Tuesday, March 13, 2007
Platonic Forms and Front Loading
I’m currently working through Robustness and Evolvability in Living Systems, and came across the following information which seems to be right in line with Denton’s evolution by natural law ideas:
A final, especially counterintuitive feature of RNA sequence space is that all frequent structures are near each other in sequence space. Consider a randomly chosen sequence that folds into a frequent structure and ask how far one has to step away from the original sequence to find a sequence that folds into this second structure…For instance, for RNAs of length n = 100 nucleotides, a sphere of r = 15 mutational steps contains with probability one a sequence for any common structure. This implies that one has to search a vanishingly small fraction of sequence space…to find all common structures.
One of the issues that came up was this -- just how much does this really have to do with intelligent design? If the evolutionary space itself is set up to evolve, does this really have anything to do with Intelligent Design?
Anyway, this is a very good question, and to begin to understand it we need to understand two distinct (but not mutually exclusive hypotheses) in the Intelligent Design world -- front-loading and platonic forms. Just to note, however, not everyone uses these terms equivalently. Especially front-loading, which has a different meaning for nearly every individual who uses it. Anyway, I will attempt to define these two positions, talk about how they relate to each other, and finally, how they relate to this quote from Wagner.
Front-loading is the hypothesis that at some point in the past (usually at the origin of life), organisms were given a rather large deposit of information. The history of life from that point onward has been primarily governed by that information, specializing into the different species we have today. In this view, the original organisms potentially had _more_ information in them than individual organisms do today. The original organisms contained the major patterns and building blocks for all of the major evolutionary themes. Biological patterns may have been stochastically mixed and distributed throughout life (i.e. there was necessarily not _one_ preset destination in mind, just a set of biologically-meaningful patterns from which organisms could choose), but ultimately, the main driver for life patterns comes from this deposit of information.
Now, information, in order to be information, is essentially physics-independent. For example, when you install software on a computer, while the physics of the computer are very much applicable for running the software, you cannot use the physics of the computer to create the software. Intelligent agents use creativity to create the software in a manner that is wholly independent of the physics of computers. Front-loading focuses on the informational aspect of biology -- the physics-independent deposit of information given to the original organism or organisms. Most people who hold to this view think that there is at least some of that original deposit left hanging around in "junk DNA".
The person most active in this area of thinking (if I understand him correctly) is MikeGene and others at TelicThoughts.
Platonic forms is the idea that physics is set up so that there are only a small set of possible configurations that life could have. The reason that life keeps coming up with the same type of solutions over and over again is that these are the forms allowed by physics itself. Therefore, biological forms are time-independent and almost completely non-contingent since they are mandated by physics (in the front-loading scenario, they were completely contingent on the information given in the front-loading act).
Michael Denton is the primary proponent of this view, contained in his papers The Protein Folds as Platonic Forms, Physical law not natural selection as the major determinant of biological complexity in the subcellular realm: new support for the pre-Darwinian, and The Laws for Form Revisited. I believe that this is similar to Conway-Morris's claim in Life's Solution, but I have not read it.
Benton's main point is that physics only allows certain biological motifs. And thus, while there are untold numbers of protein sequences, there are only a small number of folds available to them. Therefore, life, even life as we know it, is not contingent on accidents, but instead is the planned outcome of physics.
Comparing the Theories
Both theories are at least somewhat related to Intelligent Design, though platonic forms is more difficult to examine as a hypothesis. Front-loading puts the infusion of information into biological systems as a process that happens essentially within-time, while platonic forms has the design process occurring pre-time or outside of time.
Note that, on a detail level, the two are mutually exclusive. Information requires freedom of choice for the agent, while platonic forms is specifically about excluding freedom of choice through time. That doesn't mean there aren't both mechanisms in play, only that a single mechanisms cannot be simultaneously part of both (though there can be aspects of each -- the point is that one is the result of a limitting process and the other the result of a freeing process).
My own opinion is that there is a platonic-defined set of biological forms, but that they are fundamentally unreachable without an infusion of information. Both processes are active, with platonic forms being the part that keeps system perterbations from becoming catastrophic, but that the preloaded information is what helps adapt to new situations.
Theologically, I think that God did this in order to show both His transcendence and His immanence. He is transcendent, and made physics according to a specific biological plan. However, so that we would know that God is active (and required) in Creation, he made it so that the platonic forms were unreachable except through design (front-loading). That way, we know that biology is not necessary, but instead chosen.
So, in reference to the above-posted Wagner reference, the existence of a clustered functional space in RNA is clearly platonic, while the ability to find that space initially in the original organisms smacks of front-loading. I was going to include another instance of front-loading in this section of Wagner in this post, but it's already too late and I need to study for a test tomorrow and get some sleep.
As I have already told, I found johnnyb’s summary about frontloading (on his name-linked site) very illuminating. So I would like to add some comments, completely accepting, as a starting point, his definitions and classification of the various positions. If I understand well, we could agree that, from a design point of view (we will here leave alone the darwinist points of view, assuming that we already agree they are completely unsubstantiated and self-contradictory) there are at least three great lines of thought, as follows:
1) Platonic forms. I cite from johnnyb: “Platonic forms is the idea that physics is set up so that there are only a small set of possible configurations that life could have. The reason that life keeps coming up with the same type of solutions over and over again is that these are the forms allowed by physics itself”.
I think we have to make a distinction here. There is a weak form of this argument, which is in essence the same as the general fine tuning argument of physics: physical laws and constants are intelligently selected from an almost infinite configuration space, so that our ordered universe is possible. Any slight deviation from the specific set of values we observe would make the universe chaotic (no atoms, molecules, galaxies, and obviously no life). In this form, which I strongly support, it is evident that very important information has been selected by the designer in planning the whole universe, and that kind of design is absolutely “necessary” for life to emerge. But the point remains open about the “sufficiency” of that information for life. Life could still have emerged by completely deterministic mechanisms (once the correct premises were set before the big bang), like in the darwinian model. Or it could still need other information additions.
I think we all can agree on this weak kind of “platonic forms” in the basic physical laws, and it still remains a very strong cosmological argument for the existence of God.
But I understand that your definition of the “platonic forms” view is stronger than that. It implies that “physics only allows certain biological motifs. And thus, while there are untold numbers of protein sequences, there are only a small number of folds available to them”.
I must say that I have difficulties to accept this stronger formulation. First of all, one should demonstrate that such a “restriction” of possible motifs and protein folds is a necessary consequence of basic physical laws, and I am not aware of any argument in that sense. Second, I think that the only purpose of such a formulation seems to be that of allowing a purely “natural” mechanism for the emergence of biological information, once the basic physical laws are fixed. But all the ID arguments, CSI and IC first of all, have demonstrated that such an emergence is impossible by chance, and is best explained by design. The only other possibility is necessity, but that would require “new” physical laws, of which we have presently no knowledge, or detailed mechanisms linking known physical laws to he above said restrictions.
Besides, I can’t see how any “platonic forms” explanation could account for OOL. Not only there is no known physical law which can account for the spontaneous generation of complex organic molecules from inorganic matter, just the contrary is true: known physical laws clearly demonstrate the impossibility of such an event. Indeed, all OOL scenarios, from urey-miller ponds to rna worlds, are mere fiction, as has been well discussed elsewhere.
And even if we admitted (but I don’t!) that only few protein folds are accessible, and that for some mysterious reason they are the ones which can bear function, one should still explain the higher informational levels, like the order and control of function, the procedure code (still unknown to everyone), the error management, the general plan for multicellular organisms, etc. It seems obvious to me that all these aspects cannot be explained in terms of mere protein folding, however platonic.
So, while I perfectly agree that specific fine tuning is necessary for life to emerge, I believe that all the ID approach is evidence against a purely mechanistic explanation of biological information, even allowing for very intelligent initial choices.
2) Frontloading. I cite from johnnyb: “ Front-loading is the hypothesis that at some point in the past (usually at the origin of life), organisms were given a rather large deposit of information. The history of life from that point onward has been primarily governed by that information, specializing into the different species we have today”.
That’s an interesting point of view, but again I have some difficulties, although of a different nature.
First of all, if I understand well, this theory assumes a specific act of “information imparting” (or creation, if we prefer), at OOL. On that I agree. I think that, however one considers what happened after, OOL can only be explained by a very unusual event. Maybe natural laws were not violated (it is possible, in principle), but it is difficult to conceive OOL as a “gradual” event, because nothing we know or can conceive of is even near to a “precursor” of the simplest living beings. So I strongly believe that the only plausible scenario for OOL is that bacteria, or archea, probably very similar or identical to those we know today, must have been “assembled” from non-organic matter according to a specific plan by a designer.
But, always if I understand well, according to the frontloading hypothesis the designer not only provided these first living beings with the information for their existence, survival and reproduction, but also with the information to “evolve” to the more complex species, up to humans. In other words, once that information was frontloaded in the first living beings, the following “evolution” can be explained by physical laws and mechanistic events, obviously exploiting the initial information. I understand also that, according to some supporters of this theory, the initial “extra” information is no more present, having exhausted its role, or as johnnyb states: “ most people who hold to this view think that there is at least some of that original deposit left hanging around in “junk DNA””. Well, that seems consistent enough, but here are my objections:
a) First of all, why? I mean, obviously one can postulate any model one likes, but usually a specific model tries to address some specific difficulty. Well, it seems to me that one of the main fights about “evolution” is that some (the darwinists) are convinced that everything must be explained in terms of known physical laws and deterministic mechanisms, while others (IDists) believe that living beings (or at least the information in them) can best be explained by the intervention of a designer. Please note that the intervention of the designer needs not (although it certainly can) violate physical laws, but it is absolutely necessary (according to ID) to explain the observable fact of the appearance of biological information. Then my problem is, what is the need for a third hypothesis which shares the difficulties of both the others? Because frontloading certainly postulates at least one intervention of the designer, at the beginning of life, and indeed such “intervention” should be “heavier” than is supposed by normal ID, having to explain not only the emergence of life, but also any future evolution of it. So, front-loading creates a problem for “naturalistic” thinkers as much as any other ID model. But, at the same time, it seems to postulate that the designer acted only once, and inside time, leaving all the rest to deterministic mechanisms. Again, I don’t understand: why? What is the advantage of thinking that way? What kind of facts are more easily explained that way? I can’t see any reason why a designer, who can impart information once, should not do that other times, or even continuously. And we know that the only “observable” facts which are brought up by darwinists are in essence omologies, either morphologic or genomic, and we know that omologies are, at best, evidence of common descent (or of reutilization of the code), and are never evidence of a specific mechanism. So again, why?
b) Second: how? It’s not enough to postulate something, you must also have a credible model for your hypothesis (unles you are a darwinist, of course…). So, what is the model? Was the extra information stored in the primordial genome of the first bacteria? How? How bigger was then that genome? How much extra information was necessary to “guide” evolution up to humans? It seems to me that here we are again in pure speculation, unsupported by any fact, but again maybe I am not aware of something.
Moreover, if many think that part of the extra information survives in non coding DNA, I see a very difficult contradiction here. I am the first to believe in the importance of non coding DNA. With Mattick and others, I am convinced that much of the “missing code” which can explain procedures and regulations must be there. But Mattick has shown that non coding DNA is the only “quantity” which constantly increases with the complexity of the species. Indeed, the ratio between non coding DNA and total DNA is the best quantitative marker of complexity in living beings, being extremely law in prokaryotes, and going up to 98-99% in humans. If non coding DNA were the repository of the initial extra information, I think we should observe the opposite.
In other words, while frontloading is a possible model, bypassing at least some of the difficulties of naturalism, still I find it unnecessarily complicated. Besides, unless a specific and credible model of this “extra information” is provided, I believe that the general ID arguments against the spontaneous generation of information by natural mechanisms still apply.
3) ID proper. This is obviously my favourite scenario. In this model, design is imparted to living beings by one or more designers, inside time, and during time, many times or continuosly. In this model, again, it is not specifically important if the imparting of information is implementing by an intelligent “manipulation” of known physical laws, like in the case of human designers, or violating them (miracle – creation acts), or by superior laws at present unknown to us. Or if it is implemented continuously or in a “punctuated” way. Personally I have always been attracted by a continuous model, mainly because I believe (not a scientific statement, this one) that God acts in the world practically always. But I must admit that the OOL problem and the cambrian explosion fossils strongly support a special intermittent implementation.