Today is Darwin Day, or so say a bunch of yahoos from Palo Alto. The Darwin in question, of course, is Charles Robert Darwin. He's the fellow who penned "Origin of the Species" and introduced and promoted environmental selection and common ancestry, Evolution in other words, as the dominant explanation for the variety of life on this planet. It is somewhat ironic that Darwin started out a deeply religious man. However, investigation of natural philosophy led him to question traditional expression of faith. It all but disappeared when his daughter died when he was in his early 40's. Thereafter he considered himself an Agnostic.
It is probably not well known, but one of the specific things that caused erosion of his faith was the idea of Intelligent Design. At the time it was promoted as Beneficent Design. As he traveled the globe in the Beagle, observing the creatures of the world, he found examples that led him to question the "beneficent" bit and he sought a better answer for himself.
OK so "Beneficent Design" was off the mark, but that does not invalidate Intelligent Design, does it? Well, this will be another post that tries to address the difference between what is real and what gets spun in politics. "Evolution is just a theory", it's detractors claim and say "Intelligent Design is a theory too". On the basis of this it is proposed that Intelligent Design should get equal time in education. This is not consistent with what "Theory" means in science.
Intelligent Design has some good science in it under the layers of politics. Specifically, the arguments based on information theory show some promise. What is that, you ask? Well, if you have not investigated Intelligent Design as a theory, then you probably just know it as: "Life on this planet is too complex to have arisen by happenstance and therefore it requires a creator." However, in science, this can not be taken simply on faith, it must be tested and not found lacking. Those working on Intelligent Design work in several directions. Mostly this involves looking for examples that simply could not explained by evolution. The least valuable of these take the form of, "look at the human eye, there is no way it could have been created by evolution." I take that only to indicate a lack of imagination on the part of the speaker. The more valuable arguments are being made with the application of hard science and mathematics.
DNA is a code for life. This code carries information. Not all of the sequences in DNA carry useful information. The ratio of useful to non-useful information is 1) a number that can be pinned down and 2) a number that can be predicted by the application of Evolution. The folks working this aspect of Intelligent Design are thinking that evolutionary processes could not produce the density of information based on information theory. I'll leave further understanding of this as an exercise for the reader, but suffice it to say that this has some scientific merit.
I used a word in the last paragraph, "predicted". That word is important in this debate because it characterizes the concept of "theory". A theory, amongst other things, is something that can be used to predict. This is one of the reasons why Intelligent Design is unpalatable presented as a theory. It seems to seek to throw it's hands up and fall back on faith, especially as it is presented by those seeking to shape public opinion. One of the things about Evolution is that it is a fine predictor. I'll explain more a little further down.
The main thing that characterizes a theory, though, is that it has been tested, either through observation of past evidence or through controlled experiments. Evolution has stood the test of both. No other theory has so well explained the apparent progression of species over time as evolution. Yes, there are holes in the record, but they are un-proven opportunities to invalidate Evolution. Those opportunities, as yet, are unsatisfied. Based on the overwhelming evidence in the rest of the historical record, it is reasonable to assume that those holes would be filled similarly. To date, no strong evidence of Evolution being false has been promoted and found valid.
Furthermore, Evolution has been tested. Scientists have observed speciation 30 or so times. Unfortunately for the layperson, none of these have been obvious enough (such as a dog morphing into an otter or some such) as to be convincing. This is, of course, mainly because our opportunity to observe these changes has been pitifully short compared to the history of life. However, tests have been conducted with Evolution as the idea being tested and the results have been positive.
Detractors of evolution then turn to something that Evolution does not attempt to explain and say that because it does not it must be false. This is tantamount to claiming that because a car can not fly it is not a vehicle. The something is abiogenesis. Abiogenesis the creation of life from non-living matter. For some faithful, there is nothing contradictory with saying God stuck his finger in the muck and, "zap," we have cyanobacteria and the rest is Evolutionary history. Others are offended at the thought of having a bacteria as a great great ... great grandfather. Nevertheless, Abiogenesis is a separate topic that has many theories of it's own.
Intelligent Design is a theory that has barely made it past the stage of hypothesis. It is very poorly tested at this point. That does not mean it is false, it simply means that it is very questionable. Evolution, on the other hand, has been very well investigated and found to be, many times over, an adequate theory to explain what has happened and what will happen. When theories become this well established, this is when they should be introduced into lower education. Intelligent Design, as it stands, is pretty much still a topic for graduate level scholars to puzzle through.
Happy Darwin Day!
Subscribe to:
Post Comments (Atom)
8 comments:
You wrote:
The ratio of useful to non-useful information is 1) a number that can be pinned down and 2) a number that can be predicted by the application of Evolution.
I submit that the ratio is very near 1. Just because we haven't figured out the usefulness of each "bit" doesn't mean all "bits" are not useful. Huge sections of "non-coding" DNA used to be thought of as mere "junk". Now we know that some of it has uses beyond basic protein construction. We are continuously learning new connections. When do you suppose that we will be done deciphering the entire mystery so that we can "pin down" the ratio?
How would you propose that the "application of evolution" could "predict" this number? Perhaps you are referring to some time in the far distant future when our undestanding of complexity theory and information theory are much more advanced than today?
So, to be clear, the accuracy of using Evolution to predict this number was not described, just that it could be used, and thereby tested using information theory. Further, since, as you suggest, that ratio is moving towards 1 as we learn more, that tends to lend more credence to this arm of investigation into Intelligent Design. However, there is much more supposition to this argument than simply that ratio so this remains much farther from proven than evolution. In other words, it has not yet actually been generally satisfied that DNA is too complex.
As for the specifics of how, a model can be built based on certain rules of natural selection, otherwise known as "fitness criteria", starting conditions for life can be described and rates of mutation applied. I am sure there are a great many other variables that would need to be accounted for. Using this process, one might make a guess how much organized complexity can be generated in a certain period of time. Without such a guess, it would be otherwise impossible to say that DNA is "too complex" and that further, that complexity denies Evolution. If that was, in fact, the finding, then all the rules and variables would have to be carefully justified and verified through peer review. If the argument was sufficiently convincing, then, eventually, it would become a dominant theory. That may take years or decades and the theory may even languish for political reasons, but if it was truly better, it would eventually take precedence.
You wrote:
Evolution has been tested. Scientists have observed speciation 30 or so times.
Define "speciation".
Biologist often define speciation as the point where organisms become reproductively isolated.
There are a couple of problems with this definition as it applies to the theory of evolution.
First, it limits the usefulness of the definition to observations of only living organisms. It cannot be used to say anything about the evidence of speciation in the fossil record. Archeologists have a different definition and I don't think we want to explore the scientific foundation and testability of morphological speciation...
Second, it doesn't even say anything reliable about our observations of living organisms. For instance, lions and tigers are generally considered to be different species, yet they are not reproductively isolated. They can mate to produce ligers and tigons. Of course, there are other examples of interspecies reproduction (AKA hybrids) in both the animal and plant kingdoms - the mule being the most well known.
To further confound the situation, let's consider dogs. All dogs are considered to be of the same species, yet some are reproductively isolated (from a practical perspective). In other words, a Chihuahua and a Mastiff would be incapable of breeding naturally (setting aside the possibility of artificial insemination - that would be cheating).
Speciation is often tossed about as if it was some evolutionary event, but in reality it's just a bunch of people who agree to draw arbitrary lines and create artificial categories. When these same people get together and draw an arbitrary line between one organism and another and call that a speciation event - that is NOT scientific evidence of evolution... That's politics...
Of course, there are other definitions of speciation, but I contend that they all suffer from the same fallacy - they try to define something that doesn't exist. The process of evolution creates a continuum, not a sequence of distinct, individuated categories.
On the issue of speciation, I disagree. You've picked out corner cases and some exceptions to the rule, but in general speciation means two populations of organisms are no longer capable of exchanging genetic material. However, two organism not so capable are not necessarily separate species, i.e. dogs. In many cases, classification of various species predate modern understanding of genetics and so err. For a much more comprehensive basis on which to consider speciation, check out: http://en.wikipedia.org/wiki/Speciation
You wrote:
speciation means two populations of organisms are no longer capable of exchanging genetic material
This just replaces one flavor of arbitrariness with two more flavors.
Let's say that we have two populations of an organism that become geographically separated. A million years later scientist decide that the resultant generations now represent two different species. At what point do we say that speciation occurred? At the time of separation? A million years later when the scientists all agree? Both of these boundary conditions are obviously arbitrary. Since there is no way to tell when (or if) the two populations became unable to share genetic material, any attempt to define when or how speciation occurred is purely arbitrary and subjective.
Also, it is not clear what is meant by "capable" of exchanging genetic material? Capable could mean "reasonably able" or it could mean "genetically compatible"? If we mean "reasonably able", then we have to define arbitrary criteria for what we perceive is reasonable and what is not. If we mean "genetically compatible" then we need some way of measuring compatibility. The obvious criteria would be based on DNA, but what percent difference in DNA differentiates a species? A horse and a donkey have a greater divergence in their DNA than does a Monarch butterfly and a Viceroy butterfly, yet the former pair can exchange genetic material and the latter pair cannot. Again, any criteria you come up with is likely to be arbitrary and subjective.
As for the Wikipedia reference, I think it's ironic that there several explanations of the causes of speciation, yet no real definition...
Evolution is a scientific process, but speciation is a democratic process...
Actually, science is a democratic process, too. General acceptance is necessary. One hopes that this process is objective just as one hopes that the democratic process is objective. In practice, however, it is lacking. On a positive note, science needs to be useful. Therefore we are much less likely to accept a sub-optimal proposal in science than in politics.
I think the definition for speciation is objective. From the Wiki article: "When the populations come back into contact, they have evolved such that they are reproductively isolated and are no longer capable of exchanging genes."
Again, there may be corner cases where this is violated, the horse/mule hybrid, for example, but this definition works for an extremely high percentage of species.
You wrote:
Again, there may be corner cases where this is violated...
That's the second time you've used this expression, but I have to point out that we don't know just how big the 'corner' is. It is estimated that there are on the order of 15 million living species' and we have, thus far, taxonomized only about 1.5 million. Of these, we have identified thousands of hybrids.
In fact, since you seem to put some credence in Wikipedia, so I refer you to this statement:
Hybridisation between two closely related species is actually a common occurrence in nature.
See:
http://en.wikipedia.org/wiki/Hybrid
And this refers merely the ones we know about.
We have no reliable, scientific measure of which 'species' are no longer capable of exchanging genes. Perhaps the dog and the otter are the real "corner cases"... in the grand scheme of things.
One doesn't need to test all NXN combinations. Statistics gives us good enough tools to validate this metric. An event that occurs .1% of the time (if that high) seems like a good candidate to be called a corner cases.
I think this is an example of a "Red Herring". The application of speciation is not the issue, but the fact that organisms changed substantially enough such that the event could be correctly called speciation is. Just because the term speciation has been mis-applied in the past and may be mis-applied in the future does not mean that it was mis-applied in the observed cases.
Post a Comment