Abiogenesis

Post Reply New Topic

Meta, before I give a detailed answer I need to know one more thing. You wrote before:

Unless the references I gave you change your opinion, it appears you argue only intelligence can generate hierarchical modular organization. Do you also think that life and intelligence need hierarchical modular organization? Is it possible to have life and/or intelligence without that?

Is it just me or is this (very interesting) discussion asking a lot of time/energy/attention/focus?
I haven't read them all but until now I haven't read anything substantial with regard to our arguments. A lot of unproven/unprovable assumptions, not much fundamental evidence, just idle speculation pretending to be science.
That's not really my argument... My argument: The only proven source of a hierarchical modular organization(HMO) is an intelligence.
I don't know. Any answer would be purely speculation.

All I know is that (1) all known examples of life have a HMO and that (2) the only known source of HMO is an intelligence? Given available evidence I can't explain the origin of either life* or intelligence*; but the evidence with regard to life as we know it indicates that this implementation of life has been designed and constructed by an [unknown] intelligence. This implementation of life is a technological artifact, not a natural phenomenon.

* an hypothetical category, the term interpreted independently from the (at the moment) only know implementation.

I don't have a scientific answer, only unscientific speculations: Maybe that intelligent life has always existed? (panspermia?) Maybe there are lifeforms/aliens which have no HMO? (e.g. A hyperintelligent share of the colour blue? (just joking)). Maybe the universe itself is intelligent? Who knows? Something no one has considered yet? Given the evidence available, at the moment it's outside of the domain of science: There is no scientific answer to your question

Question. Imagine that I where to genetically modify (by a secret protocol of my own invention) a plant or animal. An without telling anyone I would set it free in the wild. If now one of its decedents where to be captured and genetically sequenced by an evolutionary biologist. Would this biologist be able to identify intelligent design (my genetical modification) or would said biologist assume that this genetic material must have evolved by itself? How would you make the distinction between the two? Note that genetic modification is just an example of horizontal gene transfer...

Really? Would you care to enlighten me? Nice try but my objection to the above statement has nothing to do with organic chemistry. It's a plain non sequitur. This is what Dr Sarfati said, word for word:



Now contrast this with the following:

"If NASA engineers did design spacesuits for the Apollo astronauts to survive on the moon, guess what it would prove. It would prove that intelligence was necessary for humans breathe. Most people would think that because the spacesuits only recreated the necessary conditions on Earth, it would not require an intelligence to breathe. As though the NASA engineers weren't intelligent"

So from the same argument, it requires an intelligent designer to intervene in order for us to breathe oxygen on Earth.



I know what Stanley Miller's experiment is about. In any case, Sarfati's argument that both kinds amino acids always being produced is false. Although initially the experiments always produced both left and right handed amino acids, it's recently been shown that some processes can produce an excess of one of them:

http://www.rsc.org/publishing/journals/CC/article.asp?doi=b409941a

Leslie Orgel's last essay The Implausibility of Metabolic Cycles on the Prebiotic Earth points out some very interesting problems that need to be solved/proven. Most of these problems are currently ignored. Much OoL research is speculative/fringe science, or as he puts it in final line of his conclusion:

I didn't say it was in itself evidence for a non-intelligent origin. I say it is a plausible alternative to your claim that "similar results imply similar origins". You can get similar results even from different origins.


The non-intelligent process in question is variation and selection. The first example is one you probably are familiar with. Compare the body shapes of a greyhound, a dachshund, a bulldog. They weren't designed to be these shapes. They got those shapes through variation and selection.

You might object that this is selection by intelligent agents. You'd have to show that this matters, and you'd have to show how selection by intelligent agents explains the observations of Peter and Rosemary Grant in their work on Darwin finches in the Galapagos. They could see the shapes of the beaks changing across generations in response to selection pressures in just a few decades of data collection. I see form following function to be well supported by observation. If you want to explain that by appeal to the intervention of an intelligent designer, that designer must have been operating on Galapagos in the last few decades without the knowledge of any of the zoologists who study those finches for several months every year, during their breeding season.

I also expected that you would see this is a trivial problem. That's one reason I chose wing shape, as an example so obvious that we wouldn't need to discuss that. Aspect ratio, wing loading and profile have easily quantifiable effects on performance measures like glide angle or sink rate. Do you really think selection would have trouble finding a combination of the three parameters that is at least close to optimal?

You are keen on observation and evidence. That technology borrows from biology is a known fact. An intelligent agent making a life form has not yet been observed.


What makes you believe that? Do you think there is only one path to any outcome? There is some fundamental assumption you make that I just don't get. You appear to think this is obvious. I don't see how you can even justify it.


By the standards generally applies in science, I can show that evolution is a far better theory than ID. One piece of evidence I offered you is that intelligent reuse should be accompanied by transferring best solutions across taxa. You claimed unknown constraints and tried to argue that one example really is optimal:
I already mentioned that if you go down this path, the price is that ID becomes untestable. If you see something clever, you say "that is intelligent design". If you see something that looks dumb, you say "this design is so intelligent, we don't even understand it! That is really intelligent design". And you can find things in this list that are really dumb, like the laryngeal nerve of the giraffe. It is easily explained by evolutionary theory, but it would take some ingenuity to show that it is intelligent design.

You tried with the eye:
Cephalopods are molluscs. They do have blood, and last time I looked it up I read they do have blood vessels in their eyes. Using hemocyanin instead of hemoglobin, they don't transport oxygen so well. If having the blood vessels in front of the photo receptors is so good for getting oxygen to the photo receptors, cephalopods should have a greater need for that intelligent design feature than vertebrates.


Use of blood to transport oxygen is related to body size, not to brightness of the habitat.


No. You didn't know enough about the biology to know the relevant design parameters. Putting those in and assuming best design practice still leads to a prediction from ID that is in conflict with the data, even if I accept what you say about the benefits of having blood vessels on top of photo receptors. You can only make ID fit the evidence by postulating the designer does not follow best practice.


Here is a prediction that follows from evolutionary theory and developmental biology:

Data on sea urchin larva development, reviewed in the same paper, support that prediction. Intelligent reuse of design features, combined with the modularity you present as evidence in favour of intelligent design, should also come with massive horizontal transfer of not just individual genes but whole gene complexes. That would mean the same phenotype in different taxa would be made by the same developmental pathway and the same set of genes, no matter the phylogenetic distance between the taxa. That prediction is not supported.


Short connections between neurons reduce the costs of signal transduction and increase speed of computation. The more you need to integrate information from different sensors, the more it pays to have those interconnected neurons close together. You get brains instead of nerve nets distributed over the whole body. The brain is an organ. Cost and speed of computation are features of the phenotype.


Modular design is then a sign of limited intelligence? You often write that evolution should produce non-modular and incomprehensible designs. If you then assume a very intelligent designer, you can adjust ID to predict that. Not very testable.

In response to my references you wrote:
The one I read all the way through described the result of a simulation. They did not take already available data and made them fit into a hypothesis, they did an experiment. Set up variable phenotypes, apply selection pressure, see what comes out of it. That happened to be modular.


Look at the wings of birds, bats and pterosaurs. At the level of function, this is convergence, because the most parsimonious phylogeny does not include a last common ancestor that flew. All three groups use the forelimb as wing. That aspect is parallelism. They took the same structure that all of them had and modified it in similar ways. All of them made the forelimb longer.

You could also look at the genetics and the development, to see whether similar phenotypes are made in the same way.

You should find more detailed explanations in any introductory textbook on evolution. "Introductory" means undergraduate level. If you only ever heard of evolution in school, especially some American schools, you may not have heard of this before.


Let's take hemoglobin and hemocyanin. Hemoglobin is simply better at transporting oxygen. Why is hemocyanin still around, even in very active animals that are limited by oxygen availability?

Bird lungs are more efficient than mammal lungs. They get more oxygen into the blood for less energy. Modern humans have existed for only about 200 000 years. Birds have been around for much longer. Why haven't we got that far superior lung design?


If you had several different pieces of code doing exactly the same job, but some of them fast and with low memory usage, others slow and with high memory usage, would you use the bloated piece of old junk? Your argument about reuse instead of convergence assumes that solutions can be plugged in like a piece of code. If you have a clearly superior option, tried and tested, why not use it?


Hemocyanin and mammalian lungs are not cases of trying something new. They are cases of sticking with what your ancestors had. Human engineering students would be able to tell very easily which lung would work better. Simple lab testing would show which molecule transports oxygen better. Any designer who can design whole life forms should be far more competent than that.


But it does. It shows that adding time makes a difference. That is the point I responded to.


It is a direct response to what you wrote, quoted above. You say adding time doesn't matter. I can almost guarantee getting ten 1s in a row from a random selection of 0s and 1s with probability of 0.5. All I do is either run the program until I get that sequence, or run enough copies of the program in parallel, or both. If I run one program once and let it produce a string of just ten digits, the probability of all ten digits being 1 is 2^-10 = 1/1024. If I run enough programs in parallel for long enough, the probability of getting ten 1s in a row at some time can be as close to 1 as I want.

We are discussing probabilistic events. How often I flip the coin does matter.

Last edited by Gromit on 08 Dec 2009, 5:08 pm, edited 1 time in total.

Wrong. In the initial argument, it is "create life" for antecedent and "create life" for consequent. In your parody, you equivocated "design spacesuits" with "breathe".

How about starting with "If NASA engineers did design spacesuits for the Apollo astronauts to survive on the moon, guess what it would prove. It would prove that intelligence was necessary for humans to design spacesuits."?

If you want to use that to make predictions, try this: How often was intelligent intervention necessary in the lineage from the first bacteria to humans, bats, dolphins, beetles, octopus? That should give you lots of phenotypes, changes in hierarchical modular organization, etc. There's the cell nucleus, multicelluarity, dividing tissues into organs, adding neurons and hormones for communication between cells, different developmental pathways, new sensory systems, sometimes new organs when invading a new habitat. Unless you assume that all information was front loaded into the very first life forms, you must assume additions to the hierarchy. If these come from intelligent intervention, you can work out how often that happens. If you believe that form following function is a sign of intelligent intervention, that gives you more data. How often does the designer intervene?


If you use a virus to transfer one gene, then you mimic what occurs naturally. Even then, if you inserted green fluorescent protein into an orchid from the Andes, there would be suspicion. Where is the pathogen that infects both orchids in the Andes and jellyfish and could transfer genes?

But that is not designing a species. You disputed convergent evolution. What you are talking about in the context of intelligent reuse of design features is mix and match. If you mix and match chunks of genes from widely divergent taxa, that will definitely be identifiable. If intelligent reuse of elements of phenotypes and genotypes were responsible for producing the life we see, we couldn't reconstruct phylogenies at all because phenotypes and genotypes would depend only on environment, not on ancestry. No one would have tried to explain the origin of species by evolution through natural and sexual selection.

True. So let's try a different substitution. Substitute "life" with "water from combining hydrogen with oxygen".



Here are a few more substitutions:
vapour by heating water
salt crystals by evaporating sea water
convection
combustion
bubbles

If we also substitute "laboratory" for "test tube", we get more possibilities:
turbulence
lightning
microwaves (or anything else on the EM spectrum)
magnetic fields
electric fields

I think all these can happen without intelligent intervention.

Have I made a mistake or is Sarfati's argument not all that strong?

But you never actually prove that point, do you? You would be right if and only if you could indeed present similar (non-trivial) results from different origins.

You can't point to life or biology because its origin is the one which is in dispute. So what other example do you have?
I did not say that a process of variation-and-selection was useless? An intelligent designer can use a process of variation and selection to build a system which automatically reconfigures itself for optimal fitness.

The point is that lacking hmo, a process of variation and selection will never generate hmo because unlike an intelligent designer, a process of variation and selection does not require hmo.

Darwin's finches are indeed an interesting example of how variation and selection as a system of automatically reconfiguration. With the changing weather, the distribution if the beek size/shape changes. The intelligence here is not external, but embedded as part of the design.

Like you I accept that variation-and-selection takes place. Where we have different opinions about is the range of what such a process can generate.

Even a very intelligent designer is limited in the above sense. The two are not exclusive: The hard limitations of intelligence are real and absolute, but there is some variation between intelligences among humans below these hard limitations.

I would like to point out that I understand variation-and-selection-processes of all kinds much better then all the details of biology. I don't have an answer to every possible biological detail. I do know the limitations of a variation-and-selection process in general, which even biological systems most obey.

It wasn't even the main argument. However, creating life in a laboratory is far from being as simple as setting up the right conditions and allowing it to happen on its own while you just watch. If it were that easy, then it could be classifiable with such other phenomena. As it is, you have to constantly interfere to the point where you are basically engineering it. Such things as water will break down organic molecules, and also you need to produce so many hundred of components and have them combine correctly just to produce simple parts of as simple cell, so after guiding the reactions thus far it has to be guided even further. In that way, it requires intelligence to create life in the laboratory, so it is not classifiable among the other things you listed.

Biology is engineering -- Daniel Dennett (clearly not a creationist or IDist)

(1) There’s No Such Thing as a ‘Simple’ Organism, (2) All known metabolism is cybernetic--that is, it is programmatically and algorithmically organized and controlled.

All known metabolism is cybernetic means that it requires intelligent control to maintain life in living cells. This intelligence is embedded as complicated algorithms.

You can't just expect chemistry to do something useful. Instead control is needed so that some chemically very unlikely processes will happen at the right moment.

This means: Chance and necessity do not explain the origin of life

The key question seems to be "Is life (as we know it) natural or artificial?" Each paradigm will have its own logical answer, are we however able to provide an empirical answer?

To quote Niels Bohr: "Life is consistent with, but underivable from physics and chemistry”

Biological control systems are homeostatic (they work on negative feedback) and can be accounted for by ordinary natural physical laws.

Mysterious Woo is not required.

ruveyn

It looks like an argument but if it is I don't understand it.

How I understand what you wrote:
Living organisms are able to maintain homeostasis without breaking any physical laws.

With other words: Life is consistent with physics and chemistry...

I don't see any connection to any of the arguments that I made with regard to the origin of life.

No one claimed that an external intelligence would be required for maintenance, the embedded intelligence is very capable without. The argument is not about maintenance but about origin.

I propose that living material come from non-living material by purely natural/physical processes. In the beginning there was nothing living, then somewhere along the line a series of events caused some non-living material to be able able to replicate. The rest is history.

If you object to this on the grounds that this supposes something come from nothing, I answer that the God hypothesis has the same flaw. Why have an extra step? Just assume there natural cosmos occurred and living material emerged from that.

ruveyn

ruveyn

This isn't "something com[ing] from nothing", but rather proposing that molecules which break down very rapidly and have equilibria set against their production, that these molecules, would not only hang around long enough to combine, but combine multiple times without breaking even though for each additional unit to the chain there is a multiplicative chance of breaking and becoming even more irrelevant. Forget learning organic chemistry, just mix everything and hope for the best.

The Origin of Postings:

A CMI feedback related to this,