Evolution of complexity/information

Dembski's argument contains the concept of the 4th law of thermodynamics. He argues that for a 'closed system' CSI or complex specified information cannot increase. Although he does allow for a small increase he basically claims thatI(A&B) = I(A) Mod UCBI(A) = - log(2)P(A)
I(A): Information in event A
P(A): Probability of event A
UCB: Universal complexity bound I(UCB)<500 bits or P(UCB)<= 10^-150But there seem to be several examples which appear to contradict his claims.First of all Tom Schneider's Evolution of biological information shows how a simple mutation selection algorithm can increase the information in the genome.Adami as well shows in Evolution of biological complexity how selection/mutation acts like a Maxwell demon, increasing the information of the genome, but without violating any laws of thermodynamics.In fact, Dembski's argument applies to closed systems (sounds familiar 2nd law fans?)

The problem, when it comes to down to it, is that
'information' in the genome is an analogy, rather
than a fact.The genome does not contain the information to build an organism,
it contains templates for proteins that, when formed and
operating together, produce an organism.These are two very different things.

Your assessment,is dead on.Too bad Dembski was not as careful as the creator of the 2nd law in taking into consideration open and closed systems.I've been waiting for Dembski to address this issue for a while, instead of running around asserting he came up with a new law.I wonder what his "law" would make out of libraries?------------------
holmes

Peter, I'm not sure if your criticism is totally accurate. I'm not taking a definite position, but I'm currently leaning away from yours.Dembski talks about complexity and information as pretty much the same thing. While one can certainly call "information in the genome" an analogy, and can criticize Dembski's use of loaded terms (like information)to muddy the waters (it creates an air of equivocation), I don't think that automatically refutes what he is saying.There is entropy and order in chemical systems and this is essentially what he is talking about, just cleverly reclothing them in terms of complexity and information.In order for a complex molecule to be a template, and stay a template, there must be a certain order (or lack of entropy)in that system. Its ability to change into a new and workable template can be considered an increase in order (or a further loss of entropy). This requires energy, and of course not too much energy as well.Likewise one can also look at the entropy of the protein "system" itself in coordinated action to form an organism. There is a redistribution of energy and entropy, to build a more ordered system.So Dembski calls it "information." It can still be evaluated as an increase or decrease of whatever term he uses.I think his argument falls harder by realizing how close it really is to the 2nd law and that he failed to include concepts of open and closed systems.------------------
holmes

Peter I'd like a little clarification here. When you say that 'information' is an anaology are you saying it doesn't contain information at all? Or are you simply saying it doesn't contain [/b]all[/b] the information needed?If the former then please offer you definition of information.If the latter then please tell us where you think all the information is.

I view information to be that which an intelligent
agent obtains from data.You are assimilating a collection of data (the letters) as
you read this, and if you have been taught to read english,
and have sufficient context you will obtain information
from it (the meaning).In 'design' you can view the blue-print as data, and what
the 'engineer' (say) does with that blue print is to
interpret it to form an implementation. Give the same blue-print
to different engineers and you cannot guarantee that the
outcome will be identicle.It requires an intelligent, interpretive act.There is no 'information' in the genome, any more than there
is the 'information' within hydrogen and oxygen atoms on
how to make water, or to bond with carbon to make alcohol, or ...Within living things there are a set of chemical reactions going
on, the emergent property of that system of chemical reactions
is the organism.The 'information' on 'how to build an organism' doesn't exist.
The organism emerges from the complex set of chemical reactions
that go on.'Information' in the genome is an unfortunate consequence of
references to the 'genetic code'.

Doesn't the concept of an increase in 'order' presuppose
a direction or intent for the change?The problem with extending the analogy of information in
the genome is that one starts to think of 'increasing'
or 'decreasing' the information content.All that is happening in evolution is that the set of chemical
reactions available is being changed. Some changes provide the
system (organism) with a benefit in its environment, some don't.Some old greek bloke once said 'things change', and that's all
that evolution is about ... change. Concepts of information
or even complexity, and defintely of 'order' obfuscate the
simplicity of what is going on.There are a vast array of chemical reactions, and the way they
interact is so complex that the human brain cannot fully
encompass them (according to another thread we can never understand
them since we are a part of the same system ).If someone claimed that there was 'information' involved in
production of water and carbon dioxide from the combustion of
wood or coal would anyone listen?

And just to add one to your collection:Breivik, J, 2001 Self-Organization of Template-Replicating Polymers and the Spontaneous Rise of Genetic Information, Entropy 3:273-279

quote:

---

Living systems imply self-reproducing constructs capable of Darwinian evolution. How such dynamics can arise from undirected interactions between simple monomeric objects remains an open question. Here we circumvent difficulties related to the manipulation of chemical interactions, and present a system of ferromagnetic objects that self-organize into template-replicating polymers due to environmental fluctuations in temperature. Initially random sequences of monomers direct the formation of complementary sequences, and structural information is inherited from one structure to another. Selective replication of sequences occurs in dynamic interaction with the environment, and the system demonstrates the fundamental link between thermodynamics, information theory, and life science in an unprecedented manner.

---

Absolutely not. This is simply an evaluation of some relative characteristic that is measurable.For example you could base "order" on the number of different molecules found within a mixture (more molecule types meaning less order).Entropy as a concept in chemistry does not assume intent or direction, just a relative differences between molecular structures.And as it turns out this concept proves quite useful for chemists, especially in thermodynamics. I think the problem here is in equivocating between information and "information". There are certainly increasing or decreasing amounts of order, or potential for specific chemical reactions, in biochemical systems.One can call that order information, and it may very well be a handy analogy when talking about DNA as it is a storehouse of potential chemical reactions. However, you are completely right that in using that term it becomes easily confused with "information", by which we mean something that an intelligent being inputs, interprets, then acts on.That's one of my criticisms of ID as a whole. It is filled with unnecessarily loaded or misinterpretable terminology. I'm sure this is a calculated effort to win converts, and muddy the waters of debate, but it is not helpful to real discussion. Yes there would. Especially as computer-modeling of chemical systems increases, there is a very real identification of chemical systems as information. The problem, once again, is coming to believe that everything is information.Just as an economist could use economic analogies to understand chemical equations, and driving forces, it would be an error to come to view chemistry as a form of economics.So goes it for the computational-philosophic-mathematical information theorists. Dembski has either lost the reality that he is modeling a real world system and information was just a useful term, or he finds that term useful to obfuscate.Without that mistake, "information content" of chemical systems is a neutral assessment.------------------
holmes

Well, then pick another word. I think "meaning" is the closest simple English word that might do.When complexity and information content is being discussed there is a very well defined term called "information".Your agrument is that there is no 'information' in any sense of the word involved in the process of the genome being used as a basis for building a cell. I guess that, at least, blows away the Dembski's arguments against evolution or abiogenesis based on the his idea that the information increase is impossible.There is however, information there in the information theoretic sense. I don't think it helps anything if you make up new meanings that know one else uses.

I am still looking for a better discusion of the TWO laws than Wolfram's simply physical notion that optimality ONLY applies to averages of molecular motions and not to the free path's themselves. Till then it seems one simply need indicate the MILTARY assocation of the gene coding concepts and indicate that action a distance (even if one does not want as with Wolfram to consider what Collet did not understand that Thom thought was wrong about US physics decimal placing places...) As for KELVIN temps and the 2nd law that is probably a specious c/e attempt to decieve the darling clemintine.

The definition of information that I use isn't one that
I have made up ... it's the result of an extensive literature
search in the field of information theory.There has been, typically, an poor usage of the word 'information'
even within the field itself, and many have argued for a pinning
down of it's meaning. The one that I use is a concensus of
various researchers including Peter Checkland from the 'soft
systems thinking' camp.'Meaning' is something different.If I write:T697AOKIt's just alpha numeric characters ... except that to anyone in
England it brings vehicle license plates to mind, and for
myself a particular nice CBR600 motor-cycle.The reader makes associations based upon their own context to glean
information from data. The same reader/observer may obtain
different information from the same data when viewed at different
times.In the main, older works do not differentiate between data and
information.Perhaps it is simply misleading to use the term 'information' if
it has different meanings in different areas of study.You say there is a well-defined term 'information' that is used,
could you please post (or link to) that definition, since that
may be the root of my objection to it's use in reference to
genomes.

If order can be arbitrarily defined how is it a helpful
concept?I think I agree with you as far as the rest of your comments
go.Basically my position is that the term 'information' when used
by anti-evolutionists is used in the more ethereal sense than in
any mathematical sense. They then apply, to an analogy, processing
which is only consistent with 'actual' information, rather than
the 'sort-of' information that genomes actually appear to contain.I wonder if a creationist would use 2LOT to claim that the
oceans don't exist

And what don't you like about Shannon's defintion? And how does that definition relate to yours.?

Just because something is arbitrarily defined in general, does not make it less helpful as long as it is consistently defined within the model you are creating.Of course any model is more helpful if it uses common definitions.To be honest though, what is there in science which is not defined arbitrarily? There is no absolute anything, ESPECIALLY in biology. One arbitrarily picks a feature or some measurable quality as a starting point and uses it as a reference point.Well you can always be a bit picky to choose a more useful starting reference point, so its not just "eeny-meeny-miney-mo arbitrary", but it is still arbitrary in the grand scheme of things.I agree that anti-evo use of the term information is ethereal, but it doesn't start that way. Dembski starts by using a real, defined term. It's just that by then the end of his sermonizing he has equivocated on that original term, and its easy for anti-evo forces to buy into the equivocation. From that point on its consistently used in its ethereal sense.People familiar with computer modeling probably won't be as fazed by its use as a term, as long as it is consistent.------------------
holmes