Bits and Pieces of an RNA World
by BradfordZachriel linked to an article entitled Exiting an RNA world; Nature Structural Biology 7, 5 – 7 (2000) doi:10.1038/71194; authored by Paul Schimmel & Shana O. Kelley. Snippets from the article are reproduced here for the purpose of commentary. You can read the full article as linked. From the article:
"In contemporary biology the connection between the RNA world and proteins is made by the aminoacylation of transfer RNAs. In this reaction, the rules of the genetic code are established by virtue of linking each amino acid with the tRNA that bears the cognate anticodon triplet. The code in turn makes template-directed protein synthesis possible."
Amino acid encoding codons have no value in the absence of a means of translating transcribed messages. Translation is effected through tRNA which links specific amino acids with appropriate anticodons attached to tRNA.
"Given the results obtained thus far, further selections and tinkering will probably yield substantial improvements "” molecules that come closer to having the catalytic activity of modern protein-based aminoacyl-tRNA synthetases. Of course, an additional challenge remains in the attainment of RNA-based amino acid activation reactions because the availability of reactive amino acid esters was likely to be low in the aqueous environment of the proposed RNA world."
Researchers have "tinkered" to find RNA catalysts that would mimick the function of aminoacyl-tRNA synthetases. Those protein catalysts are involved in attaching the correct amino acids to tRNA; a function that needs accounting for in a theorized RNA world. We are reminded of the importance of an adaquate supply of twenty amino acids- available as needed.
"In the RNA world, ribozymes that catalyzed aminoacylations may have progressed through a succession of intermediates (Fig. 2). These intermediates could have included those with an enhanced efficiency because they were aminoacylated."
OOL papers generally have at least one part like the foregoing where authors speculate as to what might have been and then state their views as to what was likely. An argument is made for a selection process. The RNA world commences with an RNA self-replicator. An indicator of the difficulty in securing supporting data is evident by a lack of details explaining the conditions in which a succession of intermediates is thought to have evolved. Presumably there is some sort of genome but what functions are specified by it and how is replication accomplished. Do the biological structures exist within a cellular membrane and if so what structures would be included within its boundaries. What process leads from a self-replicator to tRNA groups? Why should we assume prebiotic dynamics proceed in this direction?
The weakest part of the case for an RNA world lies in the theory behind it. The RNA world often invokes a selection process without specifying details as to replicating entities and a directional indicator. Why would enhanced efficiency be a determining factor indicating the outcome of a chemical process in the absence of some precursor cell? If such a cell existed where are the steps delineating the process of getting there? Why not instead describe the likelihood of theorized chemical reactions in terms of free energy of activation?
May 10th, 2007 at 8:09 am
Hi Bradford,
YEAH! Let's do science!
Boy, you are fast, I barely got through skimming the articles and I find you have already posted your rebuttal.
You wrote…
Followed by…
My knee-jerk reaction is an interest in hearing an alternative, One that makes a stronger case.
This shouldn't be too surprising coming from a fellow IDist.
I need more time to digest before I can comment beyond my knee-jerk reactions.
Regards,
TP
Comment by Thought Provoker — May 10, 2007 @ 8:09 am
May 10th, 2007 at 2:18 pm
The weakest part of the case for an RNA world lies in the theory behind it.
TP: My knee-jerk reaction is an interest in hearing an alternative, One that makes a stronger case.
The present case is analogy reliant but without substance. OOL asserts, rather than demonstrates, natural selection as guiding prebiotic chemical reactions. It asserts that a genetic code evolved without demonstrating a step by step selection process. OOL asserts that self-replicators evolve in the direction of cells without the empirical data to indicate this. There is a failure to link a precellular process to evolutionary selection. If analogy is useful, the better one is an analogy linking the development of a coding system to an intelligent source. This analogy, unlike the prebiotic chemical one, is able to cite many documented cases.
Comment by Bradford — May 10, 2007 @ 2:18 pm
May 10th, 2007 at 2:37 pm
Bradford:
I think fairness requires that you now demonstrate or at least attempt to demonstrate a step by step design process for the first living creatures. Did the designer create self-replicating molecules first? And then what happened? Or did he just poof the first cells into existence? Scientists that attempt to model OOL postulate explicit pathways from the "ursuppe" to replicators to protocells to genetic code etc. Perhaps they are wrong, but at least they stick out their necks and try to devise testable hypotheses.
Comment by Raevmo — May 10, 2007 @ 2:37 pm
May 10th, 2007 at 2:54 pm
When I was going to school I used to work in construction building homes. I learned that there were many different variations, blueprints and ways of going about building even two similar homes. One of the hallmarks of design is creativity. You do not poof a home into existence or a code either. But neither is there a deteminism to it. Many differing approaches to the same outcome is characteristic of intelligent solutions.
One thing a designer would have to do at the outset is provide mechanisms that safeguard the integrity of genomes. Corrupting factors are ubiquitous and mitigate against gradual adapting mechanisms.
Comment by Bradford — May 10, 2007 @ 2:54 pm
May 10th, 2007 at 6:10 pm
I'd be very interested in knowing the location of papers that give testable hypotheses for the creation of the genome. Does the above paper provide that? Are there others that do? Anyone else want to add some? Thanks in advance!
In all honesty, I haven't had the time lately to even keep up with my responses to posted questions I've been asked, let alone do much supplementary reading. So I'd like all the more to focus on those papers that actually stick their necks out in a testable, falsifiable way concerning a natural process origin to the genetic code.
In particular, I don't have time to read many hand-waving, then the genome is acquired just-so stories. So I'm really glad to hear there are some out there that go beyond this, e.g. to specify a plausible sequence for the construction of mechanisms for encoding, storage, retrieval, and decoding. I'm very interested to see how they navigate those waters, given the mutual dependencies of those mechanisms.
Comment by eric — May 10, 2007 @ 6:10 pm
May 10th, 2007 at 6:21 pm
Tinkered. Appropriate word. But is abiogenetic research based on faulty premises? Let me repost a previous comment.
According to the Abiogenetic Hypothesis, life began as simple self-replicating molecules, then specialization of functions occurred driving further evolution. The discovery of self-replicating RNA is an important confirmation of the basic idea.
Let's consider a prediction from the Abiogenetic Hypothesis. In extant biological systems, the aminoacylation of tRNA is catalyzed by protein enzymes. However, a reasonable prediction from the Hypothesis is that RNA may have previously performed the functions of these modern protein enzymes. And catalytic RNA has been shown to be capable of acting as a aminoacyl-tRNA synthetase.
Does this "prove" abiogenesis". Of course not. However, it does show that the Abiogenetic Hypothesis is capable of generating fruitful hypotheses, and that further tinkering may yield more insights.
Self-replicating RNA are capable of evolution in vitro, becoming more precise in their copy mechanisms, and more efficient at acquiring local resources.
You ask some excellent questions. Only additional investigations will reveal the answers.
Comment by Zachriel — May 10, 2007 @ 6:21 pm
May 10th, 2007 at 6:29 pm
About "replication", it is worth noting the profound difference between replication Before Language (B.L.) and After Language (A.L.).
Imagine some mechanism X. Purely for illustration purposes, imagine a functioning mechanical clock or some other mechanism. It is required to replicate X. How can it be done?
After Language:
Copy the symbolic construction information needed for building X. Then use the duplicated information and raw parts to construct a new X.
Before Language:
There is no construction information on how to build X because information does not exist yet. Therefore the only guide for building another X is to actually copy from a working X itself, using the original as a template to guide the process, step by step.
Now if the original is just a single macromolecule do nothing much more than floating around in the soup, then it may be a comparitively uncomplicated matter to make use of it as a template. Perhaps a peer copy might serve as a catalyst.
But if X were some mechanism of multiple varied parts fulfilling some function (e.g. one of the functions necessary to creating language), it is not so simple. The simple peer-catalyst approach will likely break down at some point.
There is, for example, the sheer physical problem of access. If you actually had to do this with a clock, how do you get at all the gears and such without taking it apart? To use the originals as templates, physical access is needed. (There are no "eyes" to just look.) Then what becomes of the performance of the function, if replication depends on taking the original apart.
Clearly there are severe constraints on replication B.L. How then does this kind of replication serve as a sufficient means to creating all the mechanisms essential to a functional language?
p.s. Please, no pedantic responses on whether we actually need a "clock". Please listen to the point being made and address the matter of the limitations on replication B.L. and/or whether that is reasonably adequate to construct an information processing system.
Comment by eric — May 10, 2007 @ 6:29 pm
May 10th, 2007 at 6:35 pm
Genomes are the result of long processes of ad hoc evolutionary change. Most any genetics journal will provide voluminous examples of testable predictions.
There is no complete theory of the origin of the genetic code. That is not to say that nothing is known. Several plausible (albeit incomplete) pathways have been proposed.
Comment by Zachriel — May 10, 2007 @ 6:35 pm
May 10th, 2007 at 6:46 pm
RNA replicators do exist, after all.
Quite possibly. But remember, there is a lot of evolutionary change involved. RNA replicators probably didn't exist in isolation, but in chemical aggregates. The complexity may make it hard to think about, but may resolve many of the technical issues as well. Each molecule may specialize to perform important functions to the community. Protein synthesis probably started very early.
Of course, this is speculation. However, as you are trying to argue there is some sort of barrier, it is only necessary to show that there is some plausible history.
Comment by Zachriel — May 10, 2007 @ 6:46 pm
May 10th, 2007 at 7:02 pm
That's not all that's necessary. What is needed is some reason to believe reactions would produce the observed effect of a cell. What is it you learned about organic chemistry that assures you of the answer?
Comment by Bradford — May 10, 2007 @ 7:02 pm
May 10th, 2007 at 7:06 pm
To be more clear than I was, I'm not interested in the subsequent history of genomes after you have a working information processing system. I'm specifically interested in how an unguided natural process builds such a system in the first place.
Since a system to process information cannot function without all the necessary mechanisms, I would consider "plausible" and "incomplete" to not fit well together. There could be any number of blind alleys that could never reach the destination and only lead to (literally) dead ends.
Nevertheless, if you or someone else could give us a clear and accessible description of just how far one gets with some particular examples of these incomplete pathways, that could be educational.
Comment by eric — May 10, 2007 @ 7:06 pm
May 10th, 2007 at 7:11 pm
To answer the specific claim, that is all that is required. However, I am open to discussion.
What do you mean by "observed effect of a cell" If you mean "cell", there are a variety of segregating structures, including lipid membranes which can form spontaneously.
I have not claimed a valid theory of abiogenesis. Quite the contrary.
Comment by Zachriel — May 10, 2007 @ 7:11 pm
May 10th, 2007 at 7:18 pm
There could be multiple points of difficulty, and Bradford is alluding to some besides what I have mentioned.
Speaking for myself, the "some sort of barrier" is the Language Barrier. It has huge significance, not the least of which is the point I made above about what kinds of replication are possible after vs. before.
At a sheer minimum, a plausible account of the emergence of information processing would have to provide a plausible pathway for the construction of mechanisms for encoding, storage, retrieval, and decoding. This includes specifying a sequence of construction for these mutually dependent mechanisms that is plausible, despite the facts that
a) the mechanisms must be synchronized to work with each other (decoding must agree with encoding; storage must be consistent with retrieval),
and
b) each mechanism is useless and meaningless when taken alone,
and
c) the mechanisms need to be built using only Before Language replication (no hand waving blind leaps about how "mechanisms reproduce" without information).
p.s. Oops, forgot
d) Information itself is non-functional. It doesn't have the properties it describes, which makes any type of "selection" problematic.
Comment by eric — May 10, 2007 @ 7:18 pm
May 10th, 2007 at 7:21 pm
Quibble alert. Feel free to ignore.
I think it is plausible to get from New York to San Francisco by taking I-80W, but I may not be able to provide you complete directions.
Sure. Maybe they closed the highway. That doesn't mean the trip isn't plausible (superficially fair, reasonable).
Comment by Zachriel — May 10, 2007 @ 7:21 pm
May 10th, 2007 at 7:56 pm
Yes, lipid chains can form spontaneously but they hardly resemble cellular lipid membranes in which embedded proteins enable vital cellular functions. Not only do cellular structures not form spontaneously, the specificity required in order to enable interaction between them is absent in causal generating scenarios. A nucleic acid is useless unless its sequences enable functional coding and transcription/translation mechanisms exist. A lack of causal specificity is a problem for OOL.
Comment by Bradford — May 10, 2007 @ 7:56 pm
May 10th, 2007 at 8:02 pm
Hi Zachriel. The other day I was accused of being Eric by a commenter. Not that that is insulting. To the contrary Eric appears to be an intelligent and discerning commenter. Today I notice you attributing my quotes to Mike Gene. Another complementary confusion however the three of us are separate individuals.
Comment by Bradford — May 10, 2007 @ 8:02 pm
May 10th, 2007 at 8:06 pm
The weakest part of the case for an RNA world lies in the theory behind it. The RNA world often invokes a selection process without specifying details as to replicating entities and a directional indicator.
The evolution involves artificial selection but even if it were natural how do more efficient SRMs lead to precursor cells since the latter's properties entail, not separate replicators, but interacting parts?
Comment by Bradford — May 10, 2007 @ 8:06 pm
May 10th, 2007 at 9:36 pm
Oops! Quite so. It was a continuation of the previous thread by Mike Gene. If you can edit the comments, please do; then feel free to delete this comment. Otherwise, I don't think the content is affected by the misattribution.
Comment by Zachriel — May 10, 2007 @ 9:36 pm
May 10th, 2007 at 9:51 pm
No one said they did. Lipid membranes merely help provide segregation.
Hypothetically, through increasing division of function and ensuing specialization. I already mentioned this several times. RNA might evolve various enzymatic functions, including acting as aminoacyl-tRNA synthetase.
Sure. That's why no one claims to have a complete theory of abiogenesis. That's why scientists continue to investigate and collect evidence.
Comment by Zachriel — May 10, 2007 @ 9:51 pm
May 10th, 2007 at 10:03 pm
Zachriel, I substituted my name for Mike's.
Comment by Bradford — May 10, 2007 @ 10:03 pm
May 11th, 2007 at 5:40 pm
True. OTOH, I think it is implausible to get to the moon from the earth using a hot air balloon. The inability to provide complete directions in that case is an inherent consequence of the fact that it fundamentally cannot work. Time spent messing with different ropes, different balloons, different heaters, or different starting places would all be irrelevant to the fundamental barrier.
The origin of Language and semantic meaning is fundamentally unlike anything ever discovered in the study of natural chemical processes. Just as hot air wouldn't get us to the moon, information isn't a chemistry problem.
Regarding plausibility, here is a simple reality check. There are only 24 possible sequences of four objects. Considering the construction of Encoding, Storage, Retrieval, and Decoding mechanisms, no sequence of construction has plausibility. Every sequence involves the construction of mechanisms that are meaningless in isolation when considered in a prelanguage, preinformation universe.
The idea that a sufficiently complex mechanism could be replicated by Before Language replication seems also very implausible on the face of it.
Treated as a scientific hypothesis, the burden of demonstrating plausibility rests upon the proposal that there is a way across these identifiable obstacles. So far, that burden has not been met. Ergo, it remains implausible until further developments can show otherwise. The obstacles are known, many, and considerable.
So far, we don't have any scientific reason to believe that nature has the capability of crossing the Language Barrier.
Comment by eric — May 11, 2007 @ 5:40 pm
May 11th, 2007 at 6:43 pm
Then it becomes an empirical question"”as it should. And we already know quite a lot.
To continue the analogy. We already know the route (I-80) that connects New Jersey to San Francisco. You just have to find your way along the treacherous roads from New York to Teaneck, NJ.
The Abiogenetic Hypothesis has been fruitful and leads to many falsifiable scientific observations. Do you understand why that is so important in science? If you were really interested in determining how life originated, these are exactly the sorts of scientific observations that would interest you.
Comment by Zachriel — May 11, 2007 @ 6:43 pm
May 11th, 2007 at 6:57 pm
Apparently it is, as self-replicating molecules can evolve.
You keep saying that, so either you are confused on the evidence or you are confused on the meaning of "plausibility". RNA can act as an enzyme, as a synthetase. As this property is observed in vitro, it is plausible it happened in nature (though not certain by any means). And as this function was considered a plausible consequence of Exiting an RNA World, this is confirmatory evidence of exactly the sort of process you claim is not possible.
Comment by Zachriel — May 11, 2007 @ 6:57 pm
May 11th, 2007 at 6:58 pm
Hi Zachriel,
You wrote…
I know it just pushes back the problem and doesn't really detract much from the argument that you are trying to make, but are you sure of this? The Origin of Life timeline keeps moving closer and closer to the origin of Earth itself. Is it possible that the two (Earth and life) came into existance at the same time?
Just a thought and I don't have anything to back it up. I was just something that stood out as not necessarily an absolute.
Provoking Thought
Comment by Thought Provoker — May 11, 2007 @ 6:58 pm
May 11th, 2007 at 7:31 pm
eric,
Based on your last comment, would it be fair to characterize your skepticism as resting not on the technical details of any particular hypothetical pathway leading to a coding mechanism, but more on what you consider the general impossibility of getting from non-meaning to meaning through gradual change?
If so, I have some thoughts on this, though I won't be able to post them for the next several hours.
Comment by keiths — May 11, 2007 @ 7:31 pm
May 11th, 2007 at 8:38 pm
Zachriel:
I'd like to see references for this. The studies with which I am familiar, relating to self-replicators, begin with RNA, not formed as a result of natural reactions in what were reasonably believed to resemble prebiotic conditions, but rather were selected through intelligent input by researchers. Bear in mind that initial conditions are critical to the case for plausibility as needing to be accounted for are, not just initial SRMs but, also a supply of the constituent compounds found in RNA. Those are not your garden variety chemicals on a lifeless planet. Still the biggest obstacle to life is not the formation of biochemical compounds, as formidable an obstacle as that is. What needs to be explained is why nucleic acids would acquire the sequence specificity that relates to the function of basic universal catalysts one would expect to find in a precursor cell. Natural selection does not explain this in an extra-cellular environment.
Comment by Bradford — May 11, 2007 @ 8:38 pm
May 11th, 2007 at 9:48 pm
Zachriel Says: In response to my statement that the problem of information isn't a chemistry problem, you said
You appear to be missing the fact that a replicating molecule is not language processing.
As I did before, I will readily grant for the sake of discussion a world full of peer-replicating RNA molecules. That is not "the sort of process [I] claim is not possible." (although that in itself may also be impossible for its own reasons — more separately). So it seems I haven't yet succeeded in conveying to you what I am looking for.
The barrier is about the original appearance of a system to process symbolicly coded information.
Replication is not the same thing at all. With replication, like copies like, possibly with some variations along the way. You get more of what you had.
With information processing into and out of a language, some reality or meaning (e.g. a specific sequence of amino acids) is encoded into an alternate symbolic form (e.g. codons made of nucleic acids), and later converted back again (i.e. decoded) into its original meaning. The symbolic representation is not like the reality it represents.
That is why it is crucial to recognize that a code maps between symbols and a meaning. The symbols and the meaning are distinct and the association between them is not required or inherent. It is not irrelevant that there are different genetic codes, since this shows that the mapping between codons and amino acids (or stop symbols) is not required or inherent.
Comment by eric — May 11, 2007 @ 9:48 pm
May 11th, 2007 at 9:54 pm
Hi Bradford,
You wrote…
I happen to like Dawkins simple definition of life.
"Life results from the non-random survival of randomly varying replicators." link
I like it because I can definitively tell what meets the definition and what does not. This means I can identify the exact moment life originates. With this, the Origin of Life can't be due to "natural selection" because that is what happens after life starts.
To me, the only limits as to how life can originate is that life originates via a mechanism consistent with the OMA model of our universe.
Provoking Thought
Comment by Thought Provoker — May 11, 2007 @ 9:54 pm
May 11th, 2007 at 9:56 pm
Zachriel pointed to some bullet items, including about RNA, and said
Many of those fruitful falsifiable scientific hypotheses have been so "plausible" that, according to OOL scientist Shapiro, there is a movement of scientists toward leaving the RNA first scenario as though fleeing a building on fire. (Those interested, I have links in the recent OOL thread.)
Though still in the majority, the idea of starting with RNA appears to have already reached its zenith.
Comment by eric — May 11, 2007 @ 9:56 pm
May 11th, 2007 at 10:07 pm
Hi Eric,
LOL
Yes, we have had this discussion before. They aren't abandoning the "RNA First" concept because they don't believe the RNA World didn't occur. They are only moving down the regression to PRE RNA World. While this may make it an RNA Second (or Third or Fourth or etc) it is science going forward, not backwards.
If you would like, I can provide links. Do you really want to challenge?
Provoking Thought
Comment by Thought Provoker — May 11, 2007 @ 10:07 pm
May 11th, 2007 at 11:11 pm
TP, there are some OOL advocates who have abandoned the RNA world concept and there are solid reasons for doing so. Among the reasons is the nature of RNA. Unlike its DNA cousin RNA lacks the properties suited to a long term carrier of information. IOW, it is less stable. Like all other chemicals it would also be subject to covalent modifications which induce mutations (assuming there is an ordered genome to mutate). Genomic change is no savoir for this paradigm because the utility of change presupposes a means of maintaining genomic integrity. There would be none of that at the outset. If a miracle occured and RNA acquired encoding capacity and a means to express encoding information it had better also have mechanisms able to counter genomic decay. But another miracle would make a reasonable person suspicious that a guiding hand was behind all this- and he would be right.
Comment by Bradford — May 11, 2007 @ 11:11 pm
May 11th, 2007 at 11:41 pm
Hi TP. You wrote:
TP, given this definition you are unable to tell that the presence of an RNA SRM would constitute the origin of life. According to the quote the survival of SRMs would be non-random in nature. That would occur with a specified chemical reaction but that could also entail an outcome that is deterministic in nature e.g. combustion of hydrogen and oxygen leads to water. There is no variation just a predictable outcome. OTOH, variation that is non-random implies a selection process. Dawkins has supplied a definition that is meaningless in the absence of an empirical reference frame detailing the reactants involved, the conditions and the properties of the identified chemicals. There is no experimental evidence showing that SRMs evolve in the direction of organized interacting parts that replicate as a whole.
Comment by Bradford — May 11, 2007 @ 11:41 pm
May 12th, 2007 at 12:23 am
Hi Bradford,
I suggest there are some OOL advocates that question the RNA World concept and always have. If done for ethical reasons, that is science. Is Orgel one the scientists alledgedly "abandoning" the RNA World?
This has been discussed before on TT. Here is my comment with links on the subject… http://telicthoughts.com/rna-w...
Provoking Thought
P.S. It is getting late, I will have to respond to your other post later.
Comment by Thought Provoker — May 12, 2007 @ 12:23 am
May 12th, 2007 at 11:57 am
It is obvious that any OOL would have to eventually get back to RNA, DNA and proteins. I never doubted that those who are fleeing RNA first for metabolism first expect to do this — someday.
With regard to the Language Barrier, however, this is a retreat from the problem. By analogy, it is putting off until some future day the thorny problem of how to get the hot air balloon to fly beyond the earth's atmosphere in order to get to the moon. For now, they are going to struggle with other issues that are also giving them huge grief. They are working on finding a better balloon or a better starting point.
I made reference to the RNA troubles as a response to Zachriel's mention of fruitful results — so fruitful that people feel a need to start over from an entirely different direction.
But in the larger picture, while it's not a given that they can build up to RNA, it matters little whether they can or not. As Bradford said
That is a key point because it touches directly on the limitations of Before Language replication and whether it could ever reasonably construct the system of synchronised mechanisms that would be necessary to perform language processing.
Just stop and think for a bit about what is required just to decode DNA into proteins. That is just the retrieval and decoding aspects. How will SRMs build such mechanisms in a pre-language universe?
And why would they? They have no need to go symbolic. Nature can be quite content operating with non-symbolic direct actions. Nature doesn't need symbolic information processing to function or to fulfill its nature.
Comment by eric — May 12, 2007 @ 11:57 am
May 12th, 2007 at 12:21 pm
It's known with some certainty that the Earth began much too hot for liquid water, and there is some evidence that it was sterilized at least once after its formation. However, it is possible that Earth was seeded, either by organic compounds, or by life itself. But if it was seeded, the seeds were primitive.
Comment by Zachriel — May 12, 2007 @ 12:21 pm
May 12th, 2007 at 12:30 pm
You are correct that these experiments start with RNA under highly contrived conditions, but the statement remains true that these contrived chemical replicators evolve 'specified information'.
Conceivably it could, but I think segregation is probably an important aspect of the process.
Comment by Zachriel — May 12, 2007 @ 12:30 pm
May 12th, 2007 at 12:47 pm
I assume you are using "language processing" as a mechanism that encodes, stores, retrieves, and decodes. Yet, you ignore the evidence that such an evolutionary scenario is possible, meaning your claim of some sort of barrier is unsupported. That doesn't mean there isn't some other sort of barrier, just that your broad and overreaching claim is faulty.
The genome is considered the coded information. The genome is decoded into proteins. But proteins do not code back into genomes. Genomes replicate (with mutation).
Furthermore, there is evidence that RNA can evolve the capability of manufacturing proteins. Even if the particular mechanism turns out to not be the historical avenue, it shows the conceivability of such a process. Your claim, based on no empirical data but just an analogy and incredulity, is that such a process is not conceivable. Your 'proof' is faulty.
I understand that. The strongest statement you can make in this regard is that it seems to be a difficult problem. Did you even read the article, much less the linked studies?
Comment by Zachriel — May 12, 2007 @ 12:47 pm
May 12th, 2007 at 1:00 pm
I specifically avoided the term "RNA World", even using caps for emphasis; "Abiogenetic Hypothesis". Shapiro suggests that there may be even more primitive beginnings to life. He does not abandon the concept of abiogenesis, and suggests new lines of research. That's what scientists do. I would not be surprised if this story takes many turns.
That's not untypical in science. As long as it is fruitful, it will be investigated. If Shapiro's ideas on "metabolism first" are fruitful, they will continue to lead to new research. You can always tell a good Hypothesis by all the new empirical questions it raises. What empirical studies can we make to investigate the "Language Barrier"
Comment by Zachriel — May 12, 2007 @ 1:00 pm
May 12th, 2007 at 2:04 pm
What specified information are you referring to and how does it relate to biological function?
Comment by Bradford — May 12, 2007 @ 2:04 pm
May 12th, 2007 at 3:42 pm
Zachriel, I would echo Bradford's question about your "specified information" claim. By "specified information" do you mean "specified complexity" (which need not be symbolic information at all — proteins have specified complexity), or do you actually mean "symbolic information" (the topic I have been talking about)?
When you say
I notice that you do not their claim that it is a solved problem. That would be a surprising omission if it were the case that RNA have already been found to be creating symbolic information.
About the retreat away from macromolecules to a simpler metabolism first approach, I am not objecting to the fact that they are doing this, nor am I claiming science doesn't or shouldn't give up on some approaches and try others instead. The point here is that RNA first appears to be yet another example of an OOL blind alley. It is progress to learn that, but that is not the same as progress toward crossing the Language Barrier.
If that were always true, then there would never be a genome and an unguided origin to life would be impossible. Game, Set, Match. You can't get information out where no information has ever gone in. You can't forever get your information by assuming previously existing information.
That is why one of the crucial obstacles of the Language Barrier is encoding from actual proteins into symbolic genetic information describing their assembly. An unguided process cannot think or imagine or invent this information. The only possible source is to start from actual proteins and encode into coded representation.
The mechanism to do that encoding must be created by Before Language replication.
Comment by eric — May 12, 2007 @ 3:42 pm
May 12th, 2007 at 4:00 pm
The basics of RNA evolution are now considered a standard laboratory procedure and have applications in many areas. Over a decade ago, Bartel constructed a quadrillion random RNA sequences, of which 65 were synthetic ribozymes.
Isolation of new ribozymes from a large pool of random sequences:
Some other relevant research.
Experimental evolution of complexity: in vitro emergence of intermolecular ribozyme interactions.
Continuous in Vitro Evolution of Catalytic Function
'Accelerated Evolution' Converts RNA Enzyme To DNA Enzyme In Vitro
Study Offers Insights Into Evolutionary Origins Of Life
Here's an interesting article by Szostak, Bartel and Luisi, Synthesizing life and provides the basics of the RNA World Hypothesis. There is ample evidence that life diverged from primitive ancestors, and that all organic systems are consistent with chemical laws. Nor are there any known barriers to prevent such a possibility. Of course, there is no complete theory of abiogenesis. It's a Gap.
Comment by Zachriel — May 12, 2007 @ 4:00 pm
May 12th, 2007 at 4:26 pm
Zachriel, you wrote:
There is no dispute about function in accordance with chemical laws. The issue is whether laws governing chemical reactions are sufficient to explain the origin of information. It does not surprise me that directed evolution can produce biomolecules like RNA that exhibit enzymatic properties- ligation, excision and perhaps more- hydrolysis etc.? These are toolbox functions. In real organisms the information needed to generate them in a timely manner is found in DNA. It is important to distinguish between nucleic acids whose function it is to retain information and enable the gene expression that generates the toolbox biomolecules and nucleic acids having enzymatic functions. The latter are functionally equivalent to proteins which do the same. While enzymes are easily analyzed within the context of chemical properties, information bearing nucleic acids are not. Of course their chemical bonds are but their critical property- their nucleotide sequencing in accordance with a coding convention- is not reducible to underlying forces of nature.
Thus, your presumption of a gap equates to my explanation of intelligent input. You believe an answer has not been found and I believe you will not accept the one that logically presents itself.
Comment by Bradford — May 12, 2007 @ 4:26 pm
May 12th, 2007 at 4:31 pm
Let's start with information; in this case, defined as the sequence of nucleotides. And this is in reference to your statement that "information isn't a chemistry problem".
I've only stated it four times just on this thread. There is no complete valid theory of abiogenesis. I have also posted it on other blogs.
What the heck are you talking about. Why would the information have to come from proteins rather than some other source? We already know that genomes can evolve, creating new information in response to the environment.
Sure you can. What odd definition of information are you using?
Comment by Zachriel — May 12, 2007 @ 4:31 pm
May 12th, 2007 at 4:39 pm
If you are going to make this claim then you need to demonstrate either a chemical process or a selection process whose end result is an encoding convention.
Comment by Bradford — May 12, 2007 @ 4:39 pm
May 12th, 2007 at 7:29 pm
What predictions of empirical phenomena can you make from that explanation? What specific observations do we make to verify and extend our knowledge of this "intelligent input"
I was referring to the standard mathematical definition of information as defined by Claude Shannon.
Comment by Zachriel — May 12, 2007 @ 7:29 pm
May 12th, 2007 at 10:06 pm
This is a question begging position. You can't use the ability of genomes to evolve (after they exist) to explain the origin of genomes or genetic information in the first place.
You can't run a car on an empty tank of gas, and you can't evolve information in DNA if you've never had an encoding source that puts some there to start with (along with the other mechanisms needed to form a complete functioning information processing system).
If you want to insist its "just a gap", you still have a burden to show plausibility for crossing the gap. Assuming you can start on the far side with a working, evolving genome is begging the question, which is not an acceptable method of establishing plausibility.
So far you have referred either to what a genome can do after it is up and working, or else what molecules may be able to do prior to the topic of encoding, but no plausibility for creation of coded information or coded information processing mechanisms. Bradford's recent post is exactly on target in pointing out this distinction.
Please note that it is not relevant to show that RNA or other molecules have "information" merely in the sense of sequence. The "gap" (what I call a barrier) that needs plausibility is about coding and decoding to and from symbolic representations.
I understand you've acknowledged no one has a "complete" story of abiogensis, but you seem reluctant to acknowledge that this particular gap is consistently and entirely missing. There is not even a partial crossing across the symbolic coding gap, at least as far as you've shown to this point. Hence, the burden of plausibility on this matter is so far unmet.
Replication is not and never will be the same as symbolic encoding. Replication makes more copies of what you have. Symbolic encoding produces something entirely unlike what you have, namely a coded symbolic representation.
Finally, the reason symbolic instructions for constructing proteins (i.e. genetic information in DNA) would have needed to be originally derived from proteins is that mindless matter has no ability to imagine or invent or create what does not already exist. It couldn't design by imagination what coded sequence would make a protein. The idea for building a protein could not start with the information.
In a mindless world the only possible source for encoding protein design information is the protein that is to be described.
To supply the original content of DNA for protein construction instructions, the most one could possibly hope it could do is to mindlessly extract from an actual protein the instructions for its construction. The protein would become the guiding template (although for encoding, not for replication). Mindless creation of the first symbolic codes requires that an actual case provide the design, since there is no mind to do the designing.
Whereas copying is far easier, encoding requires mechanisms to convert realized forms into a dissimilar symbolic form. Yet the precoding world has only basic replication to build such mechanisms. It would not have the inheritable instructions of the post genome world, and simple replication of physical templates would not scale up to duplicating complex interacting mechanisms.
Comment by eric — May 12, 2007 @ 10:06 pm
May 12th, 2007 at 10:21 pm
As I've mentioned earlier, the ID inference rests entirely upon this hypothesis:
If this is true, the presence of one of those distinguishable effects justifies an inference to intelligent agency. Regarding a particular claim, e.g. that the creation of symbolic language is such an effect, there are two aspects to the claim.
1. Intelligent agents can do this.
2. Unguided natural processes cannot.
The first is established beyond reasonable question. The remaining issue is entirely about the second half. Consequently, the ID inference can be understood as clarifying the limiting boundaries around what unguided natural processes are capable of.
When discussing limitations of nature (i.e. natural processes would not do X, e.g. convert copper into gold by chemical processes), the common and basic way to test this empirically is to look for exceptional cases that would contradict that hypothesis. If one had a competing hypothesis, that could also suggest tests where a counter example to the first hypothesis might be found.
So, for a claim like "Only intelligent agents create symbolic languages." one would try to find counter examples that show that unguided nature can create symbolic languages.
If the evidence consistently shows that unguided nature does not do this, that consistency supports the hypothesis.
Consequently, every time OOL research runs up to the language barrier, runs out of gas, and has to turn back and try some other way, that is empirical data consistent with the hypothesis that only intelligent agents create symbolic languages.
Comment by eric — May 12, 2007 @ 10:21 pm
May 12th, 2007 at 10:25 pm
p.s. The investigation is not limited to empirical data. We can also make qualitative evaluations and reason from them. For example, we can observe how intelligent agents are able to create symbolic languages using capabilites that mindless matter manifestly does not have.
That is prima facia evidence to expect that mindless matter may not share the ability to create languages. In the absence of the discovery of alternate methods that would be available to mindless matter, the hypothesis that mindless matter can do this too would lack a theoretical foundation, as well as lacking empirical support.
p.p.s. If it is not painfully obvious, I predict that OOL will continue to be unable to cross the language barrier and that we will not find a basis for expecting mindless matter to create symbolic language.
Notice that OOL research is an essential part of providing the empirical support for the limitations of unguided natural processes. It is precisely as we learn more and more about how mindless matter does not behave that the ID inference is strengthened.
It is not based on the lack of knowledge. It cannot be. If we just didn't know what matter could or couldn't do, we couldn't justify the inference. It is only as we gain knowledge about what matter does not do that the ID inference becomes established.
This is exactly the point made in The Mystery of Life's Origin, i.e. that their inferences were based on the increase in knowledge over the previous three decades.
Comment by eric — May 12, 2007 @ 10:25 pm
May 12th, 2007 at 11:02 pm
I pointed out that this statement was untrue. You have never clarified or retracted it.
Nor did you provide specific observational predictions as requested.
Comment by Zachriel — May 12, 2007 @ 11:02 pm
May 12th, 2007 at 11:13 pm
Let's analyze the components of a system and then identify their biological structures. There will be a transmission source that generates a message signal encoded by a transmitter, an information channel, a receiver that translates the signal and a destination. There are self-checking and error correction mechanisms.
A signal is generated through the transcription (mRNA) of DNA (this is a very simplified version which omits detailing the many components of successful signaling i.e. promotor regions, RNA polymerases, transcription factors and more) tRNA and tRNA aminoacyl synthetases would be components involved in a translation process involving codons and anti-codons, the ribosome, the destination. There are self-checking mechanisms identified in RNA polymerases, aminoacyl synthetases and ribosomes and the genetic code functions as a protocol. I've seen variations of these things and you might view this differently but there should be at least some rough equivalence possible. This system is unique in that the hardware involved is encoded and manufactured by the same devices that are involved in the transmission of information. No factory outsourcing here.
There appeared to be some exasperation in your response to Eric's observation that you don't get information out unless some has already gone in but I agree with him. I also agree with you that Shannon's concepts can be used to quantify and describe information and that we can apply this to biological systems. Intelligently generated information is utilized to set up an information system and create the codes and correction mechanisms. BTW, the original term for information was intelligence. There are obvious parallels to be drawn to intelligence and the quacks like a duck idea certainly is applicable to information systems but I want to return to your first question:
This is not obvious because a hallmark of intelligence is the capacity to choose different options among multiple routes to an end. However I'll give it a shot. This information transmission system must be constructed piecemeal and the only way to do it entails a selection process. This is the ultimate irreducibly complex system. But while this indicates design and implicates a blind process as problematic it does not render a prediction. For that we must assume a high level of function. I'll ignore my own doubts that it could be possible to generate an information transmiting system and assume it exists- every part of it except for the self-checking and error correction mechanisms. There will be genomic patterns evidencing information and a means of transmitting and even receiving and translating it. But could an existing system evolve needed checking and correction mechanisms needed to maintain information and the hardware used to transmit it. The prediction- no. Sequences would become increasingly random with respect to functional patterns. Information loss would exceed gains. A corollary prediction is that correction systems would be needed at the origin of any selection process (my own version of a front loaded necessity) that could conceivably create information.
Comment by Bradford — May 12, 2007 @ 11:13 pm
May 12th, 2007 at 11:16 pm
Hi Bradford,
I wrote (actually Dawkins)…
"Life results from the non-random survival of randomly varying replicators."
I was tempted to yell, Eureka we have found life coming from non-life.
You, Dawkins and I would define RNA molecules as "life". It replicates itself. It can randomly mutate. Positive mutations tend to survive. Negative mutations tend to not survive.
Hopefully, you aren't questioning the empirical evidence that RNA SRMs do mutate.
Hopefully, you aren't questioning the empirical evidence that detrimental mutations are… well… detrimental and are less likely to survive (non-random).
These factors, alone, meet Dawkin's definition of Life.
Go ahead and argue that this is a one way ticket to nowhere, but it is "…the non-random survival of randomly varying replicators."
We could have a lively argument as to whether beneficial mutations could or would happen but it would probably be easier to just replay a tape of you and Smokey arguing.
My minor point was that I liked Dawkin's definition of life because it provided clear demarcations of what was and was not life.
Regards,
TP
Comment by Thought Provoker — May 12, 2007 @ 11:16 pm
May 12th, 2007 at 11:26 pm
I addressed this issue with Zachriel. The SRMs are effectively enzymatic in nature. They catalyze their own replication. Conceivably a genetic change could lead to another type of catalytic SRM. Life is characterized by many interacting biomolecular parts that must replicate as a unit. This in turn requires, not an enzyme, but an information retentive nucleic acid that has information enabling the replication of a complex of different functional parts. That is not envisioned by multiple SRMs. Neither is there a directional indicator providing a reason to believe mutations of an SRM are selected for their capacity to become information sources for biomolecular complexes.
Comment by Bradford — May 12, 2007 @ 11:26 pm
May 13th, 2007 at 9:05 am
Thank you for being specific. Noise *is* Shannon information, and Shannon showed that noise is *inevitable* in any copying system (regardless of mechanism).
That's not a specific prediction. It's a restatement of the Gap. It's a universal negative. We would have to know every possible system. But we are always limited by our technology and even our imagination.
Your prediction is already falsified by research I posted on this very thread.
–
Self-replicating molecules can continuously evolve improving catalytic and amplification rates. Some have 95% accuracy, the accuracy not being dependent on sequence length or sequence content. And they can persist. These replicating molecules can segregate into discrete functions, a tiny community of cooperating molecules. They can evolve into DNA enzymes. And they can evolve the ability to synthesize proteins.
–
These discoveries are neither trivial or obvious, and they are consistent with the Abiogenetic Hypothesis. Does this research "prove" that RNA-First or even RNA-World accurately represent events on Earth billions of years ago. Of course not. But they are specific empirical support for the General Hypothesis.
There are still many Gaps in human scientific knowledge. Though science doesn't know all the steps in the sponteneous emergence of life, many paths to stopping points along the way are known.
A Gap is not scientific evidence, but a statement of the extent of human ignorance (which is vast). Saying you will never fill a Gap is not a specific empirical prediction. If someone claims there is a barrier between New York and San Francisco, we might scientifically investigate that barrier. What are its characteristics? Is it concrete? Is it a great sea? Or is it fear of the unknown?
Comment by Zachriel — May 13, 2007 @ 9:05 am
May 13th, 2007 at 10:16 am
I'll have to repeat what I wrote to TP. Enzymes are not information storage nucleic acids. Asserting that enzymatic RNA can evolve the ability to synthesize proteins does not address the real issue. From the previous comment:
Cells consist of many interacting biomolecular parts that must replicate as a unit. This in turn requires, not an enzyme, but an information retentive nucleic acid that has information enabling the replication of a complex of different functional parts. That is not envisioned by multiple SRMs. Neither is there a directional indicator providing a reason to believe mutations of an SRM are selected for their capacity to become information sources for biomolecular complexes.
Comment by Bradford — May 13, 2007 @ 10:16 am
May 13th, 2007 at 10:25 am
Information systems function through signal encoding at transmisssion and decoding at reception. These require intelligent guidance.
Bradford: But could an existing system evolve needed checking and correction mechanisms needed to maintain information and the hardware used to transmit it. The prediction- no.
It suggests a means of testing. Unless you are into miracles correction systems would have to evolve through a selection process. The process you propose (SRMs) begins without correction capabilities. Yet natural corrupting forces would nevertheless exist. We do not have to know every possibility to be aware of initial conditions. The claim that functional information accumulates should be testable and not taken as dogma.
Comment by Bradford — May 13, 2007 @ 10:25 am
May 13th, 2007 at 10:36 am
All you have stated is that there is no complete theory of abiogenesis. No one claims there is. However, there's no evidence of some sort of barrier either. Meanwhile, significant supporting evidence has been found.
That was another commenter's issue, the evolution of the ability to utilize a "code".
RNA is a retentive nucleic acid, and RNA can evolve into DNA enzyme. There are other nucleic acid polymers that may have preceded RNA World.
Replicating RNA diverges into derived molecular species creating a productive and dependent intermolecular interaction"”Experimental evolution of complexity. Primitive, but certainly supportive.
This is falsified by the previous example. Nor is there evidence of any barrier.
There's Gaps. That's normal. We're dealing with processes that occurred billions of years ago and left little evidence.
Comment by Zachriel — May 13, 2007 @ 10:36 am
May 13th, 2007 at 10:48 am
No. I'm pointing out that OOLers are not addressing the real issues. There is a fundamental flaw on a theoretical level.
There is no demonstrable ability of a capacity to evolve a code. Self-replicating RNA is an uncoded chemical process.
Nucleic acids are not information retentive by definition. Randomly generated nucleic acids are informationless.
You are confusing the parts with an information molecule needed to generate multiple parts that have a capacity to replicate as a unit.
Show me a single SRM with the capacity to encode systems.
Comment by Bradford — May 13, 2007 @ 10:48 am
May 13th, 2007 at 10:51 am
Your response is a non sequitur, assumes its conclusion, and leaves the original point uncontested, concerning eric's faulty statement that "You can't get information out where no information has ever gone in."
As for a communications channel, information is *replicated* (with inevitable noise). Even short RNA molecules can replicate. And when they do, they mutate and evolve.
Functional information accumulates. And it persists. I have provided cites. Here's an interesting roundup from Szostak along those lines:
–
Note the use of informed intuition to direct research. One might intuite some sort of Barrier, but it then requires specific empirical support. Compare this research, "isolating a set of distinct aptamers, all of which bind the same target (GTP), but with a wide range of affinities" to a sweep of the hands, "But could an existing system evolve needed checking and correction mechanisms needed to maintain information and the hardware used to transmit it. The prediction- no." With the former prediction, I can replicate the precise experiment. With the latter prediction, I don't even know where to start. What system? Any system imaginable?
Comment by Zachriel — May 13, 2007 @ 10:51 am
May 13th, 2007 at 11:05 am
I'm not assuming anything. I'm pointing out that the electronic signaling systems, which inspired Shannon, were dependent on intelligently devised signal encoding to address noise issues.
Your responses point to replication systems having an encoding component. You continuously point to replication without indicating a means by which encoding conventions, needed to enable them, are evolved. Researchers selecting for a variety of self-replicating enzymes does not explain the origin of a nucleic acid like DNA whose sole function is to store information by which many thousands of enzymes are synthesized.
Comment by Bradford — May 13, 2007 @ 11:05 am
May 13th, 2007 at 11:32 am
Shannon was clearly working on resolving issues in electronic communications. But Shannon Information, given appropriate definitions, is applicable to genomes and their replication. It was eric who introduced the term on this thread. It was eric who incorrectly claimed that "You can't get information out where no information has ever gone in."
Random noise *is* Shannon Information. There is more Shannon Information in a random sequence than in the text of Hamlet. This may be counter-intuitive, but that's why it takes a genius like Claude Shannon to figure it out.
Shannon Communiation doesn't require a coding algorithms. In particular, genomes are replicated without a code and without regard to content, which is why Shannon Information is particularly appropriate. The genetic code refers to the transcription into proteins, genotype-to-phenotype. There is no direct encoding of phenotype-to-genotype. The information in the genome comes from evolutionary processes.
Comment by Zachriel — May 13, 2007 @ 11:32 am
May 13th, 2007 at 11:54 am
You are making an erroneous claim that genomes are replicated without a code. An SRM is an enzyme, not a genome. Real genomes (DNA) require a code by which their replication is enabled.
The genetic code refers to the logical linkage of codons to amino acids. Transcription involves the molecular machinery by which a signal is prepared. It is subsequently translated and corrected on multiple occasions before the polypeptide is realized.
It is the functional value of phenotype that accords genes their selective value.
A replicating process requires, in von Neumann terms, an information tape that is distinct from the replicating machinery. Information preceeds replication.
Comment by Bradford — May 13, 2007 @ 11:54 am
May 13th, 2007 at 12:24 pm
The specific pairing of DNA suggests the copying mechanism for the genetic material. They are copied base-by-base without an intermediate code.
Ok.
Of course. That's the whole point.
As we know that RNA can self-replicate, such a von Neumann machine is not the only possible replicating process.
Comment by Zachriel — May 13, 2007 @ 12:24 pm
May 13th, 2007 at 2:04 pm
Wrong. Complementarity of base pairs alone does not account for a capacity to replicate. The replication involves protein enzymes and non-enzymes whose synthesis is code dependent.
Comment by Bradford — May 13, 2007 @ 2:04 pm
May 13th, 2007 at 2:10 pm
The von Neumann machine, referencing many distinct moving parts that replicate in coordination with each other according to encoded instructions, is the relevant model for cells. Self-replication encompasses non-RNA SRMs. They can include non-biomolecules. It indicates chemical properties, not to be confused with symbolic representation of information.
Comment by Bradford — May 13, 2007 @ 2:10 pm
May 13th, 2007 at 2:12 pm
Hi Zachriel,
I haven't taken the time to read all your exchanges with Bradford, so I apologize if the following is off the mark.
Another commenter, Smokey, and Bradford went around and around in excruciating detail of why OOL was or was not possible. After all that time, they couldn't even agree on the definition of life much less the possibilities of its origin.
You might want to take a look at the comment I am about to make to Bradford and his response to it.
Regards,
TP
Comment by Thought Provoker — May 13, 2007 @ 2:12 pm
May 13th, 2007 at 2:32 pm
Hi Bradford,
I wrote…
Hopefully, you aren't questioning the empirical evidence that RNA SRMs do mutate.
You responded with…
As indicated in my comment to Zachriel, I didn't read all of your exchanges and I apologize if I am off base. I did, however, read enough to note that you copied and pasted this to Zachriel and started with "I'll have to repeat what I wrote to TP." I suspect you could copy and paste a lot of things from your exchanges with Smokey too.
How about some basics? Please identify which of the following you classify as a living thing and which are not.
1. A rust molecule
2. A star
3. A single strand of virus
4. A single RNA molecule.
5. A single Nanobe
Using Dawkin's definition I can explain why rust and stars aren't life and why the rest are. I can also do it using a single sentence for each example.
Can you?
Provoking Communication
Comment by Thought Provoker — May 13, 2007 @ 2:32 pm
May 13th, 2007 at 3:12 pm
TP, I'm undecided about nanobes but the remainder are not evidence of life. But that is not the central issue to an origins/evolution scenario. The relevant question is how to get from A to Z. "A" being a starting point on prebiotic earth and "Z" a living cell. Unless one of the above is useful in explaining the process it has no broader implications for natural history.
Comment by Bradford — May 13, 2007 @ 3:12 pm
May 13th, 2007 at 4:50 pm
Who said it did? "The specific pairing of DNA suggests the copying mechanism for the genetic material" is language reminiscent of Watson & Crick's seminal 1953 paper. DNA replication requires a mechanism, specifically template synthesis. Please provide a cite so that I know what you are referring to.
But that wasn't the claim I was addressing. Perhaps a von Neumann machine is a reasonable model for modern cellular organisms. The claim I was addressing was your assertion that replication requires "an information tape that is distinct from the replicating machinery". It simply doesn't.
Comment by Zachriel — May 13, 2007 @ 4:50 pm
May 13th, 2007 at 7:49 pm
So you realize that a genetic code needs to be in place to allow for the replication of DNA encoded genomes. That still leaves the question of how duplicating enzymes produce an encoding convention.
Comment by Bradford — May 13, 2007 @ 7:49 pm
May 13th, 2007 at 8:05 pm
The content of the DNA being replicated is immaterial to the copying process. It can be coding for proteins, retroviruses, or be utter gibberish. Watson & Crick knew nothing of the genetic code when they correctly suggested the obvious template copying mechanism.
Comment by Zachriel — May 13, 2007 @ 8:05 pm
May 13th, 2007 at 8:12 pm
Hi Bradford,
You wrote…
Yes, I am aware you want to take in the whole picture all at once. As an engineer, I look for more manageable pieces. Right now, I am looking for your non-life to life transition. Nanobes are too small so we go up a notch.
I taking it there would be no disagreement that Mycoplasma genitalium is life, since it is a full fledged organism.
Here is the link which has an applet that "…displays a gene map of the complete genome of Mycoplasma genitalium"
So, if we can create this baby in a lab using non-life things like RNA, would you agree that we have created life from non-life?
BTW, I think it should be obvious that no one will never convinced they know how OOL actually happened on Earth. I doubt anyone will ever convince you that it was even possible to happen on Earth. The best anyone can do is convince you it is possible.
Will you concede that it is possible to create life from non-life? It will save a lot of time.
Provoking Thought
Comment by Thought Provoker — May 13, 2007 @ 8:12 pm
May 13th, 2007 at 9:30 pm
I do not believe that there will ever come a time when one will be able to say: if we set up the following conditions and have the following organic chemicals in the mix this life form will arise.
It is conceivable that we could bioengineer life by manipulating biochemicals. It would be evidence for intelligent design.:mrgreen:
Comment by Bradford — May 13, 2007 @ 9:30 pm
May 13th, 2007 at 10:02 pm
Hi Bradford,
You wrote…
and you wrote…
Do you have different definitions for "organic chemicals" and "biochemicals"
It doesn't bother me that you wish to call it "Intelligent Design" when life is created by beings who have the ability to learn. It matches. At this point, I don't know if it bothers you. On one had you say "do not believe" on the other you as "It is conceivable".
BTW, Shall I take it you agree that Mycoplasma genitalium is life?
Provoking Thought
Comment by Thought Provoker — May 13, 2007 @ 10:02 pm
May 13th, 2007 at 10:18 pm
Hi TP.
Biochemistry is the study of chemical processes in living organisms. Some of this includes inorganic chemicals but I would think that an OOL scenario would focus mostly on organic chemicals. As an IDist I believe there was some intentional and intelligent meddling with the process that resulted in initial life forms. So a process that involves such meddling by a competent meddler is conceivable but I do not believe that natural forces alone would generate cells on a lifeless planet. Is this any clearer?
Comment by Bradford — May 13, 2007 @ 10:18 pm
May 13th, 2007 at 10:50 pm
Hi Bradford,
You wrote…
Yes and no. I suggest a lot of this is biased by an assumption that Earth's "organic organisms" are the only life possible. Is the term "inorganic life" a total impossibility for you? Robotic life?
As to your beliefs and disbeliefs, I already recognized and pointed out the fruitlessness of trying to argue Earth's kind of OOL with you. So I am not.
I offer a compromise. While you may have doubts Nanobes meet all the requirments to be called organic organisms. Here is a web site that contains a paper on Nanobes explaining how they act alive, have DNA and how it would be difficult to explain the data any other way besides as a biological organism. I purposely didn't use the words "biological" or "organisms" when I asked for example of "living things". Isn't there enough evidence to show Nanobes are living things even if they might not meet the textbook definition of "organic organism"
Will you consider Nanobes to be living things based on available data? If not, why not?
Provoking Thought
Comment by Thought Provoker — May 13, 2007 @ 10:50 pm
May 14th, 2007 at 7:50 am
Correction: You were the one who introduced the term on this thread. Until now, I have not referred to Shannon Information.
But in a different note, you also wrote about "specified information". Just to be sure we are clear, I trust that you do realize and agree that random noise is never specified information, correct? And that a random sequence is an unspecified sequence, making "random" and "specified" mutually exclusive attributes, correct?
So, I am a bit surprised that you should separately ask "What odd definition of information are you using?" as though you hadn't referred to specified information or to coded information, or as though Shannon Information were the only possible meaning of "information".
I would have hoped that by now it would be plain that I am not talking vaguely about Shannon Information and especially not random noise. The context of my remarks was the genetic information that holds coded instructions for specifying amino acid sequences. This information is specified and it is in a code, which means that coding is required. Since you yourself have said
then I trust you shouldn't think it odd to refer to coded information, and I expect that everyone, including yourself, will agree that we should not expect to be able to decode random noise and get a functional protein. Its obviously not "odd" to refer to coded information with decodable, functional meaning.
Understood in the context it was made according to its intended meaning of the specified genetic information that codes for proteins, my statement is correct and I do not retract it. I also clarified it here. Apparently you missed that post or misunderstood, or I trust you would have responded to the content of the clarification rather than suggest that it never happened. I pointed out that your statement
is a question begging argument. Since the question is how unguided natural processes can create symbolic instructions that code for proteins in the first place, you can't appeal to an evolving, functional genome as any part of the explanation. That begs the question.
Well before the coded instructions in DNA can change through evolution, there must first be some coded instructions in DNA to begin with. My statement points out that you cannot get coded instructions for the creation of proteins out of DNA if you have never first encoded such instructions into DNA.
Neither does adding random noise provide a viable starting point. The fact that random noise might be counted as Shannon Information is irrelevant for the purposes of explaining the origin of coded information that can be decoded into proteins.
At this point I want to be careful not to attribute to you a position you do not hold, but I cannot help get the impression that you seem to hold that nature does not (ever?) encode from proteins to coded instructions. So I'd like you to please clarify this point.
If that were always true about nature, there would be no source for the intial protein instructions of the first genome, rendering it impossible to populate. That would make it impossible for a genome to get started by unguided natural processes. Hence my reference to "Game. Set. Match."
Mindless matter has no imagination and cannot creatively invent the coded instructions prior to the creation of proteins. It could not anticipate proteins. Nor could it start with random noise and evolve that into coded instructions for protein construction. A mindless process would necessarily be required to derive the coded specified amino acid sequence from the only available source for that information, namely by encoding the symbolic sequence from an actual amino acid sequence.
Comment by eric — May 14, 2007 @ 7:50 am
May 14th, 2007 at 10:33 am
The first reference to "information" on this thread was by you. I then asked you to clarify what definition of information you were using. When Bradford interjected, I told him I was using the standard mathematical definition, Shannon Information. You allowed this to stand, and continued to use the term.
You are more than welcome to qualify your meaning. As information is a mathematical concept, please provide the appropriate definition.
We can specify a random sequence. Patterned would be a much more precise and clear terminology in context.
"Specified" qualifies "information". It doesn't change the definition of information, but delineates a subset. In other words, it's still Shannon Information. Shannon dealt primarily with specified information (information input into a communication stream).
I apologize if I misunderstood your statement. However, it is quite possible that you were conflating terminology or perhaps confused on the basic principles of Information Theory. That's why I asked for a definition.
Anyway, it is quite possible to get patterned information from genetic algorithms, or even from chaotic functions.
That is not question begging, as your modified statement "You can't get specified information out where no specified information has ever gone in," concerns information generally and not genomes specifically. We can point to evolutionary processes as an example of a process that generates patterned information. As to "specification", this is probably ill-defined in context, though I suppose you could point to the communication stream of replication; but as I pointed out, once input to the communication stream, it is specified whether it codes for proteins, retroviruses or utter gibberish. But you are more than welcome to provide a precise mathematical definition of "specification".
Comment by Zachriel — May 14, 2007 @ 10:33 am
May 14th, 2007 at 11:03 am
The Central Dogma of Genetics is the one way transmittal of information from DNA to RNA to Proteins. But like much of biology, nothing is as simple as it appears on the surface. There are a variety of exceptions; e.g. retroviruses, prions, etc. This does not account for the majority of the patterns, especially active coding patterns, found in genomes, which are primarily due to evolutionary processes from common ancestral populations.
I understand you keep saying that, but it requires ignoring the evidence that specialization of function in self-replicators can lead to molecular communities, protein synthesis, DNA enzymes, and the replacement of RNA-enzymes with protein-enzymes.
Comment by Zachriel — May 14, 2007 @ 11:03 am
May 14th, 2007 at 11:38 am
Natural patterns are observable. However, patterns alone are meaningless with respect to encoding functions. The expression of information is made possible by a coding convention that assigns meaning to patterns. The patterns you are looking at now have no meaning in Japanese because the encoding convention of that language differs. If the selective value of a biological function determines whether or not genomic patterns are retained then natural selection has no scientific utility until an encoding convention is already in place. To paraphrse Eric it is game, set and match for ID.
Comment by Bradford — May 14, 2007 @ 11:38 am
May 14th, 2007 at 11:44 am
If this refers to the references you previously cited then kindly cite the exact data you are using to support your claim that evidence for the cause of protein synthesis is linked to "specialization of function in self-replicators."
Comment by Bradford — May 14, 2007 @ 11:44 am
May 14th, 2007 at 11:51 am
That is contrary to fact. We know that an encoding convention is not required for replication and ensuing selective evolution. There is also some evidence that protein synthesis can evolve from these humble beginnings.
Comment by Zachriel — May 14, 2007 @ 11:51 am
May 14th, 2007 at 12:01 pm
Bradford: If the selective value of a biological function determines whether or not genomic patterns are retained then natural selection has no scientific utility until an encoding convention is already in place. To paraphrse Eric it is game, set and match for ID.
Sure. You can observe self-replication- a chemical process attributable to chemical properties of the substances involved. But synthesis in real cells is dependent on an encoding convention. Using SRMs as evidence for the origin of cells is like using snowflake patterns as evidence for the alphanumeric symbols used in the English language.
This is an extrapolation on your part that is contrary to causal patterns observed in encoding systems whose causal chain we can document.
Comment by Bradford — May 14, 2007 @ 12:01 pm
May 14th, 2007 at 12:39 pm
I keep reading and rereading this statement. I have no idea why you consider it to be a valid assertion in the light of self-replicating molecules. I'm sure it makes sense to you.
In a self-replicating molecule, the function is the acquisition of local resources to ensure replication. This process is subject to selective evolution and requires no coding.
Comment by Zachriel — May 14, 2007 @ 12:39 pm
May 14th, 2007 at 1:24 pm
eric wrote:
Hi Eric.
I think the question of whether codes can arise from non-codes unintentionally is very important. Codes are complex abstract entities, on my view, not material ones; which is why I don't think they can arise unintentionally. I say that by way of a general metaphysical principle to the effect that abstract objects are essentially linguistic entities (in the broadest sense of language, that includes logic, mathematics, and codes); and that no language can have a wholly mind-independent existence, for language is, at its core, the means of thought—that by which thought takes place.
Codes, like all languages, involve the normative and hence intentional notion of rules or instructions. Only thus can something ever have the property of correctness or incorrectness. There is no such property if there is no rule or instruction.
Opponents of ID will probably respond that there are unintentional codes—the laws of nature. These are just there. They don't have a programmer. They're just a program whose source code is expressible in the programming language of mathematics, and which instructs the universe what to do. Even if unintententional material processes could somehow generate codes such as DNA and human language, such processes cannot have generated the laws of nature, since those laws are logically prior to any material processes they govern. Thus, on the anti-ID story short of positing a multiverse, the program we call the laws of nature would have to be necessarily unique and necessarily existent.
But physicists have calculated quite precisely what a universe would be like if it was implementing a different program—in fact lots of alternative universes have been modeled by physicists (one can vary the value of, say, the cosmological constant, etc). Most of them would be short-lived and/or lifeless. Yet unless there's a multiverse (which is in effect also equivalent to an infinite past), none of them ever gets to exist, which is puzzling if universes get to exist merely by having their source codes dwelling eternally in program heaven. The only universe we observe is one with apparently a finite past and whose program just happens to be consistent with complex and mentally rich life. So empirically there's just one code which constitutes our universe's program (aka laws of nature).
But we just never see abstract entities causing any concrete effect in reality. That's part of what we mean by 'abstract'. We do not see the law of gravity itself attracting, say, a falling cat. To be sure, we see objects with mass gravitating, but not the law itself. It's an abstraction. Inductively we have far better evidence for minds being causally effective while using abstractions such as a program, or algebraic ideas in their thinking, than we do for abstractions being causally effective all by themselves.
Comment by stunney — May 14, 2007 @ 1:24 pm
May 14th, 2007 at 1:25 pm
What you are asserting is that an SRM will keep replicating as long as a supply of substrates exist. That's basic chemistry. You are describing chemical properties and reaction conditions as biological functions.
Why describe this as a selection process rather than a chemical reaction when the selection demonstrated by researchers is intelligently determined? Research indicates there is variety among SRMs. Enzyme variety has relevance to a cellular environment which is multi-component and interacting. Varietals can be matched to function. All of this relates to interconnected function. An SRM does not model interconnectiveness. There is a theoretical hole in your paradigm.
Comment by Bradford — May 14, 2007 @ 1:25 pm
May 14th, 2007 at 8:19 pm
No evidence need be ignored. What you are missing is that none of that matters to the question being asked.
To focus for the moment on the central question at hand, let us assume that nature is fully capable of
a) peer-replicators
b) molecular communities
c) protein synthesis
d) enzymes of various kinds
None of these are symbolic codes representing the sequence of amino acids in a protein. Let us call the first such code sequence Alpha. The question is this:
The points a-d mentioned previously are not part of the Language Barrier. The question above is. If natural processes can do all of a-d but cannot create alpha, then natural processes cannot create the language-based life we observe.
So frankly, although there are difficulties of their own in a-d (and I believe some of these are of interest to Bradford), with regard to the Language Barrier success on a-d would be beside the point. Hence, repeatedly mentioning evidence in that department is also an irrelevant detour from the question I have been putting to you. For a moment, please just suppose success on a-d, and move to the question at hand.
In regard to the question at hand, please note that any appeal to the evolutionary abilities of a genome already containing symbolic instructions for protein construction is indeed unequivocably a question begging argument. The question asks about the origin of Alpha. Prior to Alpha, there are no previous symbolic instructions for protein construction. Question begging means assuming what you are trying to show. Assuming the existence of such instructions when the question is how the first one came to exist is clear cut question begging.
Please also note that without a creative mind, a mindless natural process cannot start with the abstraction of a symbolic representation. The symbolic must derive from the actual, the realized meaning. That is why encoding from an actual protein would be necessary, not optional, to create Alpha. Other than an actual, realized sequence, there is no other source for the correct symbolic sequence in Alpha. (If you think this is mistaken, please suggest any reasonable alternative as part of a plausible scenario for creating Alpha.)
That sets the bar for creating Alpha quite high for a pregenome universe.
About "information", as I said, you were the one to use Shannon Information, I have not been. Please do not attribute it to me. I would recommend that it is more reliable to understand someone in context, rather than taking it out of context in isolation. I also clarifed my statement. When you understand my position, you will recognize that random noise cannot serve as coded protein instructions, so the broad inclusiveness of Shannon Information to even include random noise is inappropriate to the question at hand.
Regarding specification vs. randomness, if you know Orgel's coining of the term specified complexity, he chose it specifically to distinguish specified complexity from random/unspecified complexity. That distinction is widely recognized and was directly applied to the issue of origin of life. I consider that to be an appropriate understanding of the difference.
Nevertheless, to avoid further miscommunication, I restated my question in terms that I hope will not be misconstrued. Here it is again, showing the direct connection to the question above.
So how does an unguided natural process make the leaps necessary to create Alpha? That is the Alpha question at hand.
Comment by eric — May 14, 2007 @ 8:19 pm
May 14th, 2007 at 8:57 pm
I agree, most definitely. The threshold crossing over into physical symbolic code sequences mapped to meanings is not just one more step among many others. The divide between actual and symbolic is a fundamental threshold in the history of the universe.
What I notice about responses to this problem is that they tend to either slide toward pointing away from the problem toward accomplishments before the barrier, or they slide toward pointing away from the problem toward capabilities of living organisms after the barrier. As any magician knows, a great impression can be made if you can direct someone's attention away from what you don't want them to notice or think too much about.
Comment by eric — May 14, 2007 @ 8:57 pm
May 14th, 2007 at 9:45 pm
Is the Language Barrier just another "gap" in our knowledge? Zachriel wrote:
If all one had was a lack of knowledge, that might be characterized as a mere gap. Of itself the mere absence of information may not necessarily show anything more.
On the other hand, there are times when persistent and consistent failure to achieve some result can indeed be scientific evidence that supports an inference. This happens when we have independent reasons for suspecting that we should not find success at that particular task. The persistent and consistent failure at that task then becomes evidence in support of that expectation.
Example: When Einstein devised his special theory of relativity, he did not execute any lab experiments or take any new measurements. What he noted was a persistent and consistent failure to detect a directional difference when measuring the speed of light, despite the motion of the observer. His theory was devised as an explanation of this "gap" in our ability to detect such differences.
In the case of the Language Barrier, it was understood that language requires intelligence even before we knew that life depended on language. That is why Professor Alan Turing proposed a test of language ability as the Turing Test for artificial machine intelligence in 1950, before the discovery of the double helix structure and the genetic code.
Our consistent experience that language is never created apart from the contribution of intelligence gives a prima facia justification for expecting that this may be true for the language of life as well, and that the genetic code is not an exception to our consistent observation.
If this inference is correct, then we have an independent basis for expecting that origin of life investigations will encounter a barrier at the point where unguided natural processes are required to cross over from purely actual and realized interactions to the creation and manipulation of symbolic codes with associated meaningful realized physical counterparts.
Notice that the basis for making this prediction does not depend on knowing the actual results of any research investigation. Thus, it is not an ad hoc effort to seize on an observation of a current gap in results. It has a rational basis that is grounded on an independent set of observations and a uniform inference from those observations.
It is further supported by a recognition in principle of the nature of the obstacles presented by symbolic encoding, storage, retrieval, and decoding. Replication can reproduce something like what you already have, potentially with variation. A replicated RNA molecule will give you another RNA molecule like its template. But symbolic encoding produces a symbolic representation that is fundamentally different from the realized meaning it represents. True encoding crosses over into symbolic meaning.
Since there are independent grounds for predicting that the threshold of language will present a barrier to unguided natural processes, persistent and consistent failure to cross that barrier does indeed provide supporting evidence for that prediction and the hypothesis that language requires intelligence.
Comment by eric — May 14, 2007 @ 9:45 pm
May 14th, 2007 at 9:58 pm
Again. Please provide a valid mathematical definition of "information" as you are using it. Thank you.
Or specified simplicity. Orgel used it in a qualitative manner. If you would continue to use this term, please provide a rigorous mathematical definition.
This is apparently being argued on logical grounds. In other words, it would apply to any such system. The level of abstraction required to be considered a "code" cannot evolve. Is that correct?
Comment by Zachriel — May 14, 2007 @ 9:58 pm
May 14th, 2007 at 11:09 pm
eric wrote:
Assolutamente, as they say in Italy.
The rest of this is also very pertinent and good.
Zachriel wrote:
Yes. Codes, like all languages, involve the normative and hence intentional notion of rules or instructions. Only thus can something ever have the property of correctness or incorrectness. There is no such property if there is no rule or instruction.
So, that is indeed correct. And remember: correctness is impossible without an intentional rule or instruction, any more than a star or a pile of manure is a candidate for being correct or possessing the property of correctness. Whereas the sentence, "Atheism is purple and over eight feet long" is obviously a candidate for being incorrect or possessing the property of incorrectness.
Comment by stunney — May 14, 2007 @ 11:09 pm
May 14th, 2007 at 11:29 pm
What makes anyone believe a mindless process generates symbolic representations constituting a message system?
Comment by Bradford — May 14, 2007 @ 11:29 pm
May 15th, 2007 at 3:05 am
Bradford wrote:
Let me take a couple of wild guesses:
1) Jaw-dropping stupidity.
2) A desperate psychological need for control, combined with a desperate need that there be nothing which one can't even in principle control— such as a mind greater than one's own, or God for short.
Notice that these are not mutually exclusive. Their combination in a single personality often results in the appalling condition known as 'Evangelical Atheist Syndrome'.
Comment by stunney — May 15, 2007 @ 3:05 am
May 15th, 2007 at 7:25 am
The claim is that it *cannot happen*, that there is some sort of barrier. It isn't my burden of proof.
As this is a mathematical statement, please point to a peer mathematical journal that contains the proof that a coding system cannot evolve from simpler precursors. Don't worry: I read maths.
Thanks!
Comment by Zachriel — May 15, 2007 @ 7:25 am
May 15th, 2007 at 7:44 am
Einstein's Special Theory of Relativity leads to specific empirical predictions, such as the extension of the half-life of accelerated radiative elements, or most dramatically in the conversion of matter to energy in the Atomic Bomb. It's not enough in science that Einstein said there was a Gap in empirical science and fill it with metaphysical glue.
That's not a prediction. The inability of penetrating a Gap may be due to the difficulty of the problem, the lack of technical ability in humans, or even a lack of imagination. You have to make specific empirical predictions that distinguish the 'theory' from existing theories and from the null-hypothesis. A good hypothesis leads to all sorts of new discoveries.
Try our analogy again. Let's say we don't know if there is a sea-route between Europe and Asia heading West. Perhaps there is a barrier. But saying so doesn't make it so. Instead, someone heads West and finds land. He explores that land and makes new discoveries. Yes, there is a barrier, and that barrier, America, can be studied and understood like any other phenomena.
But if you posit a logical barrier, then provide the formal proof. Point to the mathematical peer journal. If it is new, then publish it and get the recognition you deserve.
Comment by Zachriel — May 15, 2007 @ 7:44 am
May 15th, 2007 at 8:00 am
I'm not ignoring this question. I appreciate you restating it without the terms you haven't been able or willing to define; other than "code", which I have thus far agreed to use the intuitive notion unless that leads to new problems.
Let's add a few things: models of the development of membranes, verifiable predictions from phylogeny e.g. the order of the appearance of different triplet codes; and the other side of the 'Barrier', life on Earth diverged through a process of evolution from a common ancestral population, biological processes consistent with physical and chemical laws, and even this chemistry shows signs of a phylogenetic tree.
However, I have not claimed to have a complete theory of abiogenesis. There may be a 'Barrier'. But all the evidence collected thus far confirms the general hypothesis that life evolved from primitive chemical replicators (whether or not metabolism preceded the replicators). So, there may be a 'Barrier'. But to say there *is* a 'Barrier' requires that you provide that burden of evidentiary proof. All you do is point to a Gap, a Gap that has been shrinking since the relatively recent discovery of self-replicating molecules by Altman and Cech.
Comment by Zachriel — May 15, 2007 @ 8:00 am
May 15th, 2007 at 9:52 am
Such and such could happen is not good science. We need plausible reasons.
Comment by Bradford — May 15, 2007 @ 9:52 am
May 15th, 2007 at 12:12 pm
That's fine. There are significant Gaps in human knowledge with regards to the origin of life. There is no complete theory of abiogenesis.
Comment by Zachriel — May 15, 2007 @ 12:12 pm
May 15th, 2007 at 2:22 pm
All of us are aware that humans are not omniscient. That is all the more reason to be open to all possibilites related to causality. Distinguisshing between a real gap and dogma can be impossible at times. That's a good reason to confine observations and conclusions to specific evidence. If abiogenesis infers prerequisite exclusion of telic or directed causes then its inability to identify a causal trail to life may be traced to doctrinal restrictions on inquiry. That's not for me.
Comment by Bradford — May 15, 2007 @ 2:22 pm
May 15th, 2007 at 3:16 pm
Zachriel wrote:
Please point to a peer-reviewed mathematical journal article that contains the proof that "Codes, like all languages, involve the normative and hence intentional notion of rules or instructions" is a mathematical statement.
Thanks!
Comment by stunney — May 15, 2007 @ 3:16 pm
May 15th, 2007 at 5:05 pm
You affirmed the statement, The level of abstraction required to be considered a "code" cannot evolve, then justified your affirmation by defining "code" as being necessarily "intentional". In other words, you assumed your conclusion.
Just applying an analogous word such as "code" or "language" doesn't substitute for empirical investigation of actual phenomena.
Comment by Zachriel — May 15, 2007 @ 5:05 pm
May 15th, 2007 at 5:12 pm
The level of abstraction required to be considered a "code" cannot evolve.
Please provide a rigorous definition of "code" as you understand that term.
Comment by Zachriel — May 15, 2007 @ 5:12 pm
May 15th, 2007 at 11:18 pm
Zachriel wrote:
I wasn't there making an argument that codes are necessarily intentional. I was stating my belief that this is indeed eric's position, and at the same time affirming that it's also mine. Stating or affirming a belief is, as far as I know, still legal in the USA.
You were asking for clarification of eric's position, and I gave my observation on the question you asked, which was:
I wasn't 'assuming the conclusion' in any premise since I wasn't asserting a syllogism aimed at proving the conclusion.
Now that bring it up, though, my belief that codes can't evolve unintentionally from non-codes is based on two points, one of which is a priori, and the other of which is empirical:
a) they are abstract entities that essentially involve the normative concept of rules or instructions, and
b) my not being aware of any codes ever having been observed to have evolved uintentionally from any non-code in nature.
Now, if you disagree, and you know of any case of a code actually having been observed to have evolved unintentionally from a non-code in nature, then I'm all ears.
Exactly. So supply your example of an actually observed case of code's evolving unintentionally from a non-code.
Please provide a rigorous definition of 'rigorous definition'.
Meanwhile, do you object strongly to any of the following and, if not, will it do for present purposes?
Comment by stunney — May 15, 2007 @ 11:18 pm
May 16th, 2007 at 6:33 am
The level of abstraction required to be considered a "code" cannot evolve.
Which (depending on the many definitions of "norm") may be assuming the conclusion. If you use a definition that assumes agency, then it becomes a circular argument.
And is limited by your awareness, which could be due to technical limitations, current scientific knowledge, or even your personal willingness to consider the data.
In any case, that sort of evidence merely suggests that the level of abstraction required to be considered a "code" may not be evolvable. But eric claims it "cannot". This statement would still require specific empirical predictions of the posited 'Barrier'.
The statement is that The level of abstraction required to be considered a "code" cannot evolve, that there is a 'Barrier'. His term. It is not my job to provide evidence refuting the existence of mythical or posited beasts. From the posited existence of a 'Barrier', please provide specific empirical predictions that we can test.
Circular reasoning. You are presupposing that life is not such a system.
You would have to pare that definition down considerably. The "Genetic Code" may or may not meet a given definition of "code". Considering the amount of circular reasoning already demonstrated in this discussion, it is prudent to have clear and consistent definitions.
Comment by Zachriel — May 16, 2007 @ 6:33 am
May 16th, 2007 at 6:49 am
If the statement, The level of abstraction required to be considered a "code" cannot evolve is being introduced as a possible hypothesis, I have no objection; but it doesn't constitute a hypothesis until it can be shown to imply specific empirical predictions.
Comment by Zachriel — May 16, 2007 @ 6:49 am
May 16th, 2007 at 6:50 am
code: symbolic representation of a concept or physical entity in which the properties of the symbols are not dictated by the nature of that which is represented.
Comment by Bradford — May 16, 2007 @ 6:50 am
May 16th, 2007 at 5:59 pm
Zachriel wrote:
'May be assuming the conclusion' isn't a worthwhile assertion.
If codes are material entities, then yes, they might evolve naturalistically. And in a different thread, I actually gave two hypothetical examples of codes appearing to arise from non-codes.
It's called a complete lack of inductive evidence supporting your naturalistic hypothesis that codes arise unintentionally from non-codes. Do you guys have any intellectual integrity?
And I'll leave it as an exercize for the reader to figure out what logically follows if codes necessarily require that level of abstraction.
Yes, it would. And here's the empirical prediction: we shall not observe any such evolution.
It just needs one observed instance of a code arising naturalistically from non-code to falsify the hypothesis. Do you know of one? A 'yes' or 'no' will suffice.
I'm not. I'm merely pointing out that life or DNA code wasn't observed to have arisen naturalistically—maybe it did, maybe it didn't. Why do you hate actual observation so much?
And if you wan't me to stipulate that pre-biotic chemistry is a non-code or that it's a code, I'll be happy to do so either way for the purpose of the present discussion.
Let's say for the sake of argument it's a non-code. Fine. But the formation of the DNA code was not observed by anyone, as far as I know. But perhaps you are:
a) an alien from another planet,
or
b) a human time-traveler
or
c) none of the above.
If a or b, then perhaps you did observe DNA code forming or evolving naturalistically. And if so, please say so.
You've failed, completely and utterly, to demonstrate any circular reasoning on my part.
Use any definition of code you want to. Any whatsoever. Then specify an observed instance of a code arising unintentionally from a non-code, with 'code' defined as per your preferred definition, whatever your preferred definition is.
As I've already stated, I'm all ears. But it looks like you're simply playing a blatant shell game. You should be thoroughly ashamed of yourself.
Yes it does. It predicts that the number of observed cases of code arising unintentionally from non-codes in the next 500 billion or less years, will total up to zero.
Comment by stunney — May 16, 2007 @ 5:59 pm
May 16th, 2007 at 9:18 pm
The level of abstraction required to be considered a "code" cannot evolve.
Um, I've never claimed a complete theory of abiogenesis. The claim is that the level of abstraction required to be considered a "code" cannot evolve.
But we are not without some knowledge.
That is not a valid empirical prediction. The inability to discover evidence of a spontaneous origin of the genetic code could be due to technical limitations, the vastness of intervening time, or simple lack of human imagination.
Not knowing what causes the planets to trace complex orbits in the sky does not scientifically justify filling the Gap with Angels.
I have already indicated that there are no *such* coding systems that are directly observed to spontaneously assemble. (Also, I also provided an algorithmic definition of "code" similar to Bradford's above. I had been happy with the intuitive notion, but it seemed that eric was including agency in his definition, hence, reasoning circularly.)
The assertion "Maybe it did, maybe it didn't" is incompatible with the level of abstraction required to be considered a "code" cannot evolve.
By the way, you seem to have a sliding definition of the meaning of scientific observation. Here you seem to imply the observer would have to actually be on Earth billions of years ago.
And here you say that we can reasonably infer past events from observation.
Comment by Zachriel — May 16, 2007 @ 9:18 pm
May 17th, 2007 at 1:07 am
Zachriel wrote:
Do you have a counter-example?
Or do you have fuck-all?
Ah. The latter.
I see.
Who knew?
Why? Because you say so?
Or it could be due to you having no empirically observed counter-example.
Let me guess…
In other words…
You're devoid of a valid counter-example.
In other words, you're disgustingly devoid of intellectual integrity.
As I suspected.
I see. Any non such? Or is that beside the point?
What. Ever.:roll:
The Boloney-Meister strikes again!
Or be God.
I say nothing of the kind. You're making this up as you go along.
And I feel embarrassed for you that you're you.
Comment by stunney — May 17, 2007 @ 1:07 am
May 17th, 2007 at 7:28 am
stunney, I read your latest post, but can't find any actual content. Sorry.
I will go back to one of your previous statements. This is your so-called empirical test:
Let us know when the test has been completed. Then we can examine the data to see if it supports your claim. Meanwhile, by your own assertion, the test is incomplete.
Comment by Zachriel — May 17, 2007 @ 7:28 am
May 17th, 2007 at 8:04 am
Zachriel wrote:
Why do you consider this bit to be not 'actual content'—Do you have a counter-example?
See the bit where I say 'or less'? Well, that means we can repeat the test every year. All we need to do is check to see if any counter-example to the hypothesis was observed in the previous year.
But perhaps you have a counter-example already.
Comment by stunney — May 17, 2007 @ 8:04 am
May 17th, 2007 at 8:08 am
Because you had already asked it, and I had already answered it (more than once).
Let us know when the test is completed. Do we have to, like, do anything? Or just wait?
Comment by Zachriel — May 17, 2007 @ 8:08 am
May 17th, 2007 at 8:17 am
…
I'm look forward to publishing our results.
…
Only 499,999,999,999 years, 11 months, 29 days, 23 hours, 51 minutes, 33 seconds to go.
…
[twiddles thumbs]
…
I just thought of somethin'. You have to be alive to win the Nobel Prize.
…
mortal researcher: Doh!
…
I wonder what Oslo will be like then.
…
Comment by Zachriel — May 17, 2007 @ 8:17 am
May 17th, 2007 at 9:04 am
Zachriel,
See the bit where I said 'or less'? The test is not dependent on waiting for 500 billion years, contrary to your implication. We can repeat the test every year. All we need to do is check to see if any counter-example to the hypothesis was observed in the previous year.
But perhaps you have a counter-example already, in which case you can book a flight to Oslo now and have yourself fitted out for a coat with tails.
If you wish to take DNA code as your counter-example, I will stipulate that pre-biotic Earth chemistry satisfies 'non-code' for the purpose of the hypothesis, and I will stipulate that you needn't have observed the initial DNA code forming yourself. (NB—I never claim to have observed DNA code being intentionally formed. So if you use my non-observation of it being intentionally formed as a debating point, I reserve the right to withdraw that stipulation.) I shall also stipulate that all you need to do to 'provide a counter-example' is come up with an explanation of how DNA code was formed unintentionally that becomes the accepted fact of the matter within the scientific community.
Let us know when you're done. (You know, rather than have us have to stick a fork in you.)
Comment by stunney — May 17, 2007 @ 9:04 am
May 17th, 2007 at 10:21 am
The claim is that there is a 'Barrier'.
Argumentum ad ignorantiam, a logical fallacy in which it is claimed that a premise is true only because it has not been proved false or that a premise is false only because it has not been proved true.
Do we have to, like, do anything? Or just wait?
Comment by Zachriel — May 17, 2007 @ 10:21 am
May 17th, 2007 at 11:06 am
Zachriel wrote:
Asking for a counter-example isn't a fallacy of any kind.
Suppose I state that no human being can be taller than five light-years, and challenge you to provide a counter-example. what logical error would I be guilty of?
Here's something else from wikipedia:
The hypothesis is that codes can't evolve unintentionally from non-codes. One counter-example suffices to establish its falsity. Why the hypothesis is true is a different question from whether it's true. A single counter-example falsifies it.
That involves no argument or fallacy. Stating a hypothesis is not an argument for the hypothesis. I've already given two arguments for it. Those could be unsound arguments and the hypothesis still be true. But not if there's an observed counter-example.
I've indicated how you can go about showing that DNA code is a counter-example. Good luck.
Comment by stunney — May 17, 2007 @ 11:06 am
May 17th, 2007 at 11:44 am
No, but drawing a conclusion strictly from the lack thereof is.
A human taller than five light-years has testable consequences. We can safely conclude that humans are never that tall.
Quite so. A valid scientific hypothesis also has to have testable consequences, but as you are not making an argument in support of the assertion, I have no contention.
Comment by Zachriel — May 17, 2007 @ 11:44 am
May 17th, 2007 at 11:39 pm
If I take your meaning incorrectly, please clarify.
For illustration, consider whether one could take a random sequence of letters and evolve it into English, i.e. a language/code that associates sequences with meanings. You start with some random, gibberish sequence
SJFUYEHBVCCMDIOWOKDJJGHRFYSLHTUTYGK
How is selection supposed to work on this to shape it into a meaningful symbolic association?
As a trick, one could write a program that contains a well formed English sentence and then allow the original to vary, but keeping all the letters that fit the final target. Obviously this is a question begging approach that assumes access to the target.
Could a random sequence of symbols evolve into the code for a protein? No. This is not a gap. It is fundamentally not possible because the random sequence has no function that can be selected for "more like something that could be decoded into a protein".
This is made doubly true by the fact that there is no unique mapping from the symbol sequence to the realized form. Until one has a specific decoding mapping, it is meaningless and therefore inherently impossible to prefer symbolic sequences that "would decode" into a functional protein.
Remember that the realized function is not in the symbolic sequence. The function is in the realized form that a correct sequence is decoded into, e.g. the realized protein. Selection might operate on a realized form, but not on symbolic sequences with no mapping to the real world.
A random sequence does not decode into a functional protein, even ignoring the issue of where the decoding mechinery came from.
To create a symbolic code such as the genetic code, it is necessary to construct the necessary conversion machinery, both for encoding and decoding. To have at least one starting sequence, i.e. Alpha, the only possible source for the symbolic sequence is the model of a realized protein, which exhibits the realized sequence.
Comment by eric — May 17, 2007 @ 11:39 pm
May 18th, 2007 at 9:34 am
Let's try a slightly different version of the same experiment. Start with a population of identical short words. Using letter mutation and recombination, we allow the words to evolve. If a mutant is not a valid word (defined by looking the word up in a dictionary), then it is immediately culled. Knowing that longer words are rarer than shorter words, how long of sequences do you think will evolve? What if we allow whole word bits of phrases? To be, To be or not, or not to be. We could use a phrase book, such as the text of Hamlet, as a lookup table of valid phrases. How long of sequences do you think could evolve?
Comment by Zachriel — May 18, 2007 @ 9:34 am
May 18th, 2007 at 12:58 pm
Self-relicating molecules are not random.
RNA can act as a synthetase and catalyze the formation of simple enzymes (without a triplet code). There is very signficant evidence of "triplets having escaped from their original function in amino acid-binding sites to become modern codons and anticodons," and experimental evidence that "at least some primordial coding assignments appear to have exploited triplets from amino acid binding sites as codons," and that "the earlier appearance of tRNA versus their metabolically related enzymes" predicted by the RNA world theory has been confirmed.
Comment by Zachriel — May 18, 2007 @ 12:58 pm
May 18th, 2007 at 1:16 pm
Zachriel, I'm familiar with papers indicating a linkage between codons and amino acids based on the presence of a disproportionate number of codons in tRNA linked to the encoded aa. This does not apply to all twenty amino acids however and ignores the structual importance of tRNA loop structures in predicting linkage. In addition if binding affinity is a gauge then the binding affinity of the active sites of aminoacyl tRNA synthetases is instructive. Having said this, Eric's main point is still not addressed namely, linking a step by step construction of an encoding process to chemical necessity or a selection process. Which is it? If the former there need be more than the links shown as they are incomplete with regard to the whole code. If selection is the guide then what are intermediate selective values along the route to a code?
Comment by Bradford — May 18, 2007 @ 1:16 pm
May 18th, 2007 at 1:25 pm
You just shifted the burden of proof. No one claims to have a step by step mechanism. However, eric claims there is a 'Barrier'. The evidence gives no indication of such a 'Barrier' and several plausible pathways have been explored.
Comment by Zachriel — May 18, 2007 @ 1:25 pm
May 18th, 2007 at 2:27 pm
Why call them pathways much less plausible pathways. There is little detail other than a linkage of codons with AAs and even this is a suspect linkage with regard to the entire code which is the real point to whether or not a barrier exists. How do you construct a code piecemeal? The question is theoretical so you need not cite the absence of empirical evidence.
Comment by Bradford — May 18, 2007 @ 2:27 pm
May 18th, 2007 at 2:35 pm
Zachriel wrote:
How long would such an evolutionary process take to produce the complete texts, in the order in which they were written, of all of Shakespeare's plays?
Also, how can such a process generate the text of Hamlet so as to be able to 'use' the text of Hamlet as a 'lookup rule-book' for the 'validity' of word sequences?
This is another example of the Empire State building issue. Let's say the genetic code was designed. There would still be a physical description of what happened, whatever it was that physically happened, just as there would still be a physical description of what happened with regard to the construction of the Empire State building even if that building was intelligently designed.
Separately, if codes are essentially abstract entities, no physical description of material processes will capture this fact about them, since no physical description will capture the fact that a code, in order to be a code, must be something that is in principle an object of understanding; and 'understanding' is not something that can be captured by a physical description. For, with respect to whatever physical description of understanding one can come up with, we can always wonder why that physical state or process is identical with 'understanding'—in other words, we can always wonder if that description is correct or not. In other sciences, that kind of perpetual wonderment about whether a given theory is correct isn't possible. Nobody wonders in that way whether, say, chemistry is correct.
The ID issue is, in essence, posing the question to science raised by two central problems in philosophy: a) the mind-body problem, and b) the problem of other minds. Evolutionary naturalists are mostly materialists about both those problems, and believe that the answers given by neuroscience are the only answers there are, and the only answers worth having. So evolutionary naturalism ought to be able to solve these two problems in the sense of explaining the natural evolution of the objects studied by neuroscience.
But that's where the rubber hits the road. It's a hugely controversial claim that anything neuroscience could even in principle ever come up with would solve those two problems. But if evolutionary naturalism can't explain the emergence of rational minds, it is necessarily incomplete as an account of the living world. And that then raises the question of how complete an explanation it is with respect to life in general. For rational minds (and even consciousness in general) are a huge anomaly for the evolutionary naturalist project:
[Emphasis added]
Comment by stunney — May 18, 2007 @ 2:35 pm
May 18th, 2007 at 3:04 pm
The genetic code is not just an abstraction, but an observable chemical relationship between codons and amino acids. By insisting on including abstraction and understanding in your definition of "code", you have assumed your conclusions. Then you added a lot of words. (Addendum: I don't mean to suggest you haven't thought deeply about your position. But you have built your entire edifice on sand.)
As to the Empire State Building, we have the plans and evidence of the intelligent organisms that built it.
Comment by Zachriel — May 18, 2007 @ 3:04 pm
May 18th, 2007 at 3:47 pm
The scientific method. Form a hypothesis, in this case that the genetic code evolved. Use the hypothesis to make an implicit prediction. Test that prediction. Repeat the process to confirm and extend your results. What we are doing is predicting a path, then finding if we can fill in the blanks. At no time has any of these investigations revealed a 'Barrier'.
A code can be constructed piecemeal by coopting existing functions. RNA can synthesize enzymes. If there is an affinity of triplets to specific enzymes that have functions in the RNA World, then it can result in the beginnings of a code. Each of these aspects of the problem are not only theoretically conceivable, but can be investigated empirically.
Comment by Zachriel — May 18, 2007 @ 3:47 pm
May 18th, 2007 at 4:09 pm
I know what real pathways are. Organisms use a number of them to effect metabolic results. Their depiction in textbooks is accompanied by great specificity detailing distinct steps, their enzymes and substrates and end products. Your use of the term pathway is much too generous.
Studies indicate the "affinity" cited is numerical in nature and even this fails
with many AAs. True affinity is observed in the active sites of protein enzymes. You have not connected how repetitve production of SRMs leads to a code. There is discontinuity to the analysis. The data you cite is extemely ambiguous with respect to the claim made.
Comment by Bradford — May 18, 2007 @ 4:09 pm
May 18th, 2007 at 4:24 pm
Zachriel wrote:
That's a nice big baloney sandwich you have there.
I've never, ever, said or believed that the DNA code is 'just an abstraction'. So that's simply an invention of yours.
DNA code is an abstraction communicated or implemented in the medium of living material bodies.
That's a nice big baloney sandwich you have there. Because you missed the word 'if' in what I said.
That's a nice big baloney sand witch, sorry, sandwich covered in sand you have there.
What's the material description of 'understanding the DNA code'?
What's the material description of how that code, or any other code, arose without any prior act of understanding it occurring?
Have any scientific observations of something matching either or both of those descriptions actually being instantiated in reality ever explained 'code' or explained 'understanding' ?
And you've got the brass neck of accusing other people's position as being built on sand? Sheeeesh.:roll:
How does science decide that bits of paper with markings on them are 'plans', as against just markings on bits of paper determined by prior physical states of the universe? How does science determine that the activity of organisms which built the Empire State building wasn't an unintended, accidental by-product of the goal of seeking the types of food that were subsequently eaten in that building, given that this hypothesis is consistent with all the observational data? See also this post about the evolution of buildings.
Comment by stunney — May 18, 2007 @ 4:24 pm
May 18th, 2007 at 4:25 pm
Sometimes yes. Textbooks like the well-trodden trails. But many organic pathways are only tentatively understood. Yet, most scientists still think that cancers have a natural causes that can be discovered. Every discovery seems to lend support to that conclusion.
I never claimed there is a complete theory of abiogenesis. Just that everytime science looks, there is more evidence connecting different aspects of spontaneous abiogenesis. Enough so as to guide research leading to more insights. The usual.
The claim is that there is a 'Barrier'. Maybe there is, but no one seems to be able to justify that assertion.
Comment by Zachriel — May 18, 2007 @ 4:25 pm
May 18th, 2007 at 4:31 pm
As cancers can generally be described as breakdowns in existing cellular mechanisms assming natural causes to them is not controversial. But that is a poor analogy to a question of origins for the genetic code which presupposes the adaquacy of of an undirected process without the goods to back up the presupposition.
Comment by Bradford — May 18, 2007 @ 4:31 pm
May 18th, 2007 at 5:52 pm
Is that so? Do you merely assume that? On what do you base this conclusion? So you don't think that cells are possessed in order to punish those who indulge vices?
Cancers seemingly learn, such as activating the growth the neighboring blood vessels. Sure, every test so far indicates natural mechanisms, but there are significant Gaps. And there is a strong correlation between vices and cancer.
Perhaps, we should consider other examples. Any issue of Microbiology will demonstrate the vast ignorance of many organic mechanisms. They publish plenty of examples of how scientists assume a natural cause (the hypothesis) in order to test the empirical implications of the assumption. Is there crosstalk on the CDA biosynthetic pathways? Is the cytoplasmic loop of XcpS involved in metabolic stabilization? Does SeqA modulate DNA replication? Burning questions to microbiologists.
Sure, whenever anyone looks, they discover natural mechanisms. But maybe when nobody is looking … (Hah, I'm just pulling your flagellum.)
–
The claim is that there is a 'Barrier' to evolution of the genetic code. Maybe there is, but no one seems to be able to justify that assertion. If it's empirical, just show us the evidence. If its logical, then an actual proof would be nice.
Comment by Zachriel — May 18, 2007 @ 5:52 pm
May 18th, 2007 at 7:26 pm
Bradford: As cancers can generally be described as breakdowns in existing cellular mechanisms assigning natural causes to them is not controversial.
No, I think your remark is a stupid comment that flows from a stereotypical mindset.
Comment by Bradford — May 18, 2007 @ 7:26 pm
May 18th, 2007 at 9:13 pm
If one thinks about this proposal a bit, its not difficult to see that this is a question begging illustration at multiple points.
First it assumes you can start with words. Words are already meaningful symbol sequences that code for the actual items they represent. The word "mountain" codes for a mountain, though the word itself is not a mountain.
By starting with words, you have literally started at the goal. That is pure question begging — assuming the objective. From there you also suggest the availability of dictionaries and phrase books which can identify legitimate coded symbol sequences, which is more question begging. Sorry, but that is not a legitimate illustration.
The goal is to get to something like the first word — a symbol sequence associated with something that is real and functional. To implement a code you need
a) symbol sequences
b) the real items they represent, i.e. that they code for
c) the machinery that implements the association, e.g. encoding, decoding, etc.
NOTE: We cannot realistically hope to have symbol sequences that magically match the realized sequences without also having the coding machinery that connects them. The coding machinery is necessary. It is the implementation of the association for the code.
Consequently, the question arises, could you start with sequences of symbols and build toward meaningful realizations? Or is that a dead end?
I did not say that so-called "Self-relicating molecules" were random. A key point you may have missed is that I did not say that the symbols themselves were internally random. What I discussed was the futility of starting from a random sequence of symbols. The code expresses meaning using a sequence — one that is neither regularly ordered nor random.
The point I was making remains true. You cannot start from a random sequence of symbols and hope to evolve into meaningful symbol sequences. This is important because it rules out one class of options.
In your analogy you alluded to dictionaries and phrase books. But in the pre-Alpha world, the only source where you could possibly look up the correct sequence of amino acids for a functional protein would have been to use an actual functional protein as the model and encode back to the symbol sequence, thereby converting the actual sequence into its symbolic equivalent. In the pre-Alpha world, preexisting proteins combined with encoding and/or decoding mechanisms would be as close as you could get to having a "dictionary" for protein sequences.
Comment by eric — May 18, 2007 @ 9:13 pm
May 18th, 2007 at 10:23 pm
Do replicating molecules shrink the Gap I have been talking about at all? No. Do they cross over across the Language Barrier? No. That is why I specifically invited Zachriel to just assume their existence.
[If others will indulge me for a moment, I will briefly consider but ultimately pass over the matter that discovery of "catalytic properties of RNA" typically does not mean truly "self" replicating molecules, but rather peer-replicating molecules. Consider the difference between trying to scratch every spot on your own back vs. scratching the back of your neighbor. For more on the troubles of RNA, including claims of "self" replication, see here, here, and here.]
For now, let us wave all peer replication problems away. Let us assume that a population of peer-replication molecules appears. Let us also assume that in the course of replication, some variations arise that do not prevent further replication. So PRM1 leads to PRM2, PRM3, … PRMn, where n is a big as you want or as the limitations of this this kind of replication (without encoded instructions, based on physical replication of templates) allows. Take as much time as you like, up to billions of years. Assume there is no shortage of raw material to feed the replications and that favorable conditions for replication prevail throughout.
However long that chain is extended, PRMn is still a physically replicating molecule, and a copy of it is still another physically replicating molecule. It is not a symbolic code and not one step in that chain gives us a symbolic code. You don't have any symbol sequences that by their sequence mean something else.
Whereas the word "mountain" represents a mountain without being a mountain, and a sequence of codons can represent an amino acid sequence without being one, everything in that pre-Alpha replicating picture above is still just itself.
This is true even if one attaches various other atoms according to chemical attractions (albeit, within the limitations of continuing to be able to replicate by this method). Every chemically motivated attachment is an attachment according to the laws of chemistry, not determined by the instructions of a symbolic code. Replicating such an attachment does not change the picture.
You could cross the universe back and forth and observe all manner of molecules obeying the dictates of chemical laws and chance, but language is neither of these.
Until you have symbol sequences that symbolically express an associated meaning by their sequence — sequences that are covertable to functional realized counterparts unlike themselves — you haven't broached the Language Barrier.
Replication of molecules does not decrease the size of the Language Barrier at all, because there is nothing symbolic about physical replication.
NOTE: Nor do unguided natural processes have any observed need, motivation, or tendency to become symbolic. The conjecture that they could or would go symbolic at present rests on an appeal to faith. But so far as we see, nature is quite content to move along without inventing sequences with attached associated meanings.
Comment by eric — May 18, 2007 @ 10:23 pm
May 18th, 2007 at 11:50 pm
You introduced the idea of a simulation using language. We would need to have an algorithm (such as a phrase-book or grammar-checker) to determine if a given sequence is a meaningful English phrase.
A word could be a single letter. Also, there has to be a fitness landscape of some sort. In nature, it is the environment an organism finds itself in. But, we can drop simulations for now.
That is clearly a mathematical claim. You keep making loosely constructed arguments. I would like to see the formal proof. Please point to the appropriate mathematical journal, or name the Theorem.
Comment by Zachriel — May 18, 2007 @ 11:50 pm
May 18th, 2007 at 11:59 pm
eric wrote:
Eric, yup. And, indeed, I hear ya.
But I'm afraid it doesn't matter what you say.
Zachriel has already had it explained to him over and over: he simply refuses to let it sink in. He won't even answer questions the fourth or fifth time you ask him, preferring instead to answer questions that he wasn't asked.
He's a fundie. Quite a sad case of one too.
Comment by stunney — May 18, 2007 @ 11:59 pm
May 19th, 2007 at 9:27 am
Before that can be expressed mathematically a quantifiable definition of evolve is required.
Comment by Bradford — May 19, 2007 @ 9:27 am
May 19th, 2007 at 9:37 am
Yes, your argument starts with a premise,
To which I responded.
As the primary premise of your conditional is false, the syllogism fails.
You have also said that
As the genetic code is a material entity, it might evolve naturalistically, probably by cooption. The hypothesis leads to specific empirical predictions that can be tested, e.g. there is signficant evidence of "triplets having escaped from their original function in amino acid-binding sites to become modern codons and anticodons."
Comment by Zachriel — May 19, 2007 @ 9:37 am
May 19th, 2007 at 9:45 am
Excellent point!
(Eric, it might help if you attempt to prove that it *can* happen in very simple cases, then extend your proof to more general systems. A reasonable fitness landscape must be devised"”yet be general enough to account for a wide variety of such systems. It is not as easy a problem as it might appear on the surface. I look forward to the formalities supporting the claim.)
Comment by Zachriel — May 19, 2007 @ 9:45 am
May 19th, 2007 at 9:58 am
Codes are expressed through matter and energy like the one you are looking at. When a claim is made that a code evolves a selection process must be specified. That means specifying in advance what types of changes would become predominant in a population and why. A major difference between self-replicating RNA and cells is the lack of control of the former over environmental substances and their metabolism. The capacity of metabolism by organisms ensures that cells have both the needed biomaterial and energy to enable cellular replication. Those factors are outside the control of SRMs. An increasing population of SRMs are dependent on their environment for a continuous supply of ribose, nucleotides and phosphate groups. Origin theories are contingent on favorable environmental conditions with respect to their supply. As that supply is highly dubious, as contrasted to the supply of sunlight enabling photosynthesis, parallels between the evolution of cells and SRMs are compromised at the outset.
Comment by Bradford — May 19, 2007 @ 9:58 am
May 19th, 2007 at 10:04 am
There is no complete theory of abiogenesis, and the origin of the genetic code is still uncertain.
Those are very important issues that have to be resolved before anyone can claim a valid theory of spontaneous abiogenesis.
Comment by Zachriel — May 19, 2007 @ 10:04 am
May 19th, 2007 at 10:08 am
An undirected confluence of natural forces should produce a predictable order of symbols. Sequencing of encoding symbols needs to be free of the determinism a chemical process would confer.
Comment by Bradford — May 19, 2007 @ 10:08 am
May 19th, 2007 at 10:49 am
That would be a reasonable prediction from the RNA World Hypothesis. From Bartel's RNA Theme Park, a new ribozyme that can use a template RNA to create a third RNA with high accuracy and not restricted by sequence content. Still limited to adding 14 nucleotides to an existing sequence, but very flexible, and demonstrates the principle. Additional sampling and evolution should generate more efficient ribozymes.
Comment by Zachriel — May 19, 2007 @ 10:49 am
May 19th, 2007 at 2:20 pm
Your response alludes to flexibility in molecule replication. But this confuses the matter of whether replication can happen (regardless of the sequence content being replicated) vs. whether there are issues of chemical determinism in the sequence content itself. It misses Bradford's point, which is correct and is not addressed by whether a sequence can be duplicated.
To illustrate, codons within DNA are free to encode for any sequence of amino acids because there are no chemical restrictions on the order of the base pairs in DNA. Do you intend to claim that the internal structure of RNA is similary free to be rearranged in any fashion (and still retain peer-replication ability)?
In any case, replication is inherently inadequate.
If the goal is to get to the moon, a hot air balloon cannot take you there.
If the goal is to traverse an ocean, a car cannot take you there.
If the goal is driving cross country, riding the Replication merry-go-round cannot take you there, no matter how long you ride.
The word "mountain" is not a mountain, and
a codon sequence in DNA is not an amino acid sequence for a functional protein, but
a replicated RNA molecule still gives you an RNA molecule.
RNA does not represent RNA. RNA is RNA.
A coded sequence of symbols represents something other than itself. If you want to travel cross country, you have to find a way off of the Replication merry-go-round.
For a bit of thought stretching, and to catch a bit of the flavor of the issue (although not an exact comparison because of distinctions between visual images and coded sequences), as long as one maintains awareness of the differences it might be of some help to consider the famous painting The Treachery of Images.
p.s. For some extra fun.
Comment by eric — May 19, 2007 @ 2:20 pm
May 19th, 2007 at 6:42 pm
The issue about a word is not primariy its length. It is the fact that words already have an associated meaning, which is question begging regardless of the length of the word. How does the first word (however short) get an associated coded meaning?
I trust you are not opposed to logic. Before one can apply mathematical analysis to a proposal, one must first have a sufficiently well-defined and meaningful proposal to analyze. I wouldn't expect anyone to be able to publish a paper showing that self-evident nonsense is nonsense.
Let's see how close we can come to a proposal that has at least prima facia plausibility and sufficiently well-defined content such that it could support analysis.
For the present purposes, let us generously set aside for the moment any problems with creating or duplicating a sequence of "items" suitable to become symbols, but without any present associated meanings (which would be question begging). Further suppose for now that their sequence is not constrained by any chemistry requirements.
After some time, this sequence changes to some other sequence (perhaps due to irregularities in duplication). Maybe later this happens again.
Now what?
We could take more trips around this merry-go-round, but so what? We just come back around to a sequence of items. What is there here to analyze?
But a sequence of items is not a living organism and the sequence does not have function with regard to the desired end goal of a code. Example: codons in DNA don't function as proteins, even though they represent amino acid sequences. Until decoding mechanisms exist, a sequence does not function as a code either.
If you meant that "there has to be a fitness landscape of some sort.", that is false. There does not have to be one just because a hypothesis wants one. If on the other hand you were meaning "there has to be a fitness landscape in order for this to work.", that may have some truth to it, but it works against your hopes because there is no such landscape for the sequence as a coded sequence.
But let's try looking for something anyway.
Suppose you wanted landscape or selection process to shape random droplets of ink on paper to read "Eat at Joe's". Well, you could use a stencil. Except that a stencil that already reads "Eat at Joe's" is a question begging assumption. It assumes the message in advance.
Well, suppose that there were variations in the paper such that some parts turned color and other parts didn't, as is done with invisble inks such as heating lemon juice on paper. Except that, once again, this assumes that the pattern of application of the invisible ink on the paper already said "Eat at Joe's".
The problem here is fundamental and cannot be removed by advance of knowledge. A complete knowledge of the physics and chemistry of paper and ink still could not determine the origin of a single coded message. The message is independent of the laws governing the medium. Attempts to coax nature into saying the message invariably suppose that the message already exists.
All these examples also pressume the prior existence of an interpreting mechanism. One configuration of items is just as nonsensical as another — until they can be decoded into some realized form. A single string of letters could mean one thing in one language or something else entirely in another. The same is true for genetic codes. A codon sometimes means one thing in one genetic code and something else in another. Without an ability to decode to a realized meaning, there cannot be a selectable function for meaning because meaning would be undefined.
The catch 22 is that nature has no reason to construct a mechanism for decoding encoded symbols prior to the existence of encoded symbol sequences to decode. Such a mechanism would be useless in a universe devoid of appropriate encoded symbol sequences.
Comment by eric — May 19, 2007 @ 6:42 pm
May 19th, 2007 at 7:24 pm
Yes, the template can be arranged in any fashion. That's why it is considered an important result. (And one sought after by the researcher as a prediction of RNA World. Funny how he knew.) The RNA uses another RNA as a pattern for producing, one nucleotide at a time, a new RNA"”regardless of template sequence. Of course, this isn't a complete theory. As Bartel said, his RNA World Theme Park is "artificial and fragmented when compared with the real thing, but still well worth a visit."
Comment by Zachriel — May 19, 2007 @ 7:24 pm
May 19th, 2007 at 7:57 pm
Well said Eric and also an apt description of how real genomes function. A distinct information molecule able to code for multiple end products is the right model. The illustrations were comical.
Comment by Bradford — May 19, 2007 @ 7:57 pm
May 19th, 2007 at 8:55 pm
Zachriel wrote:
To which I responded by saying:
Zachriel also wrote:
Let me understand what you're saying. The premise is conditional. 'If codes are abstract entities, then they can't evolve unintentionally from matter'. Are you claiming
a) that codes are not abstract entities?
or
b) that even though they are abstract entities, they can still evolve from matter?
Please clarify. And then when you've done that, please substantiate, even just a tiny little bit, your assertion that the conditional is false.
Just because computers and speech and microbiological entities and writing on paper are all physical media by which the rules of Latin grammar, geometry, algebra, the genetic code and computer programming languages are expressed, one can't weigh, determine the position, or measure the velocity of the rules of Latin grammar, geometry, algebra, the genetic code, or computer programming languages. They emit no radiation, don't consist of charged particles, and can't be rotated in space. Any supposition to the contrary would be to commit a category mistake.
It's you who are guilty of that. Such as you did the other day when you tried in the most diabolically cunning way to constrain a conclusion concerning minds by defining them in terms of material brains.
In fact, if you recall, I steadfastly refused, precisely so as not to be guilty of doing what you allege, to give a definition of 'code' because I knew that intellects as depraved as yours would denounce my definition for smuggling in my conclusion; and so I invited you and raevmo to use any definition you cared to, and then answer the question of whether code, so defined in your preferred way, could unintentionally arise from non-code. I entreated you. I practically went down on my knees and begged the pair of you to define code in any way you wanted.
Then, in desperation, I simply offered the wikipedia definition because I was sick and tired of your intellectual dishonesty and the twisted and dementedly incoherent way you kept attacking me for defining code to include my conclusion and
for not defining code.
You've failed so far to demonstrate any circular reasoning on my part while being outrageously guilty of it yourself.
You've failed to prove even one observed counter-example to the hypothesis that codes can't evolve unintentionally from non-codes. (I even stipulated for your benefit that pre-biotic Earth chemistry satisfies 'non-code' for this purpose, and that showing a way that DNA code evolved that would satsify the scientific community at large would qualify as a counter-example)
You've also failed abysmally to answer any of the following pertinent questions:
How long would such an evolutionary process take to produce the complete texts, in the order in which they were written, of all of Shakespeare's plays?
Also, how can such a process generate the text of Hamlet so as to be able to 'use' the text of Hamlet as a 'lookup rule-book' for the 'validity' of word sequences?
What's the material description of 'understanding the DNA code'?
And you have the cheek to accuse others of assuming their conclusion in their premises? Shhhhhhhhhhhheeeeeeeeeeeeesh.
I said, "If codes are abstract …."
You complain about assuming my conclusion, and then you say: "As the genetic code is a material entity".
How fucking hilarious can you get? Honestly, you should raffle yourself.
Eric expressed the problem succinctly:
"The message is independent of the laws governing the medium. Attempts to coax nature into saying the message invariably suppose that the message already exists."
Comment by stunney — May 19, 2007 @ 8:55 pm
May 19th, 2007 at 9:07 pm
Eric, when you get a chance please visit this location.
Comment by Bradford — May 19, 2007 @ 9:07 pm
May 19th, 2007 at 9:17 pm
Hi Zachriel,
Just a word of encouragement. I think you are doing pretty well. While stunney is relatively predicable, Eric and Bradford are tag teaming. Eric's claim of the existence of an insurmountable Language Barrier implies a burden of proof. But when you hone in on that, Bradford swoops in an puts you on the defensive by asking for your proof, and just as you get traction in explaining the reasonable interpretation of the RNA World, Eric suddenly returns with his insurmountable barriers again.
I wouldn't continue playing their game, but you seem to be handling it ok.
I am still waiting for Eric and/or Bradford to take a position on nanobes. link
It would be interesting to see their definition of life in practice. There is enough information in the ten page paper that its life status should be readily apparent. I'm not asking them to declare it organic, just living.
It's just a small step in verifying whether or not their definitions are assuming their conclusions. A bigger step would be to talk about whether the term "inorganic life" is a total impossibility. (e.g. Robotic life)
Provoking Thought
Comment by Thought Provoker — May 19, 2007 @ 9:17 pm
May 19th, 2007 at 10:27 pm
TP, one more comment like this and you get body slammed.
Comment by Bradford — May 19, 2007 @ 10:27 pm
May 19th, 2007 at 10:30 pm
Of course. Your claim was you cannot start from a random sequence of symbols and hope to evolve into meaningful symbol sequences.
Replication and abstraction are implicit in your claim which was general and not specific to biogenetics.
You apparently are using a different definition of "evolution" than that used in biology and informatics. As Bradford suggested, you need to define your terms carefully. Evolution includes replication, variation and a selection mechanism.
Nor did you provide a proof of any sort. Perhaps a cite to a general theorem.
Comment by Zachriel — May 19, 2007 @ 10:30 pm
May 19th, 2007 at 10:37 pm
Thought Provoker wrote:
This from the guy who trots out:
"I don't know know the Truth, do you?
"Let's do science.:mrgreen:"
"What is your proposal?"
"OMA/NOMA yada yada yada"
ad nauseam!:grin:
TP, let me assure you, there are no more predictable posts than your own. Yea and, indeed, verily.
Comment by stunney — May 19, 2007 @ 10:37 pm
May 19th, 2007 at 11:01 pm
But the genetic code is *essentially* an observed chemical process. And the physical description of their material processes completely captures their entire activity, and they work regardless of anyone's 'understanding'. Understanding and abstraction is what people do.
You didn't provide a reference, but I believe my statements concerned intelligence, not mind.
This is one reason why the discussion never progresses. I have already indicated several times that there are no *such* coding systems that are directly observed to spontaneously assemble. I also provided an algorithmic definition of "code". I have also repeatedly stated that there is no complete theory of abiogenesis.
It is known to have taken a few billion years.
Comment by Zachriel — May 19, 2007 @ 11:01 pm
May 19th, 2007 at 11:07 pm
Thanks for the tips, but I don't play games and I don't keep score. The discussion speaks for itself.
I'm still waiting for a valid cite to support eric's claim that "you cannot start from a random sequence of symbols and hope to evolve into meaningful symbol sequences." But apparently there's some definitional problems with "evolution", "symbols" and "meaningful".
Comment by Zachriel — May 19, 2007 @ 11:07 pm
May 19th, 2007 at 11:15 pm
Hi Zachriel,
You wrote…
ROFL
NO KIDDING?!?
And don't forget the definition of "life". I think there might be a little fuzziness there too.
Now look what you made me do. I really, really try to resist sarcasm in posts.
Regards,
TP
Comment by Thought Provoker — May 19, 2007 @ 11:15 pm
May 19th, 2007 at 11:53 pm
eric wrote:
Eric, I am always struck by those in the 'bright' community who a) revel in their self-alleged superior powers of reasoning, and b) labor tirelessly to propagate the view that rationality is an illusion caused by purely physical states of human organisms resulting from a series of evolutionary accidents. They're like people who see a sign saying WELCOME TO SCOTLAND made from boulders on a hill beside a railroad, which arrangement of boulders they attribute to chance, unintentional, material processes; and yet despite their explanation of the boulder arrangement, they conclude that they've now just crossed the border into Scotland.
The same bizarre intellectual dissonance emanates from 'brights' when they start prattling on about the probable, in their view, unintentional origin and nature of codes (for even Morse code is, on their view, simply part of the materialist's natural continuum, minds being naturally evolved brains and all, dontchaknow).
A vital question here, as you clearly grasp, is the question of intentionality, or 'aboutness', or representation. This is central to the origin and nature of codes. How can one bit or state of matter intend or represent or be about another bit or state of matter?
What gives great difficulty to all materialist proposals is that representation can be more or less accurate, and it needs a mind to judge how accurate some representation R is of S, whether R represents S or S represents R, and whether R or S represents anything at all.
If my height in my teen years increased by an identical amount to that of Mount Everest in the same period, would my increase in height represent Mount Everest's increase in height, or vice-versa, or did neither my increase in height nor Mount Everest's increase represent anything at all?
The essential point is that neither my changing stature nor Mount Everest's would be able to tell or settle the matter just by their physical selves. Only a rational mind could do so.
The key point materialists are missing is this: it is intrinsic to the very idea of representation that it can fail to be adequate as a representation, that it can be more or less good as a representation; and that you need some irreducibly mental properties to discern that.
The thermometer, to return to that spectacularly lame example, might be faulty. It might misrepresent the temperature. It might not be working at all. But how is the thermometer supposed to know if it's faulty, accurate, or not working at all? It doesn't and it can't. Thus it might not be successfully representing anything. But just by itself it cannot act rationally to find out. That's why you need irreducibly mental intentional states for representations, and hence for codes too.
If a thermometer gauge accidentally but accurately predicted the ups and downs of the stock market, would we say it was a natural and purely 'material representation' of the stock market? Puhleeze.
Paintings and photos are representational not because they possess mental properties, but because intentional properties possessed by minds are involved in what makes those paintings and photos representations. Mental properties have to be involved for something to be a representation. They're not possessed intrinsically (only extrinsically or derivatively) by, say, a thermostat or a computer program. By contrast they're possessed intrinsically by the agent who intentionally designed the thermostat or designed the computer program, because they're possessed intrinsically, and irreducibly, by rational minds.
I've given previously the example of a world which only ever had one unchanging thing in it, something that just happens,/i> to be shaped exactly like Winston Churchill in his famous cigar pose. Since that world never has minds, that thing does not represent Churchill (unless it is a world that was designed by an intentional agent who transcends the world, who obviously possesses mental properties, and the designer intended that world to resemble Winston Churchill).
I reiterate that for any designating of degree of representational accuracy, you need things that are capable of assessing representational accuracy (i.e. checking for errors). They don't need to be infallible or perfect in their judgements. They just need to be able to assess representational accuracy.
Quality control technology in a factory might be thought of as being capable of assessing representational accuracy; but only if it is already the product of intelligently design, and only if it's working properly. To assess whether it's working properly, or whether its purpose is quality control at all and not something else, you will at some point along the line need something with a mind possessing original and irreducibly intentional capacity, and who can assess representational performance. Such as an engineer.
I hope the relevance of these remarks to the question of codes is evident.
A bit more from a previous post of mine on this topic, which I also think is highly relevant to the question of the origin and nature of codes:
Comment by stunney — May 19, 2007 @ 11:53 pm
May 20th, 2007 at 12:01 am
That may be true for mathematical formulations but there is more to this.
It is the last of the three that engenders confusion. Much is made of the fact that RNA comes in enzyme varieties. One can select for this and that and eventually might derive RNA representing all major enzyme groups. Selection value is linked to the utility the enzymes afford a cell. But that's the problem. Cellular selection qualities are relevant to cells which are non-existent. What really is needed is a demonstration that RNA enzymes are more chemically stable than other environmental substances. Their existence within cells is secured by some distinct cellular mechanisms that would not be present in a prebiotic environment. Before selection can be invoked on an empirical footing it needs to be shown at the level it is invoked at i.e. a precellular world. Arguing that cellular selection qualities are the determining standard assumes the conclusion.
Comment by Bradford — May 20, 2007 @ 12:01 am
May 20th, 2007 at 12:06 am
Stunney you must be a fast typist. You probably lead the blog in length of posts if not number.
Comment by Bradford — May 20, 2007 @ 12:06 am
May 20th, 2007 at 3:05 am
Zachriel wrote:
You're at it again. 'It' being doing what you accuse others of doing, namely smuggling in your conclusion by definitional fiat. Don't think we don't spot it. 'IS (NOT MAYBE, IS) ESSENTIALLY A Chemical Process'. See? Spotted it.
One may as well say that 'human language is essentially an observed chemical process', since human existence and hence human language essentially depend on chemical processes. And one could add that human language is essentially an observed solar process since it depends on the sun shining. And then one could say human language is essentially a Big Bang process or a Multiverse process since it depends on the Big Bang or the Multiverse. Or even, if one was really bold, that human language is ESSENTIALLY a, er, process.
It would be extremely unenlightening and stupefyingly silly to say such things, of course. But yes, one could say such things.
Oh shit, we're back to the Empire State building.
Does the physical description of the material processes involved in its construction capture that entire activity? Does the physical description of Shakespeare composing all his literary work capture that entire activity? Does the physical description of a programmer inventing a new programming language capture that entire activity? Does the physical description of a scholar translating a text in Sanskrit into Finnish capture that entire activity? Does the physical description of Godel proving his celebrated incompleteness theorems capture that entire activity? Does the physical description of a NASA engineer solving a differential equation capture that entire activity? Does the physical description of being in pain or of seeing that a tomato is red capture that entire activity? Does the physical description of you making a logical error capture that entire activity?
That's not true, in my view. At best you are begging the question.
DNA code only works, in my view, because it was designed to by its designer who understood it thoroughly. Without the designer's understanding, it wouldn't even exist, let alone work. As you may have noticed, much of this thread has involved contributions from folks who strongly doubt whether a code could even arise unintentionally from non-code stuff. Hence the hypothesis that the origin of all codes must be an intentional designer.
Hey, why do you hate Copernicus so much? Quit with the anthropocentrism. It's what any rational mind does. Human, alien, angelic, divine.
Right. I forgot that you're still unsure if there are minds.
Previously, in another thread, you daringly asserted that there is some evidence for the existence of minds. But perhaps you now feel you were a trifle incautious before.
So perhaps now we should interpret you as saying that there's some evidence that the Empire State building involved intelligence in its construction, and that intelligence involves brains.
Who knew? Did anyone here know that? Why wasn't I told?
Not, of course, that your trying to assume your conclusion with an implied premise that if something's intelligent, it's necessarily physical. Of course. You wouldn't stoop to such a low-down dirty trick like that, oh no.
Sane people believe we have a vast amount of evidence for humans having irreducibly mental properties (such as a capacity to understand the concept of evidence). But I guess you'll just ignore it because such properties don't lend themselves to physical observation. Rather than conclude that physical observation is too limiting to be a defining criterion of rational inquiry, like a sane person would, you become so ridiculously slavish to that criterion that it forces you to pretend some uncertainty about whether scientists, logicians, and mathematicians have rational minds.
As a wise man once said, there's crazy. And then there's batshit crazy.
So, in recorded history the hypothesis that codes cannot evolve unintentionally from non-codes hasn't been falsified. Still, we might discover a counter-example. We might discover a fairy as well.
But phew, I'm certainly glad that's settled.
That is false. It is most decidedly not known that such an evolutionary process was sufficient to produce, or did produce Shakespeare's plays.
Talk about begging the fuck out of a question.:roll:
Here is what you wrote:
My question was:
How long would such an evolutionary process take to produce the complete texts, in the order in which they were written, of all of Shakespeare's plays?
So, just to be sure: is your answer really that the process you hypothesized (as per your above-quoted statement) was really the actual pathway that resulted in Shakespeare's plays being written?
Or, to put it another way:
Are You Completely Bonkers?
(PS–The other question was: Also, how can such a process generate the text of Hamlet so as to be able to 'use' the text of Hamlet as a 'lookup rule-book' for the 'validity' of word sequences?)
Comment by stunney — May 20, 2007 @ 3:05 am
May 20th, 2007 at 4:32 am
One more thought, sorry, physical process before I hit the hay. It's a couple of questions.
Zachriel had said:
It's worth asking, Zachriel, because your formulation is suggestive of a rather severe materialist reductionism:
Given that all biological processes apparently reduce to subatomic particles of matter interacting with each other and with waves of radiation in spacetime, is it fair to say that biology is essentially physics?
And if it is, is it also fair to say that the genetic code or any other code does not really exist, given that non-living matter, living matter, and even abstract human thinking (such as identifying and understanding 'the genetic code') must all be essentially physics?
It strikes me that affirmative answers to these questions entail the view that the lines between non-life and life, non-thought and thought, and between codes and non-codes, are essentially arbitrary, and justified, if at all, merely instrumentally.
Care to comment?
Comment by stunney — May 20, 2007 @ 4:32 am
May 20th, 2007 at 8:10 am
Eric's claim is general, not specific to RNA; and that this general rule concerning evolution, meaning and symbols. Concerning your empirical quetions, and remembering that there is no complete theory of abiogeneis:
Many scientists do not think that RNA was the first replicators. Related molecules, such as PNA and TNA have been proposed. Others think metabolism preceded replication. I agree with your previous suggestion that membranes are a prerequisite for any reasonable scenario. Nevertheless, as you pointed out, there are significant problems for each of these proposals.
Various assumptions are made in order to test their implications. For instance, Eric assumes replication in order to test his Language Barrier idea. (On a journey West to find a route to the Pacific Ocean in North America, we might assume we know how to get to St. Charles, Missouri, but hypothesize various routes further on. Meet you in St. Louis, Louis!) But if a complete explanation for spontaneous abiogenesis is possible, then each of these assumptions will have to be filled with supported theory.
Comment by Zachriel — May 20, 2007 @ 8:10 am
May 20th, 2007 at 8:48 am
Yes!! Thank you. Just as when you said the genetic code was "essentially abstract entities". See? Spotted it.
No one knows. There is no complete theory of cognition. By contrast, there is a reasonably complete chemical theory of transcription.
I will rephrase that. There is no more scientific evidence that transcription would cease in the absence of someone 'understanding' the process than there is that water forms crystals when cold.
There is no more scientific evidence to support that than there is that angels push planets on crystal spheres while singing Hallelujahs. Maybe it's true, but there's no scientific evidence to support the claim.
We can certainly suppose that there is a Divine Abstractor. However, there is no scientific support for such a view.
I'm rather surprised you still haven't got this basic point. We infer intelligence from actions; memory, learning, problem-solving, self-consciousness. You may assume a black-box. However, I would note that large-brained apes are more intelligent than small-brained apes.
I'm not sure why you think I am uncertain. The scientific and philosophical evidence is convincing. It's only a few philosophers who fret.
Nor has it been supported. This again reveals your misunderstanding of the scientific method. A hypothesis is a tenative assertion held for the purposes of testing specific empirical predictions.
There are scientific reasons to believe that life spontaneous originated on the primordial Earth. That's what is being discussed; in particular, Eric's claim that there is some sort of 'Barrier' which he says he can deduce from the nature of evolution, symbols and meaning.
There is strong scientific evidence that Shakespeare is a type of evolved ape, and that his abilities are the result of a long period of cultural evolution. However, whether Shakespeare also received Divine Inspiration is not currently subject to any plausible scientific investigation. From a scientific view, there is no evidence of Divine Inspiration. But there is a sufficient Gap to hide your metaphysics.
Comment by Zachriel — May 20, 2007 @ 8:48 am
May 20th, 2007 at 9:33 am
This may be your best post yet; asking pointed questions to try and understand another's point-of-view.
From a scientific vantage, most biological processes appear to be
emergentfrom physical properties. Élan vital has been generally discarded in science as having no explanatory power. However, there are significant gaps in scientific knowledge with regards to cognition and thought. There are also a few unresolved philosophical questions (which you touch on, but then seem to conflate with other issues).Why would that imply the non-existence of the genetic code? It is an observed chemical process. It's as real as planetary orbits which, after all, are just maps. (There are no crystal spheres.) Science isn't as austere as your metaphysics, but is practical. We have scientific reasons to believe that planets exist, that they move, that we can draw an eclipse and say that this represents their orbit, and then use that map to make predictions. So do planets orbit? To an austere philosopher, it might have to do with what we mean by orbit. In science, it's just an empirical question.
Science is as science does.
We observe some aspect of the natural world. We form a generalization concerning those observations. We then deduce from that generalization a distinguishing empirical test. We then verify this prediction with new observations. We modify or discard our generalization as necessary, all the while shaving off extraneous assumptions. We communicate our results so that other observers can replicate and extend our findings. We continue this process and eventually build confidence in our model.
Comment by Zachriel — May 20, 2007 @ 9:33 am
May 20th, 2007 at 10:36 am
Just to make sure I understand your intended meaning, you also said,
That looks like a claim, but the exact claim depends on the definition of "abstraction", "communicated", etc. you are using, and whether they implicitly include agency. You could try to approach this issue by explaining what you mean by abstraction.
Well, here you out-and-out claim it is agency. No one has provided scientific evidence that such is the case. You could just as easily say that "water freezes because it was designed to by its designer". While possibly true, it does not represent a valid scientific claim.
Comment by Zachriel — May 20, 2007 @ 10:36 am
May 20th, 2007 at 10:46 am
Historically, intuition has been a problematic identifier of intentional codes. Planetary orbits, comets, black cats, cold chills, tea-leaves, and even the entrails of chickens were once thought to include coded messages.
I have provided an algorithmic definition of code as the relationship between a source, the key and the result. The transcription can conceivably be done by agency or by a natural process; e.g. the former, a simple substitution cipher; the latter, the genetic code. Of note, the encoding and decoding may use different algorithms; the former, modern digital encryption; the latter, evolution 'encodes', genetic transcription decodes.
Eric's claim, as I understand it, is that transcription must be done by agency, or by a process devised by agency. He reasons via the apparent irreducibility of such a process.
Comment by Zachriel — May 20, 2007 @ 10:46 am
May 20th, 2007 at 11:48 am
Hi Stunney,
You wrote…
Zachriel indicated to you that "This may be your best post yet…"
I agree. While I suspect you will continue to adamantly disagree with us, this goes to show that you might understand what we are saying.
It "… must all be essentially physics". That is why my ID proposal doesn't just focus on biological questions. To me, it makes no sense to try and arbitrarily decide there is a special barrier preventing the origin of life (ala Bradford) or the origin of language (ala Eric) when the origin of light (ala me) is a fundamental barrier that must be crossed before anything else can be understood.
Maxwell's equations combine to make one big "just so" claim. What is the probability that electromagnetic waves could originate through random processes? Electromagnetic waves only come as a result of other electromagnetic waves causing them. Information in, information out. The origin of electomagnetic waves is the big barrier.
Claims of special knowledge of other barriers engenders a "…view that the lines between non-life and life, non-thought and thought, and between codes and non-codes, are essentially arbitrary…".
This is why I continue to ask Eric, Bradford and, now, you. If your definition of "life" isn't "essentially arbitrary" please indicate whether nanobes are life. Can "inorganic life" exist? Robotic life?
Provoking Thought
Comment by Thought Provoker — May 20, 2007 @ 11:48 am
May 20th, 2007 at 12:07 pm
And so shall there always be, forever and ever, amen
Comment by WedgeHead — May 20, 2007 @ 12:07 pm
May 20th, 2007 at 12:42 pm
TP,
It's just too early for one to mention nanobes being life.
John Cisar, while trying to research these buggers himself, stated that the apparent growth and replication of the nanobes is more than likely mineral crystallization. This crystallization process utilizing apatite, calcium and phospate. He might be wrong, but we're not at the stage yet where one can jump to the assumption that it is life.
Comment by Doug — May 20, 2007 @ 12:42 pm
May 20th, 2007 at 1:00 pm
Hi Doug,
You wrote…
It is too early to tell if it is biological and/or organic.
It is not too early to tell if it is "life".
It meets Dawkin's definition of life. Does it meet Bradford's, Eric's, Stunney's or yours?
If not, why not? What piece of evidence is missing?
Is the arbitrary definition of "life" such that the term "inorganic life" is nonsensical? Are Robots prohibited from being called "life" no matter how well they converse and replicate themselves?
It helps focus the debate we are having here. Is it just a semantic argument where one side is defining their way to a conclusion they wish to make.
Provoking Thought
Comment by Thought Provoker — May 20, 2007 @ 1:00 pm
May 20th, 2007 at 1:04 pm
TP,
How does mineral crystallizations meet Darwin's definition of life?
If Cisar is correct with his comment:
then how does any definition of life get met?
Comment by Doug — May 20, 2007 @ 1:04 pm
May 20th, 2007 at 2:12 pm
Zachriel wrote:
Yawn. Boooooooorrrrrrriiing.
I said if. IF. IF. if. if. if. IF. Do you understand English? Or can't you read? Or are you simply a moron?
I already pointed out explicitly that you had missed the word 'if'.
You yourself noted the premise was conditional.
Seriously, you should raffle yourself. "Buy a ticket, get a 'bright'! Hours of fun guaranteed!"
I can't be bothered scrolling up to the various posts where I've already laid out the two arguments I've given in support of the conditional premise. One is analytical, the other is empirical. You've conceded the empirical one. And you haven't done anything to refute the analytical one beyond asserting the opposite of its conclusion.
So, you are talking to no purpose. Your claim that I'm assuming the the truth of my conclusion in my conditional premise is demonstrably false. Just scroll up and read. It doesn't even make sense to say such a thing.
I don't have time for more right now.
Comment by stunney — May 20, 2007 @ 2:12 pm
May 20th, 2007 at 2:22 pm
Stunney, you make some good points in your posts and we all get frustrated at times but do try to refrain from name calling. That applies to all of us.
Comment by Bradford — May 20, 2007 @ 2:22 pm
May 20th, 2007 at 2:24 pm
I addressed that issue.
Comment by Zachriel — May 20, 2007 @ 2:24 pm
May 20th, 2007 at 2:28 pm
TP wrote:
I'm not sure you and Zach are in agreement that it's all just physics.
But to the extent you are, that position is open to refutation by parity of reasoning. We could as easily assert that none of it is physics. It's all just conscious experience.
This is why I've urged you to ponder the excerpts from Dummett I've posted.
Comment by stunney — May 20, 2007 @ 2:28 pm
May 20th, 2007 at 2:52 pm
Hi Doug,
You asked…
Please provide a link to what Cisar is talking about. For all I know he could be talking about a strawman like calcium deposits in blood.
Here is a link to the work of Uwins, Webb and Taylor including a ten-page research paper describing "…living colonies of nano-organisms on Triassic and Jurassic sandstones and other substraits."
Now maybe you have evidence that these three scientists are frauds or incorrectly reported their findings. But from what I could understand (which also could be wrong) these Nanobes form colonies that show signs of diversification (different colonies are different).
Growth and adaptation = life?
Provoking Thought
Comment by Thought Provoker — May 20, 2007 @ 2:52 pm
May 20th, 2007 at 2:59 pm
TP, I do not get caught up in the debate over a definition of life because it does not address the central issue which is tracing a pathway from a lifeless environment to a cell. Precursor cell? OK, but you still need empirical support for the viability of that contruct.
Comment by Bradford — May 20, 2007 @ 2:59 pm
May 20th, 2007 at 5:41 pm
Zachriel wrote:
You're defining 'scientific evidence' idiosyncratically.
First, your definition depends upon a highly disputable and completely undefined notion of what constitutes an 'observation'. On a different notion of 'observation', one could as easily assert there is a vast amount of scientific evidence that conscious agency and experience exist, and no evidence that matter exists, since one might easily say that matter is never 'observed', only experiences are.
That is, one might say (as many atheistic logical positivists did say), that 'matter' is an abstract construct. By contrast, 'thought' or 'experience' or 'sense-data' are experienced directly. It's a 'given'. Matter is not a 'given'.
The problem both with your view and with the view I've just outlined, is that there is no such thing as self-interpreting data. On this crucial point, see my reference to the seminal work of Wilfrid Sellars I reference in this post.
'Planet' is not a fundamental term in physics. Nor is 'map'. If physics is all there is, then the only permissible existential quantifications do not include reference to 'planets' and 'maps'. Nor to 'people' or 'conscious minds' or 'donkeys' or 'railroads' or 'rules' or 'codes'.
That's why I asked you to clarify your extraordinarily tendentious notion that the genetic code is essentially a material process and that this captures all there is to it—i.e. captures its entire activity.
No, you're the one who misunderstanding.
The existence of matter or energy or physical causes or space or force is not 'a hypothesis held tentatively for the purposes of testing specific empirical predictions'.
Much of science is a conceptual framework within which 'empirical testing' and 'observation' and 'inference' and 'symmetry' and "invariance' and 'conservation' and 'falsification' and a great many other physical concepts are given meaning. The literature on this is vast see especially:
# 1.1 Objectivity of observations in science
# 1.2 Theory-dependence of observation
# 1.3 Indeterminacy of theory under empirical testing.)
I refer you to the work of Kuhn, Quine, Lakatos and many others, and what I refer to in this post about Christopher Peacocke's analysis of rationalism.
Then, I invite you to think again about whether 'rational agency' (or 'matter', for that matter) is an empirical hypothesis or even needs to be. (Hint: I'm arguing by reductio ad absurdum against your notion that such concepts are, can be, and need to be viewed as 'empirical hypotheses'; and secondly I'm arguing for the view that this stricture applies equally to fundamental physical concepts as much as it does to fundamental mental concepts.)
And do try to define 'observer'.
Self-consciousness is an action?
How does some material process constitute 'solving' as against 'not solving' or 'failing to solve' or 'not being the correct resolution of the problem'?
What empirical observations are evidence of the existence 'memory' or 'learning'? as against 'repetition and change'?
Does the sun have memory because it repeats some processes with variations? Is the sun a problem-solver with memory?
Because you've said previously we have 'some evidence for the existence of minds, while seeming oblivious that the concept of evidence presupposes the existence of minds who understand the concept. If there were no minds, ever, there would be no such thing as evidence, ever.
For more elaboration of this very important point, see the excerpts from Dummett I've quoted here and here.
As I've said before, we have a vast amount of evidence for humans having irreducibly mental properties (such as a capacity to understand the concept of evidence). But I guess you'll just ignore it because such properties don't lend themselves to your definition of observation. Rather than conclude that observation so defined is too limiting to be a defining criterion of rational inquiry, like a sane person would, you've become so ridiculously slavish to that criterion that it forces you to pretend some uncertainty about whether scientists, logicians, and mathematicians have rational minds.
Comment by stunney — May 20, 2007 @ 5:41 pm
May 20th, 2007 at 6:21 pm
Bradford wrote:
Bradford don't confuse frustration with highly amused, scornful lampooning.
I'm motivated to indulge in the latter because so many 'brights' constantly and savagely lampoon the alleged nuttiness of religious people's views. I'm a great believer in exposing double-standards.
So, if they say goddidit, I say magicmatterdidit. If they say, we might find a fairy, I say we're equally likely to find an unintentionally evolving code or a fairy. If they say, we believe in spirits, I say you don't believe in the existence of physically unobservable things? So you don't believe in rational minds or qualia or consciousness then, and you're calling us the batshit crazies? No, you're the batshit crazies. For example….
[Emphasis added]
Anyway, I take your point, Bradford, but I also note that not being able fully or infallibly to de-code actual intentions and emotions is one of the limitations of this medium, I suppose.
Which is also something of an argument for the irreducibility of mental states to physical or behavioral states, and for the view I just now mentioned in my latest reply to Zachriel about there being no such thing as self-interpreting data. Data ultimately requires minds to interpret and understand it.
Comment by stunney — May 20, 2007 @ 6:21 pm
May 20th, 2007 at 7:48 pm
My definitions are quite conventional and consistent with normal scientific practice.
Observation, an act of recognizing and noting a fact or occurrence often involving measurement with instruments. Scientific observations must be objective; that is, they must be repeatable by independent observers.
Anything. Because when you redefine terms, you can make them say anything you want.
In science, matter is an abstract concept. It has nothing to do with atheism.
A planet is a large body revolving around a star. It is not a fundamental term, but a reasonable scientific conclusion.
There may be more to the genetic code than a material process, but there is no scientific evidence of that.
Sure it is.
We observe aspects of the world and form generalizations. A generalization is a "conceptual framework".
Apparently, you don't know what memory and learning are.
That's what I thought. You don't know what memory and learning are. People, dogs, rats, remember things. It's quite easy to test.
We start with basic induction. Fire. Ouch! Fire. Ouch! Do you believe that it is reasonable to conclude that fire causes pain? And do you believe it is reasonable to teach children to avoid fire? Once we accept the basic process of learning and sharing, then we can discover all sorts of wonders.
One of the things most people discover is that there are others like them who share many of the same passions and ideas. But some people never learn this, or learn it the hard way.
Comment by Zachriel — May 20, 2007 @ 7:48 pm
May 20th, 2007 at 9:07 pm
If you still think Dawkin's definition of life is relevant, I'm not at all sure you yet understand what it is that I have been claiming. You seem to be at least somewhat aware that I am referring to a Language Barrier. Yet you seem to reason as though you imagined I were proposing a general Any Kind of "Life" Barrier.
I claim that unguided processes could not create and implement a genetic code, where sequences of base pairs of DNA symbolically code for amino acids in functional proteins. That is the Language Barrier. Implementing a genetic code requires intelligent agency.
Whether nature could produce something else that doesn't need a genetic code is completely irrelevant to the claim I am making, as is the question of whether you'd like to call that "life" or not. "Is X life?" doesn't matter at all to my position. Pick any answer you like.
I already explained this in response to you long ago in a post here so I am surprised and disappointed that you still don't seem to understand the irrelevance of the other-kinds-of-life question.
[BTW: Regarding "still waiting" for an answer, while I responded to the life-definition issue long ago, I haven't yet noticed a response from you containing a causal foundation that answers the unmet challenge I posed in that same post above regarding your own proposal.]
I have to ask you: Are you listening to first of all understand before you disagree? Or are you just looking for anything to disagree with or even for any perceived fault you can poke at, however incidental, without needing to first understand what someone's position really is?
keiths (who was not defending my position) essentially called you on the same problem regarding this same topic in other posts here and here. Though he and I do not agree, he seems to understand my position quite well, and he gets the same posts that everyone does. Yet even after that, you still don't appear to understand as well as he already did then.
I consider keiths's example in that matter to be one we should follow. It is something we should all aim for.
Comment by eric — May 20, 2007 @ 9:07 pm
May 20th, 2007 at 9:15 pm
This thread may be done, so hold the forks. At any rate, it is getting very lengthy.
If someone can start a new thread, I'd like to get some reaction from the It's All Physics crowd to my thoughts on the evolution of buildings.
Comment by stunney — May 20, 2007 @ 9:15 pm
May 20th, 2007 at 10:37 pm
Hi Eric,
If you still think Dawkin's definition of life isn't relevant. I'm not at all sure you yet understand the point I am trying to make.
Unfortunately, I had to take a leave of absence after you posted your comment. If you took my silence as affirmation, you were mistaken. Here is what you wrote then…
This is not a definition any more than saying the "sun is yellow" defines what "sun" means.
Don't you see you own question begging here? You are intentionally looking for ways to make a distinction for the "kind of life that may require intelligent agency". You are looking for the conclusion you wish to make.
Of course it does. You don't like its conclusions, ergo the proposal hasn't met YOUR challenge. BTW, by Dawkin's definition, stars and rust are not life.
Eric, you have defined yourself into the position that "coded information" doesn't even include watch faces. It might not even include the coded time signals beaming from GPS satellites. You are actively trying to consider only the kind of life that "may" result from an intelligent designer.
You can only get so much mileage out of your use of the word "may" before it becomes obvious that your definitions are directly assuming the conclusion you wish to make.
Provoking Thought
Comment by Thought Provoker — May 20, 2007 @ 10:37 pm
May 20th, 2007 at 11:11 pm
Eric: I am willing to revise my definition. I would accept this minimal requirement:
Living organisms require information encoded in a symbolic language.
TP, whether or not the requirement suffices as a definition it does do something very helpful. It identifies a property of life forms existing on earth. OOL and evolution are theories attempting to explain a historic series of events. To be successful any theory must account for observed properties. DNA is an information storage molecule containing sequences of codons and their component nucleotides. The information units represent amino acids, the proteins of which they are part and punctuation signals (start, stop etc.). Eric's argument notes that the variety of OOL theories do not detail pathways to the information storage and transmission properties we observe. The sun being yellow may not constitute its definition but a comprehensive theory about the sun should explain the yellow color.
Eric: While there are obviously many ways that "life" could be defined, the distinction I am making is central rather than arbitrary, since the key issue is that unguided natural processes cannot create the information-and-language based type of life we observe. Consequently, if there is a kind of life that may require intelligent agency, information-and-language based life provides such a distinction.
Not so. You are becoming noted for the phrase "Let's do science." That entails matching data to theory. For the origin of life via intelligent agency that means identifying properties of life that uniquely correlate to intelligent causes. Information and encoding language conventions are well known effects of intelligent causality. To look for a connection is nothing more than matching empirical observations (information bearing systems and encoding conventions) to theoretical constructs (a theory of intelligent causality). IOW, doing science.
Comment by Bradford — May 20, 2007 @ 11:11 pm
May 21st, 2007 at 12:01 am
Hi Bradford,
Am I being set up for a body slam?
You wrote…
Eric could have simply said "Living Organisms require DNA". Followed closely with DNA is "information encoded in a symbolic language."
I will be curious to see if Eric will agree to this "minor" modification to his position? And if he won't, I would be especially interested into why he won't.
You will have to excuse me since I am just an engineer. But in my line of work anyone can find "data" to support just about any conclusion one wishes to make. Simply matching data to theory is practically meaningless. In engineering, we present proposals and begin looking for data that DOESN'T match our theories of operation.
Sorry, but it sure sounds like you are actively trying to find data to match a conclusion you wish to make.
BINGO, you found what you are looking for. Or did you? To me, "well known effects" is an appeal to common sense. Which is dangerous. Let's do some critical analysis on this, shall we? Let's skip over the human intelligence because it brings in too many other factors. Let's talk about animal "intelligence". Animals use information and encoding language conventions to communicate. And, unless you want to assume your conclusion, the animals are the "causality" of this language. They created it. What is it about animals that would allow them to create language? Could it be their ability to learn or adapt? Hey, that even matches the dictionary definition of the word "intelligence".
So we are looking for learning or adapting processes to explain the presence of language. Let's see. Learning is usually done by trial and error of pseudo-random attempts. hmmm….
Ok, Let's do science!
I have to go to bed now. We can finish this tomorrow.
Provoking Thought
Comment by Thought Provoker — May 21, 2007 @ 12:01 am
May 21st, 2007 at 12:03 am
Zachriel, I refer you to my previous reply to you, and especially invite you to read and ponder carefully the various bits of material I linked to therein.
TP, let's define a property F:
For all x, x is F if and only if x is, embodies, or contains information encoded in a symbolic language, where 'information', 'encoded', 'symbolic', and 'language' are defined in whichever way you prefer, or are left undefined if that's what you prefer.
Two questions:
1) Does there exist any x that is F?
2) For all x, and for all y, does there exist some x and some y such that x is F and y is not F and y causes x to be F?
I would ask that you restrict yourself only to using scientifically justified statements in your answers, if you have any.
I'm not seeking to debate your answers. I just want to know what your answers are.
Comment by stunney — May 21, 2007 @ 12:03 am
May 21st, 2007 at 12:19 am
Hi Stunney,
I need to go now, but I think I will need clarification.
Did you mean to say…
F is a property of X
F property is defined as the embodiment, or storage of information encoded in a symbolic language encoded in a symbolic language.
Your questions are…
Does such an X exist in the known universe?
Is there a Y in the known universe that doesn't have property F?
Is there a Y' in the known universe that can change a Y into an X?
Y' being a subset of Y.
Do I get to define the terms of what property F means?
(I can't answer without common definition of terms)
Regards,
TP
Comment by Thought Provoker — May 21, 2007 @ 12:19 am
May 21st, 2007 at 12:45 am
TP, the simple logical notation I used is a pretty standard one.
Let's define a different property F*:
For all x, x is F* if and only if x is an odd integer between 4 and 26.
With that definition, we can now construct the set of all the items that have the property F*.
There's no special significance to the notation, if that's what's worrying you.
Comment by stunney — May 21, 2007 @ 12:45 am
May 21st, 2007 at 12:48 am
That's the other side of the coin and in this case would involve searching for chemical pathways to the systems I mentioned. The selection dilemna has yet to be resolved for OOL scenarios. Without a viable selection criteria one is left with pure determinism or blind chance- hardly appealing options.
Information and encoding language conventions are well known effects of intelligent causality.
This brings up a good point. Do higher forms of intelligent life actually utilize symbolic coding comparable to that found in cells? I previously mentioned examples of what codons actually represent. But I think it was oversimplified. A series of amino acids may be correlated to a series of codons but the logical linkage can be more like (a) create an acidic surface with the following amino acids (b) place these polar ones over there to enable an interface with another protein (c) disrupt an alpha helix with proline etc. This is sophisticated stuff and none of it explained in terms of chemical necessity until a very complex biological structure like a cell is already in place. If one specified the propertites of just two interacting proteins in detail, explained the nature of codons and individual amino acids and provided 3D pieces with which to construct models of the two proteins and their genes so that they would fit together as needed to function properly, I would argue it would take a highly developed intellect to put the pieces together properly. Yet we are expected to believe a mindless process accomplished the same before information systems came into existence. This takes more faith to believe than I possess. Time for bed.
Comment by Bradford — May 21, 2007 @ 12:48 am
May 21st, 2007 at 7:43 am
Offtopic:
I read your cites. I am quite familiar with the philosophy of science. I am quite aware of limitations in the scientific method and the nature of observation. However, that doesn't mean that we can't make scientific assertions concerning the existence of planets. The philosophical questions you raise surrounding more complex areas of scientific investigation are no different than those that surround dropping stones from the Tower of Pisae. It is not necessary to accept your metaphysics to do science, or understand science, or to accept scientific results.
(Wouldn't you agree that this particular thread makes the assumption of the scientific method? Your points may have some merit and interest, but I've noticed that every thread you participate in gets diverted. It would also help if you would respond to my questions.)
Start with fire. How many times does it take before it is reasonable conclude that fire hurts? Review this list. Let us know where you feel induction no longer works.
If you are stuck on 1), then just keep in mind that there is no absolute assurance in induction. Rather, we have a level of confidence in our knowledge. Modern science allows us to assign a metric to this level of confidence (including in comparison with the null-hypothesis)
Comment by Zachriel — May 21, 2007 @ 7:43 am
May 21st, 2007 at 8:04 am
Ontopic:
We know the beginning (a lifeless primordial Earth), and the end (an Earth teaming with life). We also know many points in between. We know (with reasonable scientific certainty) that life evolved over billions of years descending from primitive cellular organisms. The question involves what happened just before that. We have some ideas, and some evidence of the molecular processes. We have bits and pieces of the thread. Replicating molecules. Plausible emergence of DNA and primitive proteins from RNA synthesis. Membranes. Biochemical networks. Communities of cooperating molecules.
But these bits and pieces are still disconnected. Like pearls on a string, we fill in more and more gaps. But because of the complexity of the story, every new pearl creates two new gaps.
Like Zeno, it seems we are destined to always be half-way there.
Scientists point to a specific gaps, and say that without filling that gap, our string of pearls is wanting. No scientist of note claims to have proposed a complete theory of abiogenesis. The events happened billions of years ago, leaving scant evidence. And so the search for pearls continues.
But the pearls in our possession aren't happenstance pearls. Rather, they are predictions. We predict where a new pearl might be found. When acquired, it gives us some measure of confidence that more pearls can be attained.
Comment by Zachriel — May 21, 2007 @ 8:04 am
May 21st, 2007 at 9:20 am
I'm not part of the All Physics Crowd, but I would still be interested to see you expand on your ideas.
Comment by Zachriel — May 21, 2007 @ 9:20 am
May 21st, 2007 at 10:23 am
Hi Bradford,
I wrote…"BINGO, you found what you are looking for. Or did you? … Let's talk about animal "intelligence". Animals use information and encoding language conventions to communicate."
At which point you interrupted the flow with…
It is easy to make a simple things more complicated. As you just demonstrated. The real challenge is to make complicated things simple. Simple enough for people to understand. Here is the continuation of the interrupted flow…
And, unless you want to assume your conclusion, the animals are the "causality" of this language. They created it. What is it about animals that would allow them to create language? Could it be their ability to learn or adapt? Hey, that even matches the dictionary definition of the word "intelligence".
Joy just yelled at me for suggesting the term "intelligence" gets redefined in Telic Thoughts. Here's your chance to prove she was right. Here is a dictionary definition…
Merriam-Webster's Medical DictionaryMain Entry: in·tel·li·gence
Pronunciation: in-'tel-&-j&n(t)s
Function: noun
1 a : the ability to learn or understand or to deal with new or trying situations b : the ability to apply knowledge to manipulate one's environment or to think abstractly as measured by objective criteria (as tests)
2 : mental acuteness
Now, it seems to me the birds' development of language would be attributable to "the ability to learn". If you disagree, then how do you explain how birds CREATED language. Or are you assuming the language of birds was created for them by some other form of intelligence?
Which brings us back to asking for your definition of "intelligence"
Provoking Thought
Comment by Thought Provoker — May 21, 2007 @ 10:23 am
May 21st, 2007 at 12:47 pm
TP, a definition for intelligence applicable to one who has the ability to tinker with nature would be the part of the definition you supplied that reads:
Comment by Bradford — May 21, 2007 @ 12:47 pm
May 21st, 2007 at 1:05 pm
Hi Bradford,
And this is how birds created language?
Provoking Thought
P.S. After birds, get ready for communicating slugs link
Comment by Thought Provoker — May 21, 2007 @ 1:05 pm
May 21st, 2007 at 2:03 pm
the ability to apply knowledge to manipulate one's environment or to think abstractly as measured by objective criteria
TP, I do not know enough about birds and their language capacities to comment but would say that the capacity to think abstractly is causally linked to a human capacity for communication.
Comment by Bradford — May 21, 2007 @ 2:03 pm
May 21st, 2007 at 3:37 pm
Thought Provoker wrote:
Two weeks ago on a different thread, I wrote:
There are two obvious and fundamental codes: DNA, and human language; and the first is of course deeply implicated in the second, because you need a human brain to speak English, Greek, Urdu, etc. If animals also have quasi-linguistic abilities, that just adds to the data set that one is trying to explain"”positing the existence of a bunch of additional codes merely poses the question in a less anthropocentric way. [Emphasis added]
So I'm not at all sure why you seem to think that pointing to possible examples of non-human forms of coded communication is any kind of point in naturalism's favor. ID already hypothesizes that there is non-human coded communication.
What you need to do is show that the origin of all such coded information is probably unintentional and unintelligent. If this had been scientifically established as being the case, then there would be little to talk about and we could all go home. But we all know that it hasn't been established. Not even remotely.
On another issue, which we have spoken about before, I'm not sure I follow what you mean to say by quoting dictionary definitions of the word 'intelligence'…
Note there are two disjunctives there. An ability to understand isn't the same as an ability to learn. If a very bright 'bright', Richard Dawkins say, begins on June 12, 2007 to understand everything there is and could possibly ever be to understand, then he will cease on that date to have the ability to learn or grasp new truths or facts, but he wouldn't thereby cease to be intelligent.
Aristotle's Prime Mover would be intelligent on this definition, even though it is itself necessarily unchanging.
Comment by stunney — May 21, 2007 @ 3:37 pm
May 21st, 2007 at 6:16 pm
Hi Stunney,
Excuse me while I let Eric and Bradford deal with the Animal Intelligence issue while I focus on your logic problem. Assuming you are still interested.
You wrote…
I wasn't at all worried about the form. I used a different form to confirm I understood what you are saying. You "odd integer between 4 and 26" works because we have agreement on what you mean. We don't on the other.
Let me offer what I think is a good faith attempt at addressing this problem logically. Instead of "X" and "Y", I will use "I" and "U".
"I" = "intentional" and "U" = "unintentional".
Ie = Intentional Encoder
Id = Intentional Decoder
It = Intentional transporter of information
Ue = Unintentional Encoder
Ud = Unintentional Decoder
Ut = Unintentional transporter of information
Examples…
Ie = Remote Control (for garage door), Sundial
Id = Automatic Garage Door Opener
It = Telephone wire
Ue = Jupiter's moons, Rocks casting shadows.
Ut = Air transmitting speech
Ud = ????
In engineering parlance, this is called nodal analysis. You determine the different types of nodes and then connect them to define interactions.
Ie to Id = Coded Information
Ix to Ix = Coded Information
Ie to It to Id = Coded Information
Ie to Ut to Id = Coded Information
This process clearly defines some unknown. Does there exist an Unintentional Decoder? I suggest this is your challenge to which I have no ready answer.
There is another undefined transaction…
Ue to Id = ?????
I think you refer to it as just "information", not "coded information" That is fine with me for now.
I need to run. The next step is to figure out how to turn an Ue into an Ie (moving a rock and calling it a sundial).
Regards,
TP
Comment by Thought Provoker — May 21, 2007 @ 6:16 pm
May 21st, 2007 at 6:39 pm
Thought Provoker wrote:
TP, can we get off this treadmill?
I invited you to define or not define terms as you please. It's not a question of whether we agree. It's a question of whether you, using whatever definitions and/or undefined terms you prefer, have any answers to the questions, interpreted in the way you think they ought to be interpreted, which I posed.
What possible relevance does my agreeing or disagreeing have to what you think or believe, whatever that happens to be?
You said:
Okay.
Sometimes I think I should have been a dentist.
Comment by stunney — May 21, 2007 @ 6:39 pm
May 21st, 2007 at 7:40 pm
Stunney, the haggling over definitions is diversionary. But, just to simplify things intelligence can be viewed as a capacity to effect outcomes that would not result from undirected natural forces alone. There are degrees of intelligence; a relevant consideration when one analyzes the origin of life and the genetic code.
Comment by Bradford — May 21, 2007 @ 7:40 pm
May 21st, 2007 at 7:44 pm
Bradford:
Again you are assuming that intelligence itself cannot result from undirected natural forces alone.
Comment by Raevmo — May 21, 2007 @ 7:44 pm
May 21st, 2007 at 7:48 pm
But, just to simplify things intelligence can be viewed as a capacity to effect outcomes that would not result from undirected natural forces alone.
I do not have to assume that to have a working definition on prebiotic earth.
Comment by Bradford — May 21, 2007 @ 7:48 pm
May 21st, 2007 at 7:51 pm
Are you saying there were intelligent beings on prebiotic earth?
Comment by Raevmo — May 21, 2007 @ 7:51 pm
May 21st, 2007 at 8:07 pm
Geologic evidence indicates that upon formation earth was uninhabitable. What occured subsequently is the point of this blog. My claim is that the origin of a cell is consistent with an intelligent cause or, to be more specific, the origin of genetic information and the encoding convention by which it is expressed. If a genome were to result from an unspecified chain of chemical reactions, not guided by an intelligent hand, the claim to intelligent causality would be falsified. The negation of the foregoing would affirm it.
Comment by Bradford — May 21, 2007 @ 8:07 pm
May 21st, 2007 at 8:17 pm
Bradford:
Yeah, I know that's what you claim. But nobody has ever observed an intelligent being creating a cell from scratch. Since you are an intelligent being, perhaps you'd like to tell me how you would do it.
Comment by Raevmo — May 21, 2007 @ 8:17 pm
May 21st, 2007 at 8:18 pm
Hi Stunney,
You asked…
I think we are.
I wrote "Does there exist an Unintentional Decoder? I suggest this is your challenge to which I have no ready answer."
You responded with…
Good, now if you don't mind my use of "I" and "U" instead of your "X" and "Y" we can continue making progress.
Ie = Intentional Encoder
Id = Intentional Decoder
It = Intentional transporter of information
Ue = Unintentional Encoder
Ud = Unintentional Decoder
Ut = Unintentional transporter of information
A while back we expended a lot of energy to come to the conclusion that a random set of rocks casting shadows didn't provide "coded information" but the moment an intelligent agent intentionally moved a single rock to make a "sundial", the rocks then provided "coded information".
Ue = pile of rocks casting shadows
Ie = pile of rocks intentionally adjusted to make a "sundial"
How did "Ue" turn into "Ie"
One way to look at it is to label the act of moving rocks as an intentional transfer of information. This is consistent with your conservation of "I" ("intelligence" and/or "intent") principle. Only an intentional encoder could create another intentional encoder through coded information.
Here is the thought provoking part. If energy used to intentionally move the rock can be thought of as coded information than random energy that just happens to move rocks would be uncoded information.
Therefore, the moon is constantly decoding gravity "information" from the earth.
The moon is an unintentional decoder. Ergo I found an example of a "Ud".
In fact, I already found a Ud in the rock to sundial conversion, because information flow requires both an encoder and a decoder. The pile of rocks were the unintentional decoder being commanded to move.
I believe this is a logically consistent definition describing information flow that closely matches yours. BTW, I don't think it matches Eric's.
We can continue this if you wish since we have now gotten off the "treadmill" and have a common frame of reference.
Regards,
TP
Comment by Thought Provoker — May 21, 2007 @ 8:18 pm
May 21st, 2007 at 8:24 pm
Is your observation intended to inspire confidence that an unknown, and unspecified chemical process, incidentally accomplishing what the best and brightest among us are not able to intentionally accomplish, is a more plausible explanation for life's origin?
Comment by Bradford — May 21, 2007 @ 8:24 pm
May 21st, 2007 at 8:30 pm
Don't you agree that a chemical process accomplishes the development of a zygote into a multicellular organism? Yes or no? I've never seen the best and brightest among us accomplish such a thing intentionally. Don't underestimate the power of chemistry.
Comment by Raevmo — May 21, 2007 @ 8:30 pm
May 21st, 2007 at 9:00 pm
What you describe is possible only because an information storage system exists capable of expression by virtue of functional, encoded mechanisms. The origin of the information and the means by which it is expressed was the problem I related intelligent causality to. There is no equivalent of an observable series of reactions effecting that origin.
Comment by Bradford — May 21, 2007 @ 9:00 pm
May 21st, 2007 at 9:02 pm
Hi Bradford,
You wrote…
Even with all the information and encoded information created by humans it is a fraction of what is created by the animal kingdom.
This is more than "haggling over definitions". You are basing your presumptions on an observed correlation between coded information and a human trait.
With this loose correlation, you suggest that ALL coded information must come from this human-like trait.
The correlation of "coded information" to "ability to learn" has far more examples of occurrence and has explanatory power. So why do you resist it?
Could it be because you think humans are special?
Provoking Thought
Comment by Thought Provoker — May 21, 2007 @ 9:02 pm
May 21st, 2007 at 9:22 pm
Zachriel wrote:
It would help if you answered my questions, a number of which I posed already.
You're not really engaging with many of the issues I have raised (for instance here), but simply assert with mindnumbing monotony that we can do science, 'science is as science does', and 'hello meat'. I'm not aware of anyone at TT who has denied that we can do science. But as an extremely wise man once said—-Big. Fucking. Deal.
The issue is not the feasibility of science. And if you want to do what you regard as science, here's my advice: go do it.
Nor is the issue whether science generates empirical results. The issue is, how are we to interpret results which are agreed to have been generated by scientific methods. And that's the key issue here at TT, because the results do not have self-authenticating interpretations stamped on their foreheads.
Furthermore, 'planet' is neither a fundamental term of physics nor an absolutely objective term. The same goes for 'intelligence', 'code', 'natural selection', and 'gene'.
Hence it's a tiresome exercize in futility when you try to imply otherwise, or suggest that there is always a crystal clear dividing line between a testable hypothesis and an untestable one; since on one interpretation of data, a given hypothesis may be judged to have been properly tested and given a certain result; while on a different interpretation of the selfsame data, that supposed result has not been found. Experiments and the data they generate always have to be interpreted and understood (by rational minds, no less) in a particular way in order for it even to be the case that a certain experiment confirmed or failed to confirm or falsified a certain hypothesis. And often there is no agreement on the correct interpretation of the resultant data. Your idealized view of science is a far cry from the messy reality of science. You can cover your ears and jump up and down while yelling, "I'm not listening, I'm not listening," all you want. But I refer you again to that extremely wise man I mentioned earlier.
I've already referred you to the disputed interpretation of Libet's experimental data as one very notable example. An even better and more widely known example is that of quantum mechanics. Bohm gives one interpretation. Most physicists probably disagree with it, and there are a number of other proposed interpretations. Einstein would probably side with Bohm against the majority viewpoint, and would even have considered himself to be 'doing science' in so doing.
In terms of the main topic of this blog, the scientific data, including biological data are open to different interpretations.
You will say in your habitually and obviously question-begging way, "there is no evidence for ID". I and others think you are badly mistaken, and that there is in fact a ton of evidence for ID. For instance, the cosmological fine-tuning data and the genetic code itself, properly and most plausibly interpreted strongly point away from chance and impersonal necessity, and toward ID as a valid abductively inferred hypothesis.
You might complain "ID is unfalsifiable". But you are wrong. If we lived in a significantly irrational, irregular and hence a drastically unpredictable world in which there is no rhyme nor reason apparent, ID would be falsified; while, by contrast, given those same data, naturalism wouldn't.
It's completely disingenuous and utterly bogus of ID opponents to prattle on about 'just provide the data' and 'ID is unfalsifiable' for three related reasons: a) the data are there alright; your side just refuses to interpret it in any way whatsoever that's inconsistent with your philosophical prejudices; b) your side refuses to specify the kind of data that you'd ever accept as 'evidence' for ID (one can always interpret the Empire State building and everything else in the world as not necessarily involving any intelligent designing agency); and c) the evolutionary naturalist paradigm is itself unfalsifiable in practice.
You also say things with contradictory implications. For example:
and then
Let me remind of what I wrote in that connection:
Does the physical description of the material processes involved in its construction capture that entire activity? Does the physical description of Shakespeare composing all his literary work capture that entire activity? Does the physical description of a programmer inventing a new programming language capture that entire activity? Does the physical description of a scholar translating a text in Sanskrit into Finnish capture that entire activity? Does the physical description of Godel proving his celebrated incompleteness theorems capture that entire activity? Does the physical description of a NASA engineer solving a differential equation capture that entire activity? Does the physical description of being in pain or of seeing that a tomato is red capture that entire activity? Does the physical description of you making a logical error capture that entire activity?
Also, I don't know many more ways I can tell you what I think about induction—that I agree with Hume's take—that it's simply a strong instinctprovided naturalism is true. But, as I've explained before:
The point is that if one assumes naturalism, induction has no rational foundation, because naturalism posits no ultimate cause that's endowed with a rational and moral will (whereas theism does). It may be an ineradicable habit of thought, as Hume also held. But that it works on the whole does not explain why it works, why the world is predictable that way. Positing an impersonal Nature that is life-consistent and predictable does not explain why Nature is life-consistent and predictable. That's the explanandum. One doesn't explain it by stating it. Rational inquiry proposes explanations. Theism is one such. But naturalism has no plausible way of explaining why impersonal nature is life-consistent and rationally intelligible or capable of generating moral experience and absolute moral prohibitions against burning babies for fun and genocide); or of explaining why the world is enduringly ordered and predictable, except by recourse to the most egregiously unparsimonious hypothesis ever conceived, viz., the multiverse.
I wish to move on to other threads. You can always take up this discussion again in an 'open' or other suitable thread.
Comment by stunney — May 21, 2007 @ 9:22 pm
May 21st, 2007 at 9:42 pm
1. Yes, we do use objective criteria (i.e. tests) to measure animal intelligence by observing how they use knowledge to manipulate their environment.
2. It is not necessarily true that all manner of animal language/signalling is acquired by learning and adaptation. It may also be instinctual, i.e. hard wired in some way.
3. You appear to be misusing dictionary definitions. As I have pointed out before, the reason dictionaries have numbers and letters in front of multiple alternative definitions is that not all definitions apply to all uses of the word. The dictionary does not tell us every use of "intelligence" is going to be "the ability to learn or understand or to deal with new or trying situations".
In particular, I've pointed out to you here why the 1b definition is applicable to the ID inference, whereas the 1a definition is not.
4. If you are instead trying to propose a connection by making a logical argument, such as
Premise: At least some animals may create language through "their ability to learn or adapt"
[Missing Premises]
Conclusion: The intelligence that created the DNA code must have acquired language ability in the same way.
there would clearly be some missing pieces to be filled in for such an argument in order for it to hold up.
5. If you are neither assuming nor attempting to prove that definition 1a must apply to the intelligence that created the genetic codes, then it would seem that whole matter of definition 1a is simply irrelevant to the question of the origin of genetic codes. If you think it has relevance, you have not yet made that relevance evident. keiths, for example, also found your references difficult to connect to the conversation.
If you believe you are making a point that is relevant to the origin of the genetic codes by invoking Dawkin's definition of life or by references to animal intelligence, I'm quite sure that I don't know what point you are trying to make.
OTOH, if you are just making other points that don't relate to the origin of the genetic codes, that is certainly your option. But then I would request that you refrain from implying that I need to respond — as though they did have some relevance to what I've been writing about. If you want to head off on some other tangent you are free to do so. Its each person's choice if they are interested to follow into a different topic.
If there is relevance, you will have to communicate it more clearly.
Comment by eric — May 21, 2007 @ 9:42 pm
May 21st, 2007 at 9:52 pm
Ability to learn is a useful concept at times. Humans have the ability to learn very abstract codes that animals cannot because humans are more intelligent and in that way special. Placing voluminous information in a molecule is consistent with a level of intelligence that would exceed human level.
Comment by Bradford — May 21, 2007 @ 9:52 pm
May 21st, 2007 at 10:02 pm
Thought Provoker wrote:
Huh? Should I read that as "if…., then…."
How does that follow from the "if…., then…. ….would be uncoded information" conditional premise?
To de-code information implies that it's previously in coded form, otherwise de-coding would not even apply. I do not see how the Earth's gravitational information was previously 'coded' prior to the Moon 'de-coding it.
You're not making sense to me. I need clarification as to intentionally directed energy versus random energy. Take the example of WELCOME TO SCOTLAND spelled out on a hillside with boulders.. Those boulders might be right inside the Scottish border and placed there by intentional agents. Or they might be on a hilly part of Mars, having been blown there by Martian winds.
Are you saying both of these are equally clearcut cases of coded information?
Comment by stunney — May 21, 2007 @ 10:02 pm
May 21st, 2007 at 10:41 pm
To Thought Provoker and Zachriel regarding question begging:
Sorry, but you are confused about what "question begging" is.
It is never question begging to declare one's claim, since the claim (i.e. the conclusion for the argument) is the proposition to be proved, not one of the premises. If it were question begging, every scientific hypothesis would be question begging merely by virtue of being proposed.
It is also not question begging to look for evidence what would support a specific position or a broader theory. Evidence is not an assumption, and looking for evidence relevant to a position is not question begging.
Consider Zachriel's depiction above of searching for the pearls that would fill the gaps in the abiogenesis scernarios. If "looking for" discoveries that support the conclusion you wish to make was question begging, every one of those pearls would become question begging. Of course, that is nonsense. It is perfectly legitimate to intentionally look for relevant cases, whether as confirming evidence or as counter examples. That is what well designed investigations are about.
Question begging is about assuming the truth of what you need to prove (in formal logic) or support/establish (in science).
Consequently, if one is asking the question of how the first codon sequences to symbolically code for a protein were created, …
- It would be question begging to assume that functioning and evolving genomes already existed.
- It would be question begging to assume that one already has the ability to resort to inheritance based on coded messages.
- It would be question begging to assume one already has a decoding mechanism to convert symbol sequences into the decoded functional counterparts that support functional selection.
- It would be question begging to assume that in some unspecified way there simply must be a fitness landscape that can discriminate one sequence of items/symbols from another in a way that leads to the future ability to be decoded, even though no decoding mechanism has yet been defined or established.
These are all legitimate examples of question begging because they assume what the question at hand needs to have shown.
Comment by eric — May 21, 2007 @ 10:41 pm
May 21st, 2007 at 10:42 pm
There is no definition of "evolution" in informatics that says or implies that a sequence of undecodable items can be evolved into decodable symbols in the absence of defined decoding mechanisms or supplying predetermined target sequence. As I pointed out, under such conditions "fitness for decoding" would be undefined.
What we have in biology (i.e. once we do have working genomes that use coded messages and evolve) cannot be assumed to apply to the prebiotic creation of coded messages. That is question begging. If you want to make the case that it does apply, an appropriate place to start would be to answer my question:
Now what?
I'm not requiring proof in the formal logical sense. But if you cannot propose even a prima facia plausible scenario forward for Now what?, you certainly are not warranted to merely assume a fitness landscape exists that solves this problem.
Even with biological selection, selection cannot select according to future function or future benefit. There is no fitness landscape that sees "That would become helpful someday."
How do you "select" for suitability as a coded sequence for some counterpart if the decoding mechanisms that would convert a sequence to a functional counterpart are in the future and yet to be created?
Comment by eric — May 21, 2007 @ 10:42 pm
May 21st, 2007 at 10:44 pm
Hi Eric,
You wrote…
I had started writing a reply to you, but I lost it. Maybe it is for the best.
You are obviously not the kind of person I thought you were.
Assume whatever conclusions you wish to assume and as long as you leave me alone, I will leave you alone.
Regards,
TP
Comment by Thought Provoker — May 21, 2007 @ 10:44 pm
May 21st, 2007 at 11:00 pm
I have tried to engage your points. I am trying to find common ground, a basis for further discussion.
How many times sticking your hand in the fire does it take before it is reasonable to conclude that fire burns? Do you consider it reasonable to teach children to avoid burning themselves in fire? (And for Heaven's Sake, don't ask Hume.)
All I see is handwaving. "Interpret" the data all you want. Ponder it. Think about it. Write long essay answers. But when you're done, let us know: Does fire burn?
That is not my position. I said, "A planet is a large body revolving around a star. It is not a fundamental term, but a reasonable scientific conclusion."
That is not my position. Nevertheless, some statements clearly do not qualify as valid scientific hypotheses. (Please think before saying the two statements are contradictory.)
Difficulties in understanding and interpreting data doesn't give scientific license to pretend that we can merely reinterpret data to avoid uncomfortable results or reach a desired conclusion. Interpret all you want. Think all you want. Write long essays. But the existence planets, large masses orbiting stars, is a strongly supported scientific conclusion. Eppur si muove!
That is why we have to have a firm understanding of what it means to make a scientific assertion.
That is precisely contrary to my view. Science is a practical endeavor and often has to deal with seemingly contradictory evidence. Unlike in logic, these contradictions don't give one license to "prove" that 1=2. Science is still bound by the rules of evidence, i.e. specific and distinguishing predictions of empirical phenomena.
When you say things like that, then it's clear that you don't understand the scientific method. But we'll work on that once we can reach some agreement on basic induction.
Comment by Zachriel — May 21, 2007 @ 11:00 pm
May 21st, 2007 at 11:04 pm
Hi Stunney,
I will answer your last question first…
I would assume the rock formation in Scotland was intentional and the one on Mars was unintentional. Either could be an incorrect assumption but unlikely. As to whether or not the Mars message was "coded information", I would say no in deference to our common definition of the term. I don't care what the definition is as long as we both know what it is.
Yes. What can I say, "I R a eenginear"
If energy used to intentionally move the rock can be thought of as coded information THEN random energy that just happens to move rocks would be uncoded information.
If "information" include energy and "decoding" includes motion THEN the moon is decoding information that was encoded by the earth to move in its orbit.
the moon is an unintentional decoder.
the earth is an unintentional encoder.
Now you can reject this whole concept but then we have to go back to discussing how an intentional encoder creates another intentional encoder.
(i.e. moving rocks to make a sundial)
Provoking Thought
Comment by Thought Provoker — May 21, 2007 @ 11:04 pm
May 21st, 2007 at 11:08 pm
Hi Bradford,
You wrote…
Is this your conclusion or your assumption or both?
Provoking Thought
Comment by Thought Provoker — May 21, 2007 @ 11:08 pm
May 21st, 2007 at 11:22 pm
Given that I earlier explicitly affirmed and explained in response to you that "the markings on the faces of either [a sundial or a watch] are indeed codes", I am amazed that you are able to somehow arrive at the exactly opposite interpretation from what I am saying.
I'd like to take your posts and your questions seriously and to respond to them seriously. I would like to believe you want to understand, before you disagree. Would any objective observer be surprised if I am in doubt about whether you are bothering to understand what you confidently disagree with?
Please show a good faith effort to engage my actual position. Please give me a reason to believe that it is worthwhile to try to respond to your questions.
What exactly did you expect a blog about Telic Thoughts to be about?
Of course we are trying to see if there are certain effects (e.g. the genetic code being one) that are best explained by an inference to an intelligent designer.
What have you thought ID was about all this time?
Suppose X doesn't require an Intelligent Designer, but there are reasons to infer that Y does. Between these two, which of these would you reasonably expect to be the focus of attention on an ID blog?
Comment by eric — May 21, 2007 @ 11:22 pm
May 21st, 2007 at 11:22 pm
It is plausible that replication and therefore evolution can occur before the origin of the genetic code. Hence, some sort of fitness landscape would apply (though not necessarily discriminatory with regards to symbols).
Gee whiz, eric. After a long comment on petitio principii, here you are assuming what you are attempting to prove. How do you know it isn't "implied". Evolutionary processes often lead to rather surprising results.
Start by providing a general definition of evolution. You can't even begin a proof without it.
You had assumed evolution, such as in replicating molecules which are known to evolve in response to their local environment and in the competitive acquisition of resources. (But the fitness landscape would presumably be blind to the internal symbiology of the replicators.)
Comment by Zachriel — May 21, 2007 @ 11:22 pm
May 21st, 2007 at 11:50 pm
Zachriel wrote:
If you know that I don't understand the scientific method (not that I was aware of any scientific experiment you performed upon me to gain this knowledge—did you do it while I was asleep?), then I'd strongly advise you to move on with your life. Forget about me. I promise I won't cry.
Of course, I don't think that 'scientific method' means what you insist it means. I think the concept has been and is disputable and, dare I say it, evolving. I also do not think it denotes just one method.
You asked me before if I thought it was unreasonable to conclude that putting one's hand in the fire would hurt. I don't know if you remember my answer. It was, "Of course not".
I then explained to you—and I'm going to do this again for what I promise will be the last time, that:
Given naturalism, reason doesn't come into it. It's an ineradicable, natural, instinctive belief to be sure; but an instinct that, as Hume correctly surmised, could have no rational foundation, since the fact that induction has worked before does not non-circularly rationally justify the belief that induction will ever work again (or, for that matter, justify the belief that our instincts won't change at any moment). Whereas, given theism and in particular its hypothesis that the world has a perfectly rational, perfectly good and perfectly almighty creator, induction is fully justified from a rational point of view, since such a creator would not frustrate his rational (and even non-rational) creatures' nature and agency/behavior by making an irrational, irregular and hence unpredictable world, precisely as has been affirmed by theists for millenia, and by theists predicted to be a continuing, enduringly observed fact about the world.
And now I will let you in on something you may not have realized before now: I am a theist.
Now if you wish to refute Humean and/or theistic arguments about induction, there are many refereed philosophy journals out there that would love to hear from you. Indeed, if you could prove that inductive reasoning is rationally valid using solely the scientific method while avoiding logical circularity, then there are many refereed scientific journals that would be even more delighted to publish your proof.
I wish you the best of luck with your mission, should you choose to accept it. Please destroy your computer as soon as you have read this message.
Comment by stunney — May 21, 2007 @ 11:50 pm
May 22nd, 2007 at 12:28 am
Placing voluminous information in a molecule is consistent with a level of intelligence that would exceed human level.
It is rationally deduced in character with my customary mode of thinking.:wink:
Comment by Bradford — May 22, 2007 @ 12:28 am
May 22nd, 2007 at 3:20 am
Thought Provoker wrote:
What did you call me?:shock:
Comment by stunney — May 22, 2007 @ 3:20 am
May 22nd, 2007 at 4:07 am
stunney wrote:
Translation:
1) If we assume God exists; and
2) if we assume God is good; and
3) if we assume a good God would guarantee the reliability of induction;
4) then we can conclude that induction is reliable.
What was that you were saying about circular logic, stunney?
Comment by keiths — May 22, 2007 @ 4:07 am
May 22nd, 2007 at 4:19 am
ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ
ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ
ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ
ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ
ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ
ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ
ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ
ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ
ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ
ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ
ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ
ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ
ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ
ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ
ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ
ZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZ
Comment by stunney — May 22, 2007 @ 4:19 am
May 22nd, 2007 at 7:49 am
According to your statement, then, a naturalist cannot *reasonably* come to the conclusion that fire hurts. They have to rely on instinct. And yet we know that a naturalist can test the idea by putting their hand near a fire, touching the fire, studying the fire. After a lifetime of boiling water, cooking food, heating the home, they can acquire a great deal of knowledge. They can even study heat engines and replicate the experiments of the great inventors. And yet, while you use reason, they are using merely instinct.
I fail to see your argument as persuasive, nor your distinction as having any validity. Apparently, you think that "reasonable" excludes appeal to experience or sound judgment.
Comment by Zachriel — May 22, 2007 @ 7:49 am
May 22nd, 2007 at 7:55 am
Except of course for those numerous instances when the world *is* irregular and unpredictable. Then the Creater works in mysterious ways and who are we to question His motives.
Comment by Raevmo — May 22, 2007 @ 7:55 am
May 22nd, 2007 at 8:00 am
When someone suggests you are assuming your conclusion that only Theism provides a valid justification for inference from experience, you quote a definition that states "Abduction, or inference to the best explanation, is a method of reasoning in which one chooses the hypothesis which would, if true, best explain the relevant evidence."
Yet, without the assumption of Theism, we cannot infer any consistency to the evidence as it might (conceivably) change at any moment. We then "predict" that it will remain consistent. So, stunney is a Theist as long as he finds the world comprehendible. This is circular reasoning at its very best.
{{Applause!}}
Comment by Zachriel — May 22, 2007 @ 8:00 am
May 22nd, 2007 at 12:32 pm
Zachriel wrote:
Wrong. That does not follow from my statement at all.
What follows is that a naturalist is perfectly capable of reasonably coming to the conclusion that fire hurts, given the truth of theism. The statement does not say the naturalist is only capable of being justified in making inductions if the naturalist believes that theism is true, since that's a logically incoherent notion. Naturalists by definition (in this context) are not believers in theism.
But given theism, induction is (in fact or probably) a reliable belief-formation process. Given naturalism, induction may not in fact be reliable, or if it is, may cease to be at any moment.
You need to have a look at externalism and reliabilism in epistemology.
But even on internalist theories of knowledge, one is, as Hume realized, still faced with the problem of induction being itself unjustifed without circularity, given naturalism; whereas, if we have reasons to accept theism that are logically independent of the issue of justifying induction (and all philosophical theists do think we have such reasons), then we have a warrant for believing induction is reliable that is independent of induction and hence non-circular.
If you want to learn more about abduction, you can start by going here.
But since you are now boring me to tears, I'm afraid this will have to be it for me on this thread.
Hi Raevmo. Good to see you saying the world is unpredictable. I think you have a duty to alert the scientific community to your important finding.
Comment by stunney — May 22, 2007 @ 12:32 pm
May 22nd, 2007 at 1:21 pm
I have and I did. To appear in Nature any day now.
Comment by Raevmo — May 22, 2007 @ 1:21 pm
May 22nd, 2007 at 2:16 pm
One postscript.
In Plantinga's list of TWO DOZEN (OR SO) THEISTIC ARGUMENTS which I linked to in my preceding comment (at 'such reasons'), the one from induction is at letter M.
I attended the lectures based on these notes when Plantinga delivered them at Oxford about 20 years ago, and from memory people laughed a lot at the facetious bit about, "This may have been known even before Hume."
I can't remember how many lectures there were in total, but I do remember that he did cover 26 arguments altogether. The one that had people the most interested in was letter P, "The Kripke-Wittgenstein Argument From Plus and Quus" (for which we got a special handout which sadly I've lost), since this was not long after the electrifying appearance of Kripke's superb book, now recognized as a masterpiece of analytic philosophy, Wittgenstein on Rules and Private Language.
Comment by stunney — May 22, 2007 @ 2:16 pm
May 22nd, 2007 at 5:24 pm
Inductive conclusions are always tentative and subject being revised or discarded; hence, the second option more accurately reflects the nature of induction.
Furthermore, many theists do not believe that induction is a "reliable belief-formation process". The End is Nigh, and what we learn today may be of no value in the world hereafter.
Rather we have a measure of confidence in induction which, while demonstrably better than the null-hypothesis, is necessarily limited by our experience. However, with language, writing, libraries, printing, and modern advances, human experience encompasses the memories of the entire culture.
The first link states that Abductive Reasoning is a form of inference based on evidence and is used to generate an empirically testable hypothesis. The other links have similar, consistent definitions.
So again. You are using inference to prove the validity of the process of reaching an inference.
The world is full of unpredictability. There are many chaotic phenomena, e.g. it is not even known if the Earth is in a stable orbit, or not. Other processes are so contingent that their effects can't be predicted, e.g. a single cosmic ray may cause a cancer and the death of a community of trillions of cells we call a person. And it appears that many natural systems have quantum randomness at their core, e.g. the next uranium atom to disintegrate.
Comment by Zachriel — May 22, 2007 @ 5:24 pm
May 22nd, 2007 at 5:46 pm
Zachriel:
Thanks for clarifying my point. It's easy and trivial to say that theism predicts predictability (haha), but if theism were truly useful as an explanatory tool it should be able to explain why certain things are predictable and others aren't. Perhaps stunney would like to stick out his neck and predict something that we don't already know.
Comment by Raevmo — May 22, 2007 @ 5:46 pm
May 22nd, 2007 at 7:04 pm
Yeah, no doubt about it. Science is predictive. It predicts orbital paths and cellular functions. Here is another prediction. Cellular enzymatic functions will be used as pseudo- predictors for life's origins.
Comment by Bradford — May 22, 2007 @ 7:04 pm
May 22nd, 2007 at 8:46 pm
Yes, that is a very reasonable statement and I don't see any obvious reason to disagree with it, although some clarifying qualifications come to mind:
1. The case has not been made yet that replicating molecules can be constructed by some means yet to be discovered such that they reach a sufficient density to achieve replication. (The copying molecule needs a physically accessible template to copy from.) That is not a given, but it is also a separate matter from the Language Barrier so I'm willing to just suppose it for our present purposes.
2. Prebiotic replication is chemically more limited than reproduction based on the genetic code and decoding+construction mechanisms. One doesn't have the luxury of building new components from copied instructions. One can only replicate what one can chemically copy by using a physical original as a template. That method may not scale up sufficiently for the more complex mechanisms needed for code processing. Consequently, the method of replication may constrain what can be produced. Nor does selection remove this limitation, since selection can only select what can be produced.
3. In a prebiotic world, "selection" effectively means that some structures may replicate more prolifically under existing conditions than others, while other variations may be extinguished. As with biological selection, selection cannot consider benefits that are in the future. It works only on present differences.
The most significant part of your statement is this:
"Hence, some sort of fitness landscape would apply (though not necessarily discriminatory with regards to symbols)."
I agree. It could select for prolific replication, but not for future use as symbols. Without encoding and decoding, the idea of a "symbol" is meaningless at that point in history. Selection only works on present effective differences.
That is a key part of why I've said that there would not have been a fitness landscape for evolving items into symbols encoding messages. It would conflict with what we know about selection.
On the contrary, I'm not just assuming. I know with confidence its not there (even as "implied") for the very reasons I just mentioned. It would contradict what we know about evolution and selection.
In the absence of decoding mechanisms, "symbol" is meaningless and selection (however surprising) simply cannot pursue future realities. It would necessarily be inherently blind to symbolic considerations. Change would be directionless with regard to such issues.
Even if you accidentally constructed some sequence of items that could possibly be decoded into a protein by some coding convention, the sequence would still be meaningless because symbolic coding conventions are not unique. By other coding conventions, it wouldn't decode to a protein, and in any case, this hypothetical question is invisible to mindless processing in the absence of a decoding mechanism.
You could put bits on a disk, but it doesn't mean anything at all until you also have a synchronized and compatible associated convention for interpreting bits into music, movies, software, photos, literature, etc. Without that, its just different arrangements of noise.
That is why there is no answer for "Now what?" It is a dead end.
Comment by eric — May 22, 2007 @ 8:46 pm
May 22nd, 2007 at 10:13 pm
That is correct.
Using words like "confidence" and "what we know", you phrased your current statement as an empirical assertion"”not as a logically derived conclusion. That might explain why you haven't offered a rigorous proof.
You are merely pointing to a Gap in knowledge, can't imagine how it could happen, and believe intuitively that it couldn't. You reason based on something akin to irreducible complexity. But irreducible structures can and do evolve. Your intuition based on your everyday experience misleads you, and your confidence is not evidence. The existence of such a 'Barrier' is an empirical question, but you haven't provided any empirical predictions based on the hypothesis.
Comment by Zachriel — May 22, 2007 @ 10:13 pm
May 22nd, 2007 at 10:28 pm
Imagining that something happens is poor science. When evidence for something does not exist that's where you leave it. Asserting a "gap" is placing one's faith in the expectation of an outcome favorable to theory. It is Ken Miller's empirical argument in a cheap tuxedo.
Arguments for some IC systems are built on circumstantial evidence. For others the supporting evidence is weak or non-existent.
Comment by Bradford — May 22, 2007 @ 10:28 pm
May 23rd, 2007 at 4:46 am
Those on this thread may be interested to know that the June issue of Scientific American contains an article by Robert Shapiro on the "metabolism first" approaches to OOL.
Comment by keiths — May 23, 2007 @ 4:46 am
May 23rd, 2007 at 6:58 am
Hi Keiths,
I had already thought of using that for another post.
Comment by Bradford — May 23, 2007 @ 6:58 am
May 23rd, 2007 at 7:34 am
There is no complete theory of abiogenesis. It is eric who claims to have discovered something, a 'Barrier', that he can provide no evidence for other than his incredulity.
We know that at one time there was no life on Earth. Then there was, and it evolved into the myriad of organisms extant today. The Gap is historical and concerns the origin of life. There is no scientific evidence of agency in that origin, but there is evidence that all organic processes are consistent with physical and chemical laws.
By they way, there is nothing wrong with imagining, as long as you don't confuse the imagination with what can be scientifically demonstrated. Many valid hypotheses have originated in the imagination, e.g. Einstein pretending to ride a beam of light, racing with a clock.
There is no 'Barrier' to the evolution of irreducible structures. You will notice that modern (ID) incarnations of arguments concerning irreducible structures will be found in Gaps of modern scientific understanding, e.g. origin of cellular structures, origin of life, etc.
Comment by Zachriel — May 23, 2007 @ 7:34 am
May 23rd, 2007 at 9:47 pm
If you asserting that the origin of irreducibly complex systems is explained by unguided evolution, that would be more question begging. The belief that their origin can be accounted for in that manner is yet to be vindicated. If you mean merely that they can change once they exist, that is neither controversial nor particularly relevant.
You are right that the issue with the Language Barrier is only "something akin" to IR systems in that they both derive from a common limitation of evolution, namely
Simply put, an irreducibly complex system provides a function that ceases if any part is removed. So explaining the preparation of well fitted parts for a future function is problematic for evolution.
In the case of the Language Barrier, the problem of this limitation is compounded. Rather than needing to create single mechanism providing some function, a coded message processing system needs multiple coordinated and synchronized mechanisms in order to function, i.e. mechanisms
- for encoding (else you have no coded messages at all),
- for storage/transmission (else you lose the messages),
- for retrieval (else you cannot find or get the message you need), and
- for decoding (else the sequences remain useless without the associated function).
The value of these multiple mechanisms is in their future function as an information processing system. Evolution can only select for present effective differences, e.g. that might increase the quantity of replication, not for future utility as symbols or systems to create and read symbolic messages.
In the prebiotic present, such things would not have meaning, let alone function. Yet evolution would need to make all these mechanisms in order to have a functioning system of coded messages. The onus of burden of proof is on the vague claim that evolution can manage somehow anyway, despite its acknowledged limitation to present differences.
My confidence is not evidence. It comes from the evidence.
You seem to be confident that my "intuition based on [my] everyday experience misleads [me]". However, you have yet to show where my description regarding the limitations of evolution to present differences was incorrectly applied to the case of symbols.
For example, I've pointed out the fact (not just an intuition) that a sequence of items does not mean anything at all symbolically, if there is no associated coding convention to translate it into a realized form. Hence, there is no symbolic meaning for selection to select. Those are hard facts, not gaps in knowledge.
Your own intuitions indicate that I must be mislead about this, but your analysis has failed to locate any error in those statements. You are confident about abiogenesis and that I am misled, but your confidence is not evidence.
I know and can state why evolution cannot select for future function as symbols. You can neither state why that is incorrect, nor provide an alternative scenario that has even minimal prima facia plausibilty. (And yes, you have never claimed that you could do this since there is no complete story of abiogensis. Yet, acknowledging this truth does not remove the fact that no plausible alternative scenario has been found.)
The case of future symbolic meaning is dissimilar from the many other cases where selection has realized present function to select. How do you know that it is not your own intuitions concerning the supposedly unlimitted capabilities of evolution that are doing the misleading?
Comment by eric — May 23, 2007 @ 9:47 pm
May 24th, 2007 at 7:33 am
In general, evolution is more than capable of creating irreducible structures. If not, I would like to see the proof.
Your reasoning is based on the concept of Irreducible Complexity and your claim that evolutionary processes are not capable of creating such structures. Further, that this is a truism subject to logical proof. Start with a valid definition of what you think comprises an evolutionary process, and a clear definition of an irreducible structure.
Comment by Zachriel — May 24, 2007 @ 7:33 am
May 24th, 2007 at 11:08 pm
Let us label proposition X as "evolution is more than capable of creating irreducibly complex systems" (or alternately the idea that evolution can create and implement the first genetic code). With that in mind, let's consider the logical fallacy of negative proof.
I would invite people to examine your apparently preferred allocation for the burden of proof.
Throughout this discussion, you have repeatedly expressed your expectation that skeptics of the sufficiency of evolution need to prove a negative, e.g. that there is no possible way that natural processes could succeed in implementing a genetic code or an irreducibly complex system. If they cannot prove this to your satisfaction, all that we have is a "Gap" of knowledge, nothing more.
On the other hand, at no point have you accepted an equal burden of proof for the affirmative position that evolution can succeed in crossing the Language Barrier or creating irreducibly complex systems. Normally, the primary burden of proof would be on those who affirm something to be so, not upon the negative. However, you do not appear to believe that it is even essential to provide a prima facia plausible scenario by which creating something like a genetic code could be achieved.
In spite of this glaring omission, the audience is presumably expected to have faith as we work to fill in the gaps with promised future pearls of new discoveries we cannot even persuasively describe yet. Evolution need not even provide an explanation in order to be assumed true, regardless of clear indications to the contrary.
A key problem here is that healthy science — science that does not rest on a prior dogmatic commitment to an ideology — does not rest on appeals to believe and have faith. Healthy science makes the best inferences available from the applicable evidence we have, revising these as required by new discoveries.
Despite this double standard, I have stated clearly why it is unreasonable to believe that evolution can evolve item sequences into symbolic codes, and I invited you to show me the errors in my analysis. If you had found any, I am confident you would have pointed them out, but let's double check anyway.
1. Evolution is limited to only selecting according to differences that have effects in the present. It may not consider future benefits. You agreed.
2. In the absence of any associated coding convention, a sequence of items has no symbolic meaning. Symbolic meaning does not exist prior to establishing a coding association between the sequence and its realized/decoded counterpart. Uncontested.
3. In light of 1 and 2, before the appearance of coding conventions, selection cannot prefer item sequences that would someday become symbolic sequences. Any "fitness landscape" would be blind to such considerations. You conceded a fitness landscape was "not necessarily discriminatory with regards to symbols" and have not provided any basis to believe it could.
4. I've pointed out that information processing function of the mechanisms required to establish a symbolic coding convention would also be in the future, after their synchronized construction. (For example, you can't decode symbolic messages if there does not yet exist any symbolic messages. However, since symbolic messages cannot be selected until the associated coding convention exists, they could not be created prior to establishing encoding mechanisms.) The fact that symbolic message processing function would be a future benefit (not a present function existing prior to having symbolic messages) remains unassailed.
5. Again applying the limit of evolution to selecting for present function, the fitness landscape has no ability to prefer the future benefit of symbolic message processing in advance of the existence of symbolic messages. You've not shown any error here either, nor have you proposed any clear plausible alternative basis for selection that would lead to symbolic message processing. All plausible chemical reactions are strictly limited to non-symbolic interactions, with no evident basis for getting off the non-symbolic merry-go-round.
6. I've pointed out that Before Language (B.L.) replication is constrained to physically copying according an accessible template. Even if there were no limitations on selection whatsoever, this creates severe limitations concerning the ability of replication to construct the complex interacting mechanisms sufficient to provide the coding mechanisms necessary to establish an associated coding convention. It would not be enough to be able to replicate an individual RNA molecule over and over. The machinery of an associated coding convention is required. Yet, no prima facia case appears available to support the belief that B.L. replication can construct functional systems of molecules.
There is more that weighs against an evolutionary solution, but this is sufficient for now. What it comes down to is just this.
You acknowledge that evolution is limited to selecting for present functional differences, not considering future benefits. Yet, you appear unwilling to allow that this could mean that evolution is limited in what it could accomplish. Although limited, evolution is not limited. You do not see how evolution could still accomplish the creation of a symbolic genetic code. Yet, even though you do not see it, still you believe that evolution can do this, and no arguments based on the acknowledged limitation of evolution and its blindness to symbols and symbolic processing function overcome this belief.
This is not a description of empirical science reaching for the best supported inference from the applicable evidence. This is a description of having faith — faith that holds on despite any appearances to the contrary.
Under those conditions, the burden of proof is not being objectively and equally applied to the evolution of symbolic codes. Analysis of evolution is not functioning as a rational extrapolation according to its established properties and limitations. Instead, evolution has become a belief system that cannot fail, whatever the need.
Comment by eric — May 24, 2007 @ 11:08 pm
May 25th, 2007 at 8:08 am
I withdraw my claim for now (though it is non-controversial in the field of evolutionary computation). That still leaves you with the burden of proving your own assertion.
That is *exactly* what you claimed. I just expect you to support your own claim"”or tentatively withdraw it.
I have repeatedly stated that there is no complete theory of abiogenesis, and that no one knows how the genetic code evolved. Nor have you provided logical proof or empirical evidence of your posited 'Barrier'.
Why do you keep repeating that? I have repeatedly stated that there is no complete theory of abiogenesis or of the evolution of the genetic code. Maybe a miracle occurred. But there is no evidence of such a miracle, nor is there evidence of a 'Barrier'.
1. Okay.
2. Please define symbol in a manner consistent with the chemical processes of transcription.
3. Okay.
4. Without a valid definition of symbol, it is hard to evaluate this statement.
5. Okay.
6. Yet, no prima facia case appears available to support the belief that B.L. replication can construct functional systems of molecules. Well, I'm not sure how many times I need to repeat that there is no complete theory of abiogenesis or the evolution of the genetic code. Basically, all you have done is restate that which is to be proved, then shift the burden of proof.
Evolutionary processes are very limited in what they can accomplish. E.g., they rarely sample more than a tiny fraction of the available search space. The vast majority of possible structures are never considered.
You seem to be trying to argue that Irreducible Structures cannot evolve. You may want to try the general case to see if that leads to a more formal understanding.
You haven't provided anything close to a formal proof, and seem to be assuming your conclusion. You also may be conflating evolutionary algorithms with biological evolution. Your claim is general and applies to all evolutionary algorithms, and is based on the notion of Irreducibility. Please be skeptical of your own ideas.
If you want to say that there is no valid theory of the evolution of the genetic code, and that you are skeptical that one can be devised, that would be much more tenable position. Then the argument would concern plausible mechanisms. But you have posited a 'Barrier' that seems to be based in the logic of symbiology and the nature of evolutionary processes, yet you can't devise the formal proof necessary to prove your assertion. Hence, I am very skeptical of that particular claim.
No one is claiming to know how the genetic code evolved. Perhaps it was Divine Intervention.
You made a claim and continue to cling to it. That is your privilege. But you have failed to demonstrate that your belief follows from the logic of symbiology and the nature of evolutionary processes.
Comment by Zachriel — May 25, 2007 @ 8:08 am
May 26th, 2007 at 3:49 pm
Below I provide an easy experiment anyone can try to help understand the distinction between items, symbols, and encodings.
To preface this, Zachriel asked for a definition of symbol.
The key phrase is "represents something else". Every item or object is itself, whether it be a sun, a rock, or an RNA molecule. In order to be a symbol, the object needs to represent something else.
In DNA, sequences of codons (triplets of bases) can represent sequences of amino acids in a functional protein. A genetic code is the convention and the molecular machinery by which the symbols (e.g. codons) are decoded into what they represent (e.g. proteins).
A decoding process also happens when a web page holding text information is constructed. Most people probably suppose that the characters you see are being sent to your web browser as you see them.
In reality, digital symbols are sent and these must be decoded into what you see. The digital symbols are in the form of binary digits (bits, i.e. on vs. off, one vs. zero). This is like the bases (A, C, G, T) that appear in DNA.
A group of eight bits is a "byte", which is similar to taking a group of three bases together as one codon. When bytes represent text (instead of images, music, or something else) the sequence of bytes are decoded according to a "character encoding" in order to know what sequence of characters to display. Likewise, a codon sequence is decoded according to a genetic code to determine the amino acid sequence in a protein.
To see that the character encoding is something distinct from the symbols it decodes, try this little experiment.
Simple illustrating experiment: At least on some web pages, it is possible to control and override the character encoding that is used to decode the digital symbols that your web browser has received. For example, in a separate window or tab, you could bring up this page.
Then check the menus of your browser for something like "Character Encoding" or "Encoding" or "Text Encoding". One likely place is somewhere under the View menu.
Note which character encoding is selected (so you can also switch it back again). Then change the character encoding to something else, e.g. a Chinese encoding, or some other unrelated encoding. When you switch to an unrelated encoding, the text changes.
In fact, the text becomes gibberish. It is not translated into Chinese (or whatever you chose). Even if you knew Chinese, it would not read as Chinese.
The digital symbols that were sent to the web browser have not changed. They are simply being decoded in a way that is inconsistent with the code that was used to construct the symbols in the first place.
BTW, that is why I point out that the mechanisms of a coded message processing system must be synchronized and coordinated. If decoding doesn't match the encoding, then the message does not come through. It is not decoded correctly.
[Optional extra detail that can be skipped: I realize that for the web page there is also the matter that the associated fonts need to be matched to the symbols and the encoding. And there is a second layer of decoding that takes place when a literate reader interprets the correct, non-gibberish characters according to language. But the initial decoding from a digital sequence into a character sequence is sufficient to illustrate genetic decoding. End Extra Detail]
Suppose we have a character encoding, but instead of giving it a sequence of digital symbols we feed it a stream of digital random noise. The predictable result is that you would have nonsense gibberish. Garbage In, Garbage Out. Even if isolated short sequences corresponded to short words according to an encoding and its associated language interpretation, those words separated by long stretches of other nonsense gibberish would not be meaningful as a message. Even ignoring the gibberish, it wouldn't make any sense.
Flashback: When you were growing up, do you remember seeing cursive writing and vaguely understanding that it meant something? And do you remember ever trying to scribble some loops and squiggles on a page and then ask your parents "What does this say?" Most likely, it meant nothing according to standard conventions for cursive writing, in spite of the fact that you at least had the idea that it could mean something. End Flashback.
Now consider what happens to the random noise example if you do not even have a character encoding with which to convert the digital symbols into a sequence of characters. You just have a random stream of bits. They are not symbols because they do not represent anything. It is just a stream of stuff being itself.
That stream could certainly be changed, over and over. But that just gives you other streams of stuff being itself.
Is there an evolutionary process that could break out of this merry-go-round to nowhere and mutate the streams of stuff into a symbolic message? No. None is possible in the absence of the encoding convention.
Each stream of stuff is equivalently meaningless. If we were playing an evolutionary game of "Hotter, Colder" there would be no stream that is either hotter or colder (i.e. closer or farther to a target) because there is no answer to the question "Closer to what?" The "fitness landscape" for meaning is completely flat, without any X to mark the spot of success.
To express it another way, to have meaning it is never enough to just have a sequence on its own. The same sequence of digital bits might be part of a text message or a photo or a song or a movie or some software instructions. It might mean any of those things or none of those things.
Meaning comes from the fact that the stream of symbols was created by encoding meaningful realized content according to a coding convention such that it can be decoded according to the same coding convention to restore the original associated meaning. Meaning comes from being able to make the trip back to a realized "something else". Meaning is associated with a sequence via the coding convention.
Consequently, without a defined coding convention, it is meaningless to ask "Is this sequence of stuff more like a symbolic sequence?" There is nothing for any selection process to grab onto with regard to changing the sequence into symbols because no mere sequence of itself is symbolic.
Note: This is not a gap in our knowledge that might change someday. It is a fact about symbols and codes.
The fact that DNA sequences can be decoded into proteins shows that DNA contains true symbolic messages, and that gentic codes are true codes. The codon sequences can have meaning because there is an associated coding convention that can map from those sequences to functional proteins. The coding convention defines the association with the "something else" that they represent, which makes them symbols and not merely objects.
The question then is whether or not an unguided natural process could possibly create and implement such a system for symbolic message processing. Could it create the encoding, storage/transmission, retrieval, and decoding machinery required to attach associated meanings to sequences of objects, thereby making them to be symbols? And could it construct such a working system using only Before Language replication?
If so, it would be the very first observed exception to the otherwise universally consistent observation that codes and symbolic language require the foresight, imagination, and ability to pursue future benefit that are found in intelligent design.
[Sorry for the length, but I hope this adequately clarifies the meaning and application of "symbol" and the inherent futility of trying to start with a random sequence with no coding convention, and then evolve it into symbolic messages with meaning.]
Comment by eric — May 26, 2007 @ 3:49 pm
May 26th, 2007 at 7:26 pm
It's a start. Thank you.
Consider a literary symbol. An apple may be just an apple, or it may form an allusion to Genesis and symbolize knowledge or forbidden fruit. But "represents" is somewhat circular in definition. What do we really mean by "represents"
The key to understanding symbols is that there is a process of transcription. The apple transcribes to knowledge. The genetic code transcribes to proteins. In the former case, the transcriber is the reader of literature. In the latter, the transcriber is a chemical mechanism.
But then you go and introduce other terms, "message" and "meaning". Your entire discussion revolves around the assumption that the genetic code is teleological. You should try to strike a skeptical pose and see if you can simply describe the phenomena without such preconceptions.
As to your argument, it is the same as all arguments concerning Irreducible Structures and lacks formal merit. It is, "I don't see how the code could arise before the decoder", etc. I do understand your argument, but it remains a fallacy.
You are wrong on this point. Though the fitness landscape may not select for symbiology directly, that does not mean there is no fitness landscape or that the landscape is flat. Rather evolution selects for those replicators that can best utilize limited local resources and leave progeny. If symbiology could evolve stepwise, then it can be selected for. Each of the intermediate steps would have to be advantageous (leaving drift aside). Those intermediate steps are what you claim could not possibly occur. There is no reason to introduce this other faulty argument about a flat landscape.
Comment by Zachriel — May 26, 2007 @ 7:26 pm
May 26th, 2007 at 9:25 pm
The apple is not intrinsically symbolic of anything unless so enabled by an imposed convention. It is not clear exactly what you mean when stating that an apple transcribes to knowledge. OTOH, codons have exact representations and sequences of them can convey precisely defined chemical properties. If there is meaning in this it is gleaned at an even higher level than the transcribed protein. It is linked to overall function which in turn is linked to an organism's survival. A chemical mechanism in no way implies a chemical explanation for the source of the genetic code. The function of my computer can be explained without reference to the intelligence involved in creating the protocols according to which it operates.
Comment by Bradford — May 26, 2007 @ 9:25 pm
May 26th, 2007 at 10:38 pm
As I said, an apple may be just an apple, or it may form an allusion to Genesis and symbolize knowledge or forbidden fruit.
Comment by Zachriel — May 26, 2007 @ 10:38 pm
May 27th, 2007 at 11:50 am
At a process point that commences with self-replicating molecules (SRMs) evaluation of selection outcomes needs to encompass all competitors. That would include all molecules of the landscape and all reactions involving them. The selection criteria determining which self-replicating RNA would survive is preceeded by the question of whether any would survive competition with non-RNA substances. If the assumption is that they would the next question is what is the basis for thinking that RNAs, with properties enabling them to leave the most progeny, would also be properties that are needed to trace a trail to a cell, as opposed to change in the direction of dead-end nucleic acid variations?
It is unnecessary to show that unspecified steps could be selected for. Unlike adaptive changes in existing organisms, which have mappings between function and nucleotide sequences and causes that are pinpointed to specific changes, the evolution of molecular symbols is a concept in need of biochemical specifics. Until those specifics are forthcoming there is nothing to even falsify. This brings us back to a quote of Dembski introduced by Mike Gene in another post:
My answer, with respect to the origin of symbolic encoding systems, is an unqualified yes. Such systems do signal design. One can point to parallels with intelligent causes identified for like symbolic systems as the model.
Comment by Bradford — May 27, 2007 @ 11:50 am
May 27th, 2007 at 1:55 pm
That and your other points correctly point to significant gaps in the understanding of the origin and infancy of life.
Eric claims that there is no conceivable selectable stepwise pathway.
Eric claims there is a 'Barrier'. Without those specifics forthcoming, there is nothing to falsify.
One can point to the stars and say it is evidence of God's Benevolence; but forming a valid scientific hypothesis and discovering evidence to support the hypothesis is something else entirely.
Comment by Zachriel — May 27, 2007 @ 1:55 pm
May 27th, 2007 at 2:45 pm
Bradford: At a process point that commences with self-replicating molecules (SRMs) evaluation of selection outcomes needs to encompass all competitors.
Gaps that can be filled with an understanding that a symbolic encoding system is a designed end product of an intelligent cause.
Bradford: My answer, with respect to the origin of symbolic encoding systems, is an unqualified yes. Such systems do signal design. One can point to parallels with intelligent causes identified for like symbolic systems as the model.
Unlike much of what we observe, where events are explained by linking together causes reducible to laws of chemistry or physics, understood symbolic systems are linked to causes reducible to an intelligence that created the convention according to which they function.
Comment by Bradford — May 27, 2007 @ 2:45 pm
May 27th, 2007 at 8:57 pm
This is incorrect. There is no ambiguity about the fact that the "messages" in question are the codon sequences and the associated "meaning" in question is the resulting amino acid sequence and functional protein. Neither of these assume teleology. A nonteleological explanation would still involve creating symbolic messages (codon sequences) with associated meaning (proteins).
(For comparison, it is not uncommon for discussions of the powers of Natural Selection to speak in anthropomorphic terms about what it designs or creates, etc. But everyone understands the speaker is not requiring teleology by using such phrasing. As in that case, context removes ambiguity.)
There is a perfectly good reason to cover and exclude the possibility that rearranging item sequence can ever move closer to being symbolic. By ruling out one class of non-options, it clarifies what the necessary requirements are.
You've called my argument wrong and faulty, and you have claimed to understand my argument. Yet you misrepresent my argument in your refutation, and refute something that is not mine.
I have never claimed that there was no fitness landscape of any kind at all for any purpose. You are welcome to "evolve" RNA into other versions of RNA all you like, with no complaint from me.
What I have pointed out correctly is that mere sequence, in and of itself, never contains meaning, i.e. never contains in itself an association with something else. Consequently, no rearrangment of the sequence for some items can ever, in and of itself, move any closer at all to "This is a symbol sequence". In the absence of retrieval and decoding mechanisms, there is no selection or fitness landscape for progress by changing arrangements of items.
Notice that this is not an argument from incredulity. The impossibility of taking the path I described is inherent.
The reason, as I stated, is that the association between symbol and meaning exists in the mechanisms that use a consistent convention to convert between the symbolic form and realized counterpart. Unless and until such conversion mechanisms are constructed, all rearrangements of items are utterly ineffectual.
This is because mere arrangement, in itself, is never symbolic — not for sequences of bases, not for apples, not for electronic bits, not for any sequence of items.
In fact, every sequence of N distinct items A1, A2, … An is simultaneously potentially symbolic of any other sequence of N distinct items B1, B2, … Bn by the corresponding simple association of A1 -> B1, A2 -> B2, … An -> Bn. The immediate corollary to this is that any rearrangement of the original sequence is still necessarily unchanged in its potential for symbolic association.
In short, rearrangement of such an item sequence can never move the sequence closer to a specific realized symbolism. Something else is required to get off of the replication merry-go-round of non-symbolic changes.
As I indicated before, that "something else" is the construction of the coordinated encoding, storage/transmission, retrieval, and decoding machinery that is required for creating a symbolic association.
I trust that everyone has understood all along that any evolutionary solution would proceed in a stepwise fashion such as you describe. That much should be obvious and understood.
I have not stated that prebiotic change could not take place or that there was no fitness landscape regarding nonsymbolic considerations. So, if you want to count any of those changes as "intermediate steps" then it is incorect to say that I claim they are not possible. That is why I freely granted supposing that replicating molecules are present and changing.
What I do claim is impossible is the transition from non-symbolic, literal interactions to symbolic processing that converts between symbolic coded sequences and associated meanings (specifically codons to proteins).
Since this association and conversion ability is accomplished through molecular mechanisms for encoding, storage/transmission, retrieval, and decoding, your suggested framework translates into suggesting that these molecular mechanisms could be constructed in a stepwise fashion by selecting for some other function or functions.
That is easy to suggest. Making it seem plausible or even reasonable, given the limitations involved, is another matter entirely.
1. Is adoption from a different selected function likely?
What Orr says in general regarding adoption from a different function is even more true of the situation of prebiotic "function" becoming converted to a working system for processing symbolic codes. Try listing candidate "functions" that would be selected in a prebiotic world of replicating molecules and see how rich the possible candidates would be for adoption.
Is there any selection scenario that has even a prima facia level of reasonableness, in light of the acknowledged limitation to current (not future) effects? Or are we supposed to rely on faith in an unknown scenario as the assumed best explanation?
2. Even if we waved away all selection unreasonableness, is it reasonable to believe that Before Language replication could ever construct the encoding, storage/transmission, retrieval, and decoding mechanisms required to create associated symbolic meaning?
Ignore all time and resource limitations as well as selection limitations. Does our best understanding of physical molecular replication indicate that it could construct and replicate these molecular mechanisms for symbolic processing? For an illustration of the problem (not the specific requirement), how close could physical replication come to making and replicating the integrated retrieval and decoding mechanisms we see now?
3. We know that designing toward future function is no problem at all for intelligent agents. We know that designing coding systems that process symbolic sequences is no problem for intelligent agents.
Given what we know now (subject to revision as that changes), which is the best supported inference at present?
a) Intelligent agency is uniformally required for construction of systems that process symbolic coded sequences.
b) Unguided processes can create such a system by a means we do not know and cannot describe. Despite acknowledged limitations on replication and against selection for future benefit, for reasons we cannot yet explain or support these limitations can be assumed/trusted/believed to not prevent evolution from accomplishing this transition from literal processing to symbolic processing.
Which inference is a uniform extension of existing knowledge? Which is an ad hoc leap taken in faith, despite existing knowledge?
Comment by eric — May 27, 2007 @ 8:57 pm
May 27th, 2007 at 10:42 pm
That is a sufficient clarification.
Then it becomes an empirical question requiring detailed study of the plausibility of the precursors and mechanisms"”but you claim this:
Hence subject to rigorous proof"”no recourse to empiricism necessary.
This is an argument that irreducible structures can't evolve. If it can be proven, then your argument stands.
Did you consider Orr's actual arguments, or did you just pick up that quote from some quote-mine? Did Orr cease being an evolutionary biologist when confronting this issue? Or did he pose those comments rhetorically in order to show why Behe is "just plain wrong"
In order for you to support your claim of the "impossibility of taking the path" you have to be able to exclude every possible pathway. I don't claim to know how to determine every possible pathway, or even that such a pathway can ever be found. Maybe there was Divine Intervention.
* Do you understand what it means to claim something is "inherent"
* Do you understand the nature of a formal proof?
* Am I correct that you claim that evolution is inherently unable to create Irreducible Structures?
Not in science, of course. There is no valid theory of the origin of the genetic code. Nor is there is a valid theory or theorem of a 'Barrier' to the evolution of the genetic code, at least not presented on this thread.
Addendum: This doesn't prevent speculation or the devising of testable hypotheses.
Comment by Zachriel — May 27, 2007 @ 10:42 pm
May 28th, 2007 at 4:20 pm
You are miscontruing the point in question here. The section you refer to did not say and is not claiming that there is no possibility of taking any path whatsoever. I said "The impossibility of taking the path I described is inherent." That line of analysis is showing the inherent futility of a particular class of (non)options that I described.
You had earlier made suggestions along the lines that evolution might be able to make progress varying the item sequences (e.g. "words"). I have been pointing out that unless and until you have retrieval and decoding mechanisms that implement the association between symbol and realization, mutating the item sequences is futile and provides zero progress, no matter how much you do it.
This futility comes directly from the nature of all symbols. This issue is an inherent issue, not an empirical issue. A sequence of itself is never a symbol. A sequence is made to be a symbol by association — something outside of itself. For a physically implemented case (e.g. DNA and proteins), the symbolic association is established by the conversion mechanisms, i.e. an implemented and functioning coding convention that can convert between symbols and associated realities.
Thus, I have been saying that the only possibly viable evolutionary path is one that constructs and implements the mechanisms for a consistent and coordinated coding convention (i.e. encoding, storage/transmission, retrieval, and decoding).
I've mentioned a few of the primary problems that arise for this scenario.
1) Regardless of selection issues, it is implausible that Before Language replication of individual molecules could scale up to physically replicate and sustain the complex mechanisms that would be required for coded message translation and processing.
2) Regardless of replication issues, it is known that selection only operates on current benefit and any intermediate stages of construction of one or more of the necessary mechanisms doesn't provide symbolic processing function (or any benefits thereof). Thus, it is acknowledged that no direct route to selecting for constructing the coding machinery exists. The nature of natural selection excludes all direct evolutionary routes.
3) An indirect route is also implausible (more on this below). Apart from vague speculations and hopes, there is no actual proposal to consider here. There is nothing that proceeds from a scientific basis rather than a faith basis.
[Besides direct routes and indirect routes, do you see any other possible pathways?]
Regarding the quote from Orr, I am quite aware that he considers the ID position to be wrong and I noted specifically that he was an "ID critic". That is precisely what makes his criticism of the "adopted function" strategy all the more significant. He could never be acused of trying to favor ID.
His remarks were in the context of After Language evolution, and so much of what he and others have to say about possible ways to construct IC systems in the After Language context does not apply to the Before Language case.
However, his obvious and severe skepticism about the plausibility of accidental adoption from selecting for some other function is quite relevant. His observation is about the nature of natural selection and function, something that does apply to the suggestion that natural selection Before Language might select for something unrelated to symbolism and then accidentally get symbolic processing in the bargain.
What this means is that there is an unmet burden of proof for the suggestion that selection might somehow select for some other mysterious property and then accidentally build all of the necessary mechanisms for symbol processing. If Orr finds this implausible for After Language evolution of a single IR system, it is only more so for generating multiple coordinated and synchronized mechanisms for the very first processing of symbolic codes. Until some relevant evidence or analysis is given to overcome this recognized implausibility, it remains an implausible scenario.
We are not in the position of knowing nothing about the problem. The real problem is that the more we know, the more unreasonable it seems that an unguided process could build the mechanisms required for encoding and decoding symbolic codes — especially since it would need to build such machinery even before symbolic coded messages exist! (The creation of the working machinery makes sequences symbolic.)
Nothing in how chemicals behave, or how B.L. replication works, or how natural selection works would lead us to expect that it would produce and use a symbolic code. The idea that it could ever do so is so far a leap of faith despite the contrary nature of what we have learned and what we could predict from what we know.
Comment by eric — May 28, 2007 @ 4:20 pm
May 28th, 2007 at 5:23 pm
eric:
You're wrong, eric. A sequence of RNA nucleotides can be regarded as a "symbol". The sequence is a symbol for the 3D structure of the RNA molecule, which determines its catalytic properties among other things. Change the sequence, change the 3D structure, change the properties. We don't know very well how to compute the map sequence -> 3D structure, but there is progress.
Natural selection of self-replicating RNA molecules has been observed in the lab. An RNA molecule can mutate into a more efficient replicator. Therefore symbols can evolve. It has been demonstrated. How this might have spawned the more complicated DNA->RNA->protein map is unknown at present. But to claim that there is some inherent barrier to symbolic processing is just wishful thinking on your part. The mere fact that we can make working symbolic models of chemical reactions as a matter of routine refutes your grandiose claim.
Instead of trying to burn down the efforts of many scientists working on OOL-related chemistry, perhaps you might spend some time working out the details of your favorite alternative hypothesis of design. Who designed life and how? The bare bones of a mechanism would be nice.
Comment by Raevmo — May 28, 2007 @ 5:23 pm
May 28th, 2007 at 6:40 pm
You're right that a sequence of nucleotides can represent a specific property of an encoded end product but wrong in thinking lab evolution accounts for symbolism.
What has actually been demonstrated is a capacity to engineer both replicators and initial conditions in a controlled environment to observe autocatalytic behavoir. Artifical selection can then kick in. What is missing is the realization that SRMs are not genomic storage units and that more efficient SRMs do not get one any closer to a cell. In fact they do not even survive real live non-lab environments.
There is no barrier. When intelligent beings assign meaning to the symbols the system works great.
You have working models of enzymatic replication. It is claims relating this to cellular life forms that are grandiose.
Comment by Bradford — May 28, 2007 @ 6:40 pm
May 28th, 2007 at 7:28 pm
Bradford:
Not artificial. Spontaneous, i.e. natural.
Granted, lots of intermediate steps are missing. Your response seems to be to give up and postulate an unknown designer and be done with it. That won't work because it so contrary to human curiosity.
Comment by Raevmo — May 28, 2007 @ 7:28 pm
May 28th, 2007 at 8:44 pm
What has actually been demonstrated is a capacity to engineer both replicators and initial conditions in a controlled environment to observe autocatalytic behavoir. Artifical selection can then kick in.
The initial replicators are contrived as are the conditions. Also, you better keep a good supply of nucleotides available if the mix is to grow and keep out any inhibiting substances.
You have working models of enzymatic replication. It is claims relating this to cellular life forms that are grandiose.
I'm not giving up on anything. There is much more to learn and I'm eager to continually add to my knowledge. I'm just not excluding a possibility a priori. That's what a commitment to atheism does for you. It commits one to an outcome on origins regardless of the evidence.
Comment by Bradford — May 28, 2007 @ 8:44 pm
May 28th, 2007 at 9:29 pm
You confuse the categorization with the thing itself. Consider a given sequence. At some point an association may occur. We then may call the sequence a "symbol""”even though it is the same sequence. There is nothing magical about this.
Yes, you keep saying that.
Plausibility is an empirical issue. You just said it was inherent.
This is another unsupported conclusion. Evolution may not be able to directly select for some future possibility of symbols, but the evolutionary path could conceivably be direct if a primitive symbiological system (or precursor) is advantageous.
It's not anyone's job to propose anything. You have said it is impossible and inherent. You could save people a lot of trouble looking if you would just provide the formal proof.
That is a misrepresentation of Orr's view. He calls Behe's arguments on Irreducible Complexity to be "fatally flawed, a "colossal mistake", and a "bizarre string of confusions and contradictions". Orr does not find the evolution of 'IR systems' to be implausible.
We don't have to know anything about the problem. You said it was inherent. That means you can devise a formal proof based in the definition of evolution and symbolic associations. You have repeatedly failed to do so.
I have suggested that you modify your claim. If you want to say that there is no valid theory of the evolution of the genetic code, and that you are skeptical that one can be devised, that would be much more tenable position. Then the argument would concern plausible mechanisms.
Comment by Zachriel — May 28, 2007 @ 9:29 pm
May 28th, 2007 at 9:39 pm
Yeah, way cool! A truly incredible feat of science. I understand the discoverers of ribozymes even won the Nobel Prize.
Comment by Zachriel — May 28, 2007 @ 9:39 pm
May 28th, 2007 at 10:17 pm
If you haven't bothered to carefully read through all of the preceding messages, I wouldn't blame you. There are quite a few.
The problem with your point is that you are substituting a different meaning to "symbol" as part of showing that my statement is "wrong". Although probably not intentional (due to jumping in?), changing the definitions of words in an argument to make a point is the fallacy of equivocation.
Everything is itself. With any structure (which is itself), if you change it you can affect it. By that standard virtually any structure "can be regarded as a "symbol"", but that is not the meaning of "symbol" in the question at hand. I'm not asking "How did the first structure arise?" or even "How did the first replicating structure arise?"
Earlier I granted to Zachriel for free the assumption for the sake of discussion that replicating RNA could exist and change. So clearly, everything you mentioned about replicating RNA, while having value on its own, is beside the point of the question at hand. That much is already taken as a given.
I am specifically talking about the origin of symbols and symbolic codes in the context something representing something else.
In order to become a symbol in the sense of the question at hand, something must represent "something else". How did that ever come to happen?
The obvious clear observable example is how codon sequences in DNA represent and code for amino acid sequences in functional proteins. Although that need not be the first such case, the question is indeed about how a case of "represents something else" first appeared.
In that context, my point is still true. The association is external to the symbolic sequence. The sequence by itself is never symbolic. It is only when you add an association to something else that you have symbolism in the sense of the issue I have raised. How does that happen by unguided natural processes without the prior aid of language?
Comment by eric — May 28, 2007 @ 10:17 pm
May 28th, 2007 at 10:31 pm
I'm wondering whether you intended to prove my point rather than to disagree with me. You pose this progression.
1. Consider a given sequence. (I'll call the sequence X.)
2. At some point an association may occur. (I'll call the association Y.)
3. We then may call the sequence a "symbol""”even though it is the same sequence. (emphasis added)
So, even though X "is the same sequence" at step 3 as it was in step 1, in step 3 "We then may call the sequence a "symbol"".
What is different between step 1 and step 3? In step 2 an association Y was added. X without Y is not a symbol. X with Y is a symbol.
Also notice, since X without Y was not a symbol, Y must be something in addition to X. It cannot be X itself.
You stumped me with an irrefutable argument. Lucky for me, you are describing what I've been saying all along. For X to be a symbol, you have to add Y, i.e. the association. X in and of itself is not a symbol.
[Bringing it back to application, in the practical case X = codon sequences. Y is the association, which in a physically implemented system is the set of coding conventions and mechanisms that associate codons with amino acids.]
p.s. I forgot to say thank you for providing the proof (or at least a clear explanation) I may have been lacking on this point. It was a magical moment.
Comment by eric — May 28, 2007 @ 10:31 pm
May 28th, 2007 at 10:44 pm
Not if your analysis is pegged to what science tells us. An amino acid encoding codon is absolutely meaningless in isolation. Besides the codons and amino acids there is an inherent need for more biomolecules in the process. Some sort of enzymes are needed as time is a crucial factor. That likely means twenty varieties. A translation mechanism is essential and more than this too but we already have a group of disparate molecules and unless you are invoking miracles while labeling them "natural" then we already have need of a multi-step, indirect process.
Comment by Bradford — May 28, 2007 @ 10:44 pm
May 28th, 2007 at 10:51 pm
Zachriel, are you now going to endorse an indirect pathway?
Comment by Bradford — May 28, 2007 @ 10:51 pm
May 28th, 2007 at 11:08 pm
No, I didn't. You are still conflating two different parts of what I have been saying.
The part that is "inherent" is the discussion of the implications of what it means to be a symbol. To put it in terms of my previous post, as you so clearly showed sequence X without the association Y is not a symbol. That is true and it is inherent to the meaning of "symbol". Consequently, efforts to turn X into a symbol without having Y are inherently destined to futility.
That is the "inherent" limitation that flows from the very meaning of symbols that represent something other than themselves. You cannot make progress toward symbolism just by making changes to X, the sequence. There cannot be real progress unless and until you add Y, the association.
The question of constructing the mechanisms that implement Y is not the inherent limitation mentioned above. Regarding Y, there are empirical issues.
I am not misrepresenting Orr's view. You are misrepresenting what I am saying about Orr's view. I did not say that Orr considers the evolution of IR systems to be implausible, as you seem to have thought. I am saying that Orr considers the specific proposed scenario of adoption from a different function to be implausible as the explanation for IR systems.
This is relevant because, as you have acknowledged, prebiotic systems cannot directly select for future function. When it is observed that symbolic code processing is a future function, this implies that direct route is excluded — it cannot select for future function.
I understood you to be suggesting instead that perhaps selection was selecting for some other function and that this later was utilized for the purpose of symbolic processing. If so, that is adoption from a different function. This scenario is what Orr considers to be unlikely. His assessment of that scenario is applicable, and perhaps moreso in the Before Language world of limited structures and functions related to replication.
[Enough for now. Gotta go.]
Comment by eric — May 28, 2007 @ 11:08 pm
May 29th, 2007 at 12:18 pm
An evolutionary process (in this context) must proceed by stepwise selectable changes. As far as I can tell, you are claiming that a symbiological system cannot evolve from simple replicators due to the inherent properties of symbols; specifically, the irreducible structure of the decoding process. Have I stated your argument correctly?
In the case of biology, no one knows. There is no complete theory of the origin of the genetic code. However, the hypothesis of an evolutionary origin has led to specific predictions that have been empirically confirmed.
Adding a primitive association can be a stepwise change and something that evolutionary processes are quite capable of doing. Try substituting "association" in your argument for "symbol". Primitive associations might evolve into complex irreducible structures, including symbiological systems.
Yes, that's what I thought your argument was. However, a primitive Y can be added to X by a stepwise change. X doesn't have to change. (We presume they would coevolve in nature, but this is not a necessary assumption in order to understand the abstract case.)
Orr doesn't deny that cooption occurs, only that structures aren't "recruited wholesale". He points to the evolution of air bladders into lungs as a case of adaptive cooption that leads (with the loss of gills) to an irreducible structure. A floatation device is coopted for breathing.
In our general case, Y might have a function that gradually binds and influences the activity of X. In our specific case, amino acids gradually take over the functions of RNA enzymes with primordial coding assignments plausibly associated with amino acid binding sites.
–
Addendum: Perhaps your problem is the assumption of any function being coopted. But on the molecular level, it is possible for primitive functions to arise de novo by a process of sifting. This is an important aspect of modern RNA research.
Comment by Zachriel — May 29, 2007 @ 12:18 pm
May 31st, 2007 at 9:38 pm
(Sorry to be late in responding, but my time has been taken by other things, and it remains limited and variable.)
If you would, please clarify what you meant by "Try substituting "association" in your argument for "symbol"." I'm not sure I catch your meaning there (I've made multiple arguments about symbols), and I don't want to miss it. Thanks.
About adding a "primitive association" Y, it is critical to keep in mind that by "association" in this context we don't mean just anything at all that could be called an association in any vague sense. We mean a symbolic association where one thing X (the coded sequence) is a coded representation for something else (e.g. the amino acid sequence for a protein).
For intelligent agents, a symbolic association can be held mentally in the imagination and understanding. For a physically implemented symbol system such as with the genetic code, the symbolic association is the set of mechanisms that translate between the coded symbols and the realized counterparts they represent.
Ergo, adding a "primitive [symbolic] association" means adding a primitive set of mechanisms to translate between symbol and realized counterpart. Given that one only has physical replication from available templates, it would not be easy to build even a "primitive" set of mechanisms that provide translation.
It is not a symbolic association for one thing to merely bind to another. Being a binding site is not being a symbol. Non-symbolic nature has abundant access to numerous binding relationships, but these are not symbolic.
You are certainly free to argue that a binding relationship could somehow "escape" into a symbolic association. But without elaboration, that is merely to restate the nature of the original problem. To begin to form an answer would require filling out a plausible progression that replaces any hand waving "escape" references. "How does it escape into symbolism?" is just another way to pose the original question about the origin of the symbolic.
Not quite. You still seem to be mixing two distinct issues.
1. Issue of Inherent Properties of Symbols (and trying to change X, the sequence)
When people are inclined to believe that replication applied to a replicating molecule solves the problem or that we are at least nearly there, that is where it is important to realize that replication of potential symbols, even with variations, makes zero progress toward the goal — not slow progress, zero progress. That is due to the inherent meaning of a symbol. What is needed is translation (i.e. the association).
Physical replication using a template could take a structure with parts a-b-c-d and produce a duplicate a-b-c-d.
Symbolic translation effectively says,
- whereever you see "a" over here, put a "q" over there
- whereever you see "b" over here, put a "w" over there
- whereever you see "c" over here, put a "j" over there
- whereever you see "d" over here, put a "m" over there.
That treats a-b-c-d as a symbol sequence because each symbol represents something other than itself.
Consequently, translation of coded symbols is profoundly different from physical replication using exact copying of a supplied template. Progress toward symbolism means progress building translations between symbols and real counterparts.
2. Issue of Constructing Symbol Processing Mechanisms (i.e. adding Y)
Once someone understands that symbolism requires building the mechanisms that provide symbol processing (e.g. encoding, storage/transmission, retrieval, decoding), it is possible to move on to the issue of whether it is reasonable to expect unguided natural processes to create such translation mechanisms. These mechanisms are not a single irreducibly complex mechanism, such as a flagellum. They are distinct but synchronized and coordinated mechanisms. There are numerous issues that come up when facing this problem.
Replication is relatively straight-forward and "mindless" in the sense that there are no interpretation points. What goes in comes out. Translation of codes involves implementation and execution of symbolic associations where the constructed result is something other than the symbol. It is a very different problem than replication and much harder.
First to clarify, Orr's objection to "recruited wholesale" was not simply a rejection of "recruiting the whole mechanism or structure". He said:
In his illustration he mentions adopting "half your car's transmission". He is not merely objecting to adopting the whole transmission, but even to significant parts of it.
The plausibility of adoption from a different function would clearly depend on the difference or similarity of the other function. Air bladders into lungs? Perhaps. Transmission subsystems into airbags? Not likely.
Other than replicating, what "functions" does a world of replicating molecules engage in? Translation is profoundly different from physical replication copying from an existing template. What is the "other function" that provides selectable traction for building something very like a symbolic translation system (even when symbols do not exist)?
One further difficulty is that making a replicating molecule more elaborate easily works against replication. The more there is to replicate and the more complex it is, the more opportunity there is for something to go wrong or a lack of resources to prevent replication. In some cases, changes to physical shape could impede further replication by working against functioning as an accessible/readable template.
Take any replicating molecule and change it, either by adding something or replacing something. One of two things will happen.
1. It can continue to replicate (possibly more or less readily).
2. It no longer replicates — a dead end, historically.
Supposing that replication is preserved, you then still have a replicating molecule. It is different, but its nature is still one of a replicating molecule even if parts are replaced or added. One has not yet escaped from riding a merry-go-round.
Comment by eric — May 31, 2007 @ 9:38 pm
June 1st, 2007 at 8:39 am
You use the word "association" appropriately in your definition, here:
Correct, a symbol system is a type of elaborated association. And your claim is that this is a mechanism that *cannot* evolve.
Not at all. A primitive association might simply be a sequence of nucleotides that codes for an amino acid, and there is evidence of even more sophisticated associations.
Your premise is incorrect. We know that primitive replicators can evolve into primitive communities of cooperating molecules (partners for the ribozyme, but no longer autocatalytic).
I understand the distinction, but it is a primitive association.
Keep in mind that *you* are the one making the strong claim, that such an escape is *not* possible.
This section appears just to define the system in question.
This seems to be an argument that Irreducible Structures can't evolve. Though it is your responsibility to support your strong claim that symbolic systems *cannot* evolve, I would be happy to examine the issue in some detail.
First, can you admit that primitive associations with replicators are not only possible, but observed?
Second, can you admit that your "proof" is nothing more than an elaboration of Irreducible Complexity arguments (super-duper Irreducibility)? Keep in mind your definition that symbiology is "the set of mechanisms that translate between the coded symbols and the realized counterparts they represent."
Comment by Zachriel — June 1, 2007 @ 8:39 am
June 2nd, 2007 at 3:39 pm
Not quite accurate. You tend to say "cannot evolve" which is ambiguous. I've never said that such systems cannot change once they exist. The claim is that unguided evolution cannot create such mechanisms and systems.
Also, my claim is not that anything that might be called an "elaborated association" by any meaning of those words is excluded. My claim is specifically about the origin of the first symbols and the first symbolic processing/translation system.
My premise did not exclude communities or require replication that is "autocatalytic". What I have said is that you are limited to "physical replication from available templates".
In other words, even supposing "communities", you do not yet have construction from symbolic instructions (which would be a question begging fallacy). For "reproduction", you are strictly limited to construction by means of physical replication using some existing template as the model. That does not mean the copying must done/catalyzed by another copy of the template. It does mean that all copying must be done using an actual instance of the original as the model or template. The premise I gave is correct.
If you understand the distinction, that implies that you recognize there is a distinction. When a line of reasoning substitutes a distinct meaning of a term (e.g. for "association"), that is called the fallacy of equivocation.
By striking out "symbolic" from my statement, you are substituting a distinct, non-symbolic meaning for association. In a context where the question is whether evolution could create a symbolic association, that substitution is equivocation.
Furthermore, "A primitive association might simply be a sequence of nucleotides that codes for an amino acid" equivocates on the meaning of "codes". Unless and until there is a translation from the sequence of nucleotides to the amino acid, they are not codes in the symbolic sense we have been discussing. If one were to use "codes" in the symbolic sense we have been discussing, your claim in that paragraph would simply be false.
To be symbols in a code, they must represent the amino acid, not merely attach with it. When implemented physically, symbolic representation means that the symbols translate into the realized counterparts. Attachments are ubiquitous in nature, yet no mere binding is a symbol.
As I said before, nothing prevents you from trying to legitimately reason from a starting point where there is a binding site or a chemical attachment. You can do so without equivocation by simply using other terms that do not introduce distinct non-symbolic meanings for "code" or "association". To mix such alternate meanings into the discussion is in no way necessary, and only results in the fallacy of equivocation.
The trouble with clear communication is that it makes things clear, including the fact that "escaping" from a bond or from a role as a binding site in no way implies creating a symbolic relationship.
Thus, when papers vaguely refer to "triplets having escaped from their original function in amino acid-binding sites", the use of "escape" skillfully hides the real difficulty — as though it were simply a matter of "leaving a restriction or limitation" rather than one of constructing the very first system of coordinated mechanisms for symbol translation.
Regarding your final two questions
1. Please state your claims about replicators in chemically equivalent terms that do not lean on distinct equivocating meanings to "association" or "codes". I will not agree to equivocations about key terms. Any legitimate point can be made without equivocating on these terms.
2. It is inaccurate to characterize the problems as Irreducible Complexity. That term has a meaning that does not fit in multiple ways. For one, we are talking about multiple mechanisms, not one. For another, I have not strictly claimed that any single mechanism in the system is irreducibly complex.
As just one difficulty (among others), I do point out that selection cannot select for future function or future value. Thus, there is no traction for evolution to select for building symbol translation when this is in the future. Nor has anyone yet identified any plausible alternate function that would be available for selection and that would be nearly the same as symbol translation (such that it would give you something that could accidentally mutate into symbol translation).
Starting from what we know about how selection works, we have no way to say "From this it would be reasonable to conclude that evolution would build symbol translation mechanisms, because…" The belief that evolutionary processes are capable of this is ad hoc and without basis in evolutionary theory.
Comment by eric — June 2, 2007 @ 3:39 pm
June 3rd, 2007 at 10:22 am
*Cannot* evolve from primitive replicators.
You just called it a "symbolic association". Symbol systems are a member of the
class of associations. They are elaborated, rather than primitive.associated, to join or connect together
Nonsense. It's not equivocation. Symbol systems are members of the
class of associations.Yes. That's right. A primitive assocation does not meet our working definition of symbol system, the set of mechanisms that translate between the coded symbols and the realized counterparts they represent, as it lacks the translation mechanism.
A primitive association can be made in principle and is observed to occur.
Huh?! Remember, *you* added [symbolic] to *my* phrase. I accepted that as a rhetorical device, but it did change my meaning. I changed it back. I did not claim that a primitive association is a symbol system per our definition. However, it is a member of the
class of associations. I even used the distinct phrase "primitive association" as to avoid misunderstandings.The rest of your post elaborates on the misunderstanding.
Comment by Zachriel — June 3, 2007 @ 10:22 am
June 3rd, 2007 at 9:12 pm
In this thread, the first twenty uses of "association" where all made by me, either in posts of my own or in a few cases other posts quoting my posts.
The first 36 uses of "association" where all in the context of symbolic/coded associations, i.e. between a code or symbol and that which it represents. For some of the latter uses in this 36, I introduced the variable Y to refer to the symbolic association that is needed.
Uses 37 through 40 then occur in one of your posts, within this exchange (emphasis added):
In all previous references "association" and Y both referred to symbolic associations. In your response to those previous statements, you claim that "a primitive association" and "a primitive Y can be added to X by a stepwise change".
If by "a primitive Y" you intended "a primitive [symbolic] association" then that would be an unsupported claim — even a question begging claim, since it is what needs to be shown. However, it would at least be true to the preceding context. On the assumption that one is not equivocating, that was exactly what I wrote.
If by "a primitive Y" you are there changing the meaning of Y from a symbolic association to some non-symbolic association, then that is a clear cut case of equivocation. In your response, you've changed the meaning of Y and of association in the middle of an existing context.
To illustrate, my argument was saying "you need to be able to add Y (i.e. a symbolic association)".
Removing the equivocation, your response becomes "Yes, that's what I thought your argument was. However, a primitive Z can be added to X by a stepwise change."
Notice that once the equivocation is removed, the argument becomes a non sequiter. It breaks down. It has a complete disconnect. What is needed is Y. You are pointing out that evolution is capable of adding Z (e.g. chemical bonding).
The illusion that a Z counts in some way as a Y comes only from equivocation. They are chemically distinct realities.
You still have not shown any reason to believe that unguided processes can create Y, a true symbolic association expressed by the translation mechanisms to connect coded sequences with realized proteins. In particular, you haven't shown any reason to believe that
a) During stepwise construction, Before Language replication could physically replicate the accumulating molecular machinery required for translation and reliably retain a coherent system (no symbolic instructions to follow).
or
b) During stepwise construction, selection would have any basis for preferring to build the required molecular machinery, given that symbol processing would be a future function.
Neither question begging, equivocation, nor non-sequiters have advanced the case for the plausibility of building translation mechanisms from unguided processes. It remains a faith-based proposition, not an inference from the nature of evolution. Evolutionary theory does not give us a reason to expect it can get this job done — not even a prima facia plausible story.
Comment by eric — June 3, 2007 @ 9:12 pm
June 3rd, 2007 at 11:27 pm
This is why I don't think you understand that nature of a formal proof.
Symbol systems are members of the
class of associations. There are other types of associations, including simple or "primitive associations". I am clearly making a distinction. I really don't know why you are having so much trouble with this. It's not all that difficult.Comment by Zachriel — June 3, 2007 @ 11:27 pm
June 4th, 2007 at 12:21 am
Unlike a mathematical proof dealing with abstractions, the logic of which prevails irrespective of "empirical occurences," the type of symbolism discussed here references a physical cause, even if its physical nature is nothing more than the manifestation of a teleological process. It seems to me the perspective is off. What we should be looking for are comparative models and an assessment as to which one(s) better fit the observed outcome. A model explaining an intelligently directed outcome should contrast with a purely physical one in ways that allow for predictions. Eric and I have argued that the symbolism is the consequence of assignment. If that is so then a genetic code should be possible that scrambles current assignments but still allows for function. For example, CAG could be assigned glutamic acid instead of glutamine, UGG histidine rather than tryptophan etc. The physical manifestation of this would be altered tRNAs. An entire set of codons replete with rearranged symbolic representations would indicate that physical necessity is not a determistic factor. The practical difficulties involved in an experimental demonstration could be considerable given other structural entities involved in gene expression but in principle I see no reason why this could not occur.
A related prediction concerns a causal generator for tRNA. None is known outside a cellular environment other than expert manipulations of highly trained humans. But if a natural process were discovered capable of generating these biomolecules and the process randomly hooked up amino acids with varients of current tRNA models, physical necessity would also be called into question. That leaves selection of course but that's a story for another night.
Comment by Bradford — June 4, 2007 @ 12:21 am
June 4th, 2007 at 8:09 am
Eric claims there is some sort of 'Barrier' that is inherent in the evolution of primitive replicators to symbol systems. That is the claim I have an issue with. He has failed to provide a proof, nor have we been very successful in even defining the terms involved. We have thus far,
Then I formed a
class, theclass of associations. Symbol Systems are members of theclass of associations.I had fully expected long ago for alter-eric to say something along these lines.
Alas!
This is an entirely different question. All the data currently supports some sort of spontaneous process, but there are many gaps in current understanding. There is significant research in these areas.
There is some evidence that the current genetic code is a result of primitive bindings. There is some evidence that they escaped from these requirements and became fully flexible. There is evidence of how tRNA could evolve. The "assignment" doesn't have to be teleological, but does appear to have a discernable history.
Comment by Zachriel — June 4, 2007 @ 8:09 am
June 6th, 2007 at 11:21 pm
One indication we already have of this would be the distinctions between the various genetic codes, where a codon can be seen to code for different results.
Likewise, in regard to venturing into the realm of codon reassignment that you allude to, I found interesting the section on "3. Miller's references to biotechnology do not accurately represent the experimental literature on variant genetic codes." in this article:
Reply To Kenneth Miller On The Genetic Code
Another aspect of interest is raised by mathematician David Berlinkski:
That is because, while a given genetic code maps every codon to one amino acid, the amino acids are each mapped to by multiple codons. It is many to one, which would make the inverse one to many.
This creates a severe problem for the origin of the original sequences for proteins.
Trial and error starting from random symbol sequences is ruled out for multiple reasons. Here are some that come readily to mind
a) The search space is too vast. (It is the same search space that was already considered and rejected for the origin of proteins.)
b) There is no dictionary of correct sequences to compare to.
c) Sequences in the coded form do not have the function of the realized proteins, so that is inaccessible to selection.
d) There is no viable hypothesis scenario for constructing the decoding machinery in advance of having coded sequences to decode and/or in advance of having functional benefit from realized proteins.
In the prebiotic world, the only potentially viable source for correct protein amino acid sequences would be to derive that faithfully from actual realized and beneficial proteins. However, this runs into two other difficulties.
1. It implies that realized proteins need to precede the construction of symbolic encoding and decoding for realized proteins. (Nature cannot be expected to build complex machinery for the benefit of an improbable future invention.)
2. It implies that nature would need to overcome the point raised by Berlinski, i.e. encode from proteins to symbol sequences, despite the fact that (for the modern code at least) that is a one to many mapping.
There is another issue in that we don't see nature currently performing this encoding operation. However, I would accept that that could be because the need for it disappeared and it has been abandoned and lost. So that by itself need not be a strict objection.
Nevertheless, the problem of getting off the ground in the first place is a very serious one.
Comment by eric — June 6, 2007 @ 11:21 pm
June 6th, 2007 at 11:32 pm
p.s. There is an obvious problem with the fact that you need realized proteins before you can symbolically represent them. This means proteins need to be both created and sustained in a prebiotic world with only Before Language replication — not yet having the benefit of construction according to the future symbolically encoded instructions.
That in itself is a nest of problems even before you reach the issues of inventing symbolic representation of sequences.
Comment by eric — June 6, 2007 @ 11:32 pm
June 7th, 2007 at 6:25 am
Eric writes:
After reading this it occurred to me that the relationship between an expressed amino acid and its symbolic codon representation can be seen as follows:
If B represents one of twenty amino acids and set A represents the set of codons, which when expressed, results in the addition of B to a peptide polymer then:
If set A then B expresses a logical truth. This also:
If not set A then not B.
The syntax of the logical expresions could be more economically written but since the above describes what we observe in the present and at some time in the past we know that the logical relationship between the two did not exist, then how is it plausible to assume this logical relationship came about as the consequence of an unintelligent process? Incidentally there is no chemical necesssity evident even in cells. Only options where specified expression is concerned.
Comment by Bradford — June 7, 2007 @ 6:25 am
June 7th, 2007 at 11:33 am
That can't possibly be a statement by a mathematician. You yourself called it "a one to many mapping". We could map randomly, or it could done deterministically, e.g. some preference scheme.
However, there is no encoding scheme with regards to contemporary genomes. Rather, genomes are replicated and then change slowly in response to evolutionary processes.
It's almost as if you didn't read any of the research cited in the article that tops this thread.
Exiting an RNA World
Comment by Zachriel — June 7, 2007 @ 11:33 am