In the lab, e. coli evolved the capability to grow on citrate. This is new information being created.
Every mutation of any genome in any species results in the creation of new information. Sometimes such information is harmless (called a neutral mutation), which leads to its random spread within the population and doesn’t affect the bearer one way or the other. Most frequently, mutations are harmful, meaning they lead to fewer offspring for their carriers and they decrease in frequency within a population. Rarely, however, a mutation is beneficial, which leads its carrier having more offspring than average, which then can lead to an increased frequency of that gene in the population. Sometimes, environments change and mutations that were harmful, or at least not adaptive, become useful. For example, a mutation that leads to an increase in size can be harmful if the climate is warming and beneficial if the climate is cooling. Case closed. It’s all very simple and quite straight forward. The fact that people even provisionally accept nonsense like that propagated by IDers hows that they don’t understand what information is or how genes and mutations operate. Never accept a statement by an IDer as honest; always question all statements from any source and try to investigate its truth or falsity. IDers propagate and a lot of the latter.
While I know you are correct, I would want to introduce a VERY minor quibble.
In Information Theory, information is (in part) defined by its difference from randomness, and by its predictability. In English, for instance, the letter ‘q’ is nearly always followed by the letter ‘u.’ That’s a high degree of information. The letter ‘e’ could be followed by darn near any letter, and so when you see an ‘e’ you have a fairly low degree of information regarding what is to follow.
If a mutation involves the altering of a sequence of DNA – sort of like changing a letter in a sentence to some other random letter – doesn’t that reduce the information we have regarding that sequence? The sequence of DNA is now closer to being random than it was, because a cosmic ray or free radical or whatnot came along and made an unpredictable change?
As you note, some mutations are “fortunate” ones, and code for beneficial traits, and are thus preserved. By being preserved and disseminated widely, they become more commonplace in living things…and thus more predictable. But unfavorable mutations are (very roughly) akin to taking a manuscript and randomly changing a letter, which doesn’t increase the information in the text.
To be or not to be, thyt is the question…
Of course it is observable, if you don’t expect Monkey to Human in an observable time scale.
The domestication of Teosinte into Maize over the course of Human Civilization is a form of Evolution. Human guided, but still evolution of one plant to a completely dissimilar form.
Numerous other species have adapted and evolved to take advantage of Human civilization.
It would be silly to say that every form of Life was ‘created’ to fit it’s (often narrow) ecological niche, ignoring the fact that those niches have come and gone over the course of the epochs. Who created the life forms that occupy niches that didn’t exist 100 million years ago? 5 million years ago? I mean, if you have any sense of the true life span of the Earth and how things have evolved and changed over the last half-billion or so years of “Life”, but yet you believe in Creationism and NOT Evolution, then you would have to believe that this creative force (which somehow isn’t evolution and adaption) is still spontaneously creating new life forms even to this day, because otherwise, how would they even exist if they didn’t 5 or 500 million years ago?
So…
You would have to believe that Monkeys, Humans, etc just spontaneously sprang into existence, fully formed from the head of Zeus/Jehovah/Allah/Cthulu/Justin Bieber.
Which is way more absurd than the idea that life forms slowly changed over thousands and even millions of years to adapt to and occupy their various niches.
Is there a debate, though?
I was under the impression from other discussions that Information Theory defined information by the number of bits needed to transmit it thus random strings would have more information than a predictable string (Q always followed by U for example). Based on my understanding what you refer to above is more based around meaning than information. Was my previous understanding off base?
Part of the problem here is with terminology. A term such as “information” may have fifteen mutually independent and precise meanings in fifteen different disciplines, and none of them matching what a layperson would mean in casual conversation.
Understood, and I would not have mentioned anything except for the specific mention of Information Theory. Without that specific phrase I would have treated the answer as using a layman’s meaning and interpreted the post thusly.
As a caveat, let it be understood that I don’t know shit about “Information Theory” and everything I have heard about it makes a sophomore bull session sound like Wittgenstein.
Well, evolutionary “information” really does have “meaning,” in that the DNA code segments really do express such things as cell membranes, genes for eye-color, nitrogen-fixing in bacteria, etc.
But I was at a more abstract level: a mutation, if it is actually random (cosmic ray hits a DNA strand) will tend to reduce the amount of information. However, the natural selection process will tend to weed out those DNA strands that are “too random,” because they will lack the useful information that codes for the ultimate meaning of all: survival!
Well, no. It is a formal mathematical science. It pops up in such things as Morse Code. The more common letters of the alphabet were given the short codes – E and T as “dit” and “dah” – while the less commonly used letters were given longer codes – like J = dit-dah-dah-dah. (La di dah!) This makes the overall message shorter. Later refinements include compression (treat “qu” as one symbol) and check-sums to protect against data loss.
But, yes, some of the deeper arguments can become kinda bullish. It’s like how honest Quantum Theory led to hogwash like “The Dancing Wu Li Masters.” You can’t do gardening without getting to know both weeds and crap…and when you finally do reap, all you get are bushels of corn!
I do get that evolutionary information has meaning but I was under the impression that specifically in ‘Information Theory’ the higher degree of information would be the random string. You stated that the predictability of U following Q resulted in high information and linked that statement to ‘Information Theory’. This is the opposite of what I would expect given my other experiences when people explain ‘Information Theory’. I would describe U always following Q as providing more meaning that a random letter following E but the random letter providing more information than U following Q as you always expect the letter combination.
Oh well, I really have no understanding of ‘Information Theory’ and will just let this hijack sink slowly into the depths of time.
It’s hard to believe that some people don’t understand even the basic mechanisms of genetic mutation or that “information” is coded in the sequence of DNA bases. It’s been 30 years but I’m pretty sure I learned that in high school biology class. It’s not a very difficult concept.
Unless biology is no longer a high school requirement. Is it? Well, I suppose, even if it is, part of the problem could be that creationisms is reducing science literacy in the US because biology teachers are avoiding teaching human evolution and sometimes evolution all together. Then there’s some who are actually teaching creationism in a biology class. Really disturbing.
You are correct. Information in information theory terms is almost the opposite of what we mean by information in non-technical language. A totally random string of characters has no information in the layman’s sense, but maximum information in the information theory sense.
Which is why the new creationist tack of saying that information cannot be created is so bogus - doubling the size of a random string doubles the information therein.
If anyone wants to experiment with this, all they need to do is to generate strings of characters using different models, zip them, and look at the size of the resulting file. A strong of all As is going to come out smaller than a truly random string.
And anyone who has done an acrostic puzzle knows how little information is in English, since you can often guess large words with only a few letters.
I’ve actually taken an information theory course in grad school, and I once had a lab across from Shannon’s office - not that he ever came in by that point.
I think part of the problem is that a large segment of the population doesn’t care and quite a few truely are incapable of understanding things at the high school level.
Thank you for confirming my understanding!
Well, kind of argh. I took a couple classes on the subject, and somewhat know what I’m on about. I’m entirely capable of fucking it up, too. Anyway, it struck me as wrong that all mutations “add information.” They definitely change information, but they destroy (or reduce) the meaning.
After all, if a DNA sequence were entirely random, then, while it might contain a maximum of information in one sense…it would be of absolutely zero value in an evolutionary sense! That’d be the fastest extinction event ever!
(I had remembered information being defined as what was necessary to distinguish a sequence from randomness. The sequence 111111 has a lot of information, but the sequence 273866 has less, because it isn’t clearly distinguished from randomness. If I’ve got that backward…wouldn’t be the first time!)
Which is precisely why creationists like the term. It’s like their use of “kind”. Imprecise definitions are much easier to twist around lies.
In any case, my bad. The key is that LdavidsSkeptic is entirely correct in that the process of evolution does create new working segments of DNA. It’s perfectly scientific. Darwin showed the basic model of how it functions: everything mutates…and the beneficial mutations are reproduced. It has been observed in the lab, and it is the only explanation, so far, for the diversity of life on earth. The creationist explanation doesn’t even come close, and almost entirely depends on deliberate distortions of facts.
What I find astonishing is that Darwin, from observing living animals – breeding pigeons, in large part – actually inferred the granularity of animal genetics. He saw patterns that indicated clumping and lumping of reproductive traits, long before the discovery of chromosomes (and, of course, without knowing of Gregor Mendel’s observations.)
Now look at a maximally nonrandom DNA, say gggggg… Do you think that would be especially useful evolutionarily? (And indeed, such parts of DNA used to be considered junk, but I seem to recall that recently, some uses for it have been found, just not coding ones; rather, they may perhaps serve as structural filler, or something like that.)
And why do you think that a maximally random DNA would be evolutionarily nonsensical? If I want to transfer a message to you, the shortest way I could do that would be using a string of bits that’s maximally random, because only then does every further bit you receive also add another bit of information you get. If the string weren’t random, then it would follow some regularity which would enable you to at least to a certain extent predict the next bit you’d be getting; that bit therefore would not add a new bit of knowledge. Of course, redundancy carries the very important function of error correction: since not every bit is a bit of info, you can loose a bit and still recover the full message from redundancies; but just information-wise, novelty is tantamount to content. (Think about how few texts ever consist of just the same word being repeated thousands of times; in principle, any word you can predict is a word you can leave .)
This is indeed getting things exactly backwards. Use the analogy Voyager provided: a string of all 1s is very compressible, while a random string typically can’t be compressed at all. Random here essentially means that it has no redundancy, that is, no part of it is already implied by the rest, while in contrast, knowing just a single digit of the all-1 string already means you know all of it; the further digits don’t add any new knowledge.
ETA: Sorry, hadn’t seen your last post above.
Of course, as pointed before, Darwin still got some items wrong, but it is really silly for the creationists to expect that finding one or a few wrong items means that we should overturn all the theory.
It does not work that way in science, the problem for the contrarian is that it has to also propose an alternative that works. Otherwise the needed changes and modifications actually tell the scientists that the overall current theory is more robust now with the slight changes that were made.
And that is happening with evolution.