — the very definition of a parasite.

The single most telling point that Williams’ model makes is that information is vital to the living state; that it flows “downward” from the “top” of his model — Level (v), meta-information — and not from the “bottom” of the model flowing “upwards” by the incremental means characterizing Levels (i) and (ii) — not to mention orthodox Darwinist expectation. On this model, Levels (i) and (ii) “do not know how to fit themselves” into the “biological picture.” For that, they need the information available at Levels (iii) to (v).

Many questions relevant to our exploration of the fundaments of biology have not been touched on in this article — e.g., what is the meaning of “emergence?” What is the manner in which “complexification” takes place in nature? What do we mean by “open” and “closed” systems? What do we mean by “self-ordered” or “self-organizing” systems in nature? (And what does the prefix “self” mean with respect to such questions?)

But since we’re out of time, we won’t be dealing with such problems here and now, though I hope we may return to them later. Instead, I’ll leave you, dear reader, with yet another depiction of Figure 1, this time elaborated to show the total context in which the irreducibly complex, autopoietic model is embedded:

Note the model now sits, not only with respect to its natural environment, but also with respect to the quantum domain of pure potentiality, and also with respect to a (proposed) extra-mundane source of biological information.

I think for the biological sciences to actually progress, a model such as Williams’ IC/AP model is worthy of serious consideration. Remember, Darwin’s theory is wholly classical, meaning dimensionally limited to 3-space, to local, mechanical, largely force-field-driven material causation. Relativity and quantum theory have both moved well beyond those precincts. It’s time for the Darwinian theory of evolution to “catch up” with the current state of scientific knowledge — and especially with the implications of information science.

©2009 Jean F. Drew

[betty boop]

My deep thanks and appreciation to Alamo-Girl and CottShop for kindly agreeing to read the draft and for suggesting improvements, almost all of which were implemented. (And they'd all have been, had we not run out of space!)

Thank you so very much, A-G and CS!

[Filo]

There is no such thing as a “compelling case” for abject stupidity.

[DBCJR]

Thank you for this post!

[svcw]

As always brilliant.

[svcw]

So why did you post then?

[randog]

In layman’s terms, explain Williams’ Autopoietic Model, part v.

[RSmithOpt]

Reminds me of being forced to attend past sales meetings where dress, jargon, cell phone colors, advert pamphlet layout/impact were much much important than the speakers understanding (having a freeking clue) the function, future capacity, and design theories of a newly designed product pushed by our customers and available for market.

Sales people are not allowed into my labs unless accompanied by corporate.

[js1138]

Note here that, so far, science has not identified any naturalistic source for “information” within the universe, biological or otherwise. And yet it appears that living organisms remain living only so long as they are “successfully communicating” information.

That could be a very long time, considering that single-celled organisms are essentially immortal. Programmed cell death is a rather modern invention in living things, a requirement of having specialized cells.

[RSmithOpt]

Seriously though, my earlier post is not a slam.....the basis of the article is the thinking that has progressed for AI to become an everyday reality very soon.

[Alamo-Girl]

Thank you so much for the opportunity to preview your outstanding essay, dearest sister in Christ!

[DBCJR]

A Word about Abiogenesis
Darwin’s evolutionary theory does not deal with the origin of life. It takes life for granted, and then asks how it speciates.

---

That begged question has always screamed at me. Forget that ToE is an unproven theory when applied to inter-species, it does nothing to explain genesis.

[varmintman]

Excellent.

[js1138]

Forget that ToE is an unproven theory when applied to inter-species

If DNA cannot be used to determine kinship, there are a lot of court decisions that need to be reversed.

[varmintman]

Evolosers are always saying that. What it really translates into is something like “Hey, if we get good enough with the ad-hominems, we might could skate by with only having to defend one untenable ideological doctrine instead of TWO of em!”

[DBCJR]

Yes, OJ Simpson started that process as I recall.

[jimmyray]

There is no such thing as a “compelling case” for abject stupidity.

Regarding "compelling case", nothing like an ad hominem to validate your point of view.

[DBCJR]

Ah yes, the global warming theory! If it is factually occurring, which is a subject of debate itself, that humans “caused” it is HIGHLY debatable. What extends even further into absurdity is that algore is lobbying the Obama Administration and Pelosi’s Congress to make climate change illegal - that’s right, you heard correctly. If the climate changes, WE BROKE THE LAW!

[js1138]

Yes, OJ Simpson started that process as I recall.

Lots of OJ supporters on evolution threads.

[js1138]

I don’t see any reference to Szostak in your essay. Is there a reason?