« Movie Review: Pirates of the Caribbean 3 | Main | Book Review: Death in Holy Orders »

June 02, 2007

Comments

Nato

Unfortunately this experiment has already been done, but if comparative genomics had found no (non-functional) relationship between organisms following the patterns of common descent, then most of evolution as we know it would have been utterly overthrown. Certain portions would remain, but everything else would be up in the air. I can think of no other "single" eventuality that would obliterate so much certainty.

But of course biology is not like physics, and one cannot boil it neatly into a single system of equations. In physics you just have to present a single result that contradicts an equation and it is overthrown. Almost no other science is so clean, and certainly medicine is not. What single experimental outcome would demolish the hypothesis that tobacco smoke causes lung cancer?

But then, perhaps that is an unscientific hypothesis.

David Alexander

There is nothing that can falsify evolution because evolution is a Fact, capital F.

It is impossible that a Designer could be found through science defined as methodological materialism because a Desiger is exactly what is excluded in the question, at least the kind of Designer described in Judaism and Christiantiy. This seems to me a sophomoric point but one constantly missed. When people like Bertrand Russell ask where's the proof one is tempted to ask, what did you expect. You excluded Him in the question. "You cannot find God in the answer if you exclude Him in the question." And the question why there is something and not nothing is excluded as a form of sickness by a number of scientific materialists I have heard, such as Freud and more recent less known lecturing professors. The answer is already an article of faith for many scientific materialists.

Evolutionary theory originally was constructed with a view of the cell as being a homogeneous glob. That biology is more complex than Darwin imagined it is true but that laws of heredity which were were known for thousands of years were clarified further is true. If genetics were totally disparate I suppose that would falsify evolution but there would seemingly be no reason to anticipate that they would be disparate like that though how wonderful the genome is none had imagined. But in any case, that was something that was looked back on in hindsight and so operationally evolution has worked on a principle of unfalsifiability. Also, the genome at least equally supports the the notion of a Designer. If the Designer is anything at all like humans, we know that authors and other kinds of human geniuses often produce variations on themes that can be dsicerned in their whole body of work in looking back. Also, animals need to have genetic similarity in order that they can be consumed.

Nato

"It is impossible that a Designer could be found through science defined as methodological materialism because a Desiger is exactly what is excluded in the question, at least the kind of Designer described in Judaism and Christiantiy"

I'm sympathetic to this argument, though I don't think this is a bad thing so much as an aspect of science one must keep in mind when evaluating the philosophical implications of its findings. The fact that a car cannot fly is not a problem as long as, when piloting one, one keeps that fundamental inability in mind, and its limitation to the ground shouldn't invalidate it. I know there remains considerable disagreement on the details between David and myself, but I figured this warranted saying.

"Also, the genome at least equally supports the the notion of a Designer. If the Designer is anything at all like humans, we know that authors and other kinds of human geniuses often produce variations on themes that can be dsicerned in their whole body of work in looking back. Also, animals need to have genetic similarity in order that they can be consumed."

I do *not* sympathize with this argument, since none of this would predict that non-functional DNA would have similarities. Even in functional DNA, it's hard to see how a lot of the changes are more understandable in terms of common ancestry than common function. That said, we have nothing like a perfect understanding of genotype-phenotype links, so it's plausible (though, I maintain, highly unlikely) that eventually we'll find functions for all this stuff that show that the common functional characteristics of taxons require the common genetic structures. If we do find that, then comparative genomics' confirmation of Evolution will be invalidated.

"But in any case, that was something that was looked back on in hindsight and so operationally evolution has worked on a principle of unfalsifiability"

It's hindsight now, but it wasn't thirty years ago. Comparative genomics was invented largely as a test of evolutionary hypothesis. It's true that evolutionary biologists fully expected it to turn out pretty much the way it did, but that doesn't mean that it was a given they claimed for confirmation ex post facto.

Nathan Smith

"If comparative genomics had found no (non-functional) relationship between organisms following the patterns of common descent, then most of evolution as we know it would have been utterly overthrown..."

I don't see why. Non-functional DNA changes due to mutation. If this process is rather fast relative to the geological time scales on which evolution is supposed to have occurred, it might be that after a few hundred or a few thousand generations, the similarities between the non-functional DNA of related organisms would be very slight, so that it would be hard to tell, merely on that basis, whether animals had diverged fifty thousand or fifty million years ago.

It's like shuffling a pack of cards. How many shuffles does it take before it becomes impossible to recognize that the before and after card orderings are linked? If you shuffle the deck seven times is it random? Fifty times? Five hundred times? If comparative genomics revealed that seemingly related organisms had quite different non-functional DNA, that would just show the non-functional DNA shuffled relatively fast. (Non-functional DNA doesn't, of course, as natural selection ensures.)

Nathan Smith

David Alexander's argument about comparative genomics being consistent with a Designer-- you need similar genetic structures so that animals can be consumed-- is very interesting. It is, of course, made from a total different point of view: it assumes a creator and and asks, "Why did he do it this way?" It's difficult to bridge the gap between these two very different ways of arguing the point.

Nato

Wait, I thought of another possible disproof: Say we find life on planets orbiting other stars and they show the same non-functional (and other) gene patterns as our own. This would be prima-facie evidence that the commonalities we see are *not* due to descent.

Nato

"Non-functional DNA changes due to mutation. If this process is rather fast relative to the geological time scales on which evolution is supposed to have occurred, it might be that after a few hundred or a few thousand generations, the similarities between the non-functional DNA of related organisms would be very slight"

This case would be immediately obvious, however, and would only be an available 'save' if measured mutation/drift in extant populations would support the "fast shuffle" hypothesis. This would invalidate the experimen, thus invalidating the destructive conclusions.

Nathan Smith

I think Nato underestimates the imaginativeness of evolutionists...

In any case, in any theory about life that accepts the validity of the standard taxonomy of plants and animals etc., the similarity of non-functional DNA among related organisms would be unsurprising.

Thomas

And why would it be unsurprising? Does taxonomy predict anything about DNA?

Nathan Smith

Taxonomy came along long before DNA was discovered. But certainly, knowing what DNA is, you'd assume that functional DNA would be similar. Why not non-functional DNA? It's pretty intuitive, don't you think?

Nato

One can *argue* that functional DNA would be similar within a taxon because the similaries on which the taxonomic delineation depends require similar DNA. This is actually only partially true, since there are plenty of ways of accomplishing the same exact function in biology, yet the execution remains similar amongst related animals even at the microbiological level. I just use non-functional DNA to obviate that argument.

Which leaves me mystified as to why anyone would expect animals grouped based on functional characteristics to have similar non-functional characteristics. That's bootless speculation, not extrapolative prediction.

Nathan Smith

Organism A and Organism B are similar as phenomena. We then discover that the basis for this similarity lies in the DNA-- but not all of the DNA. Some parts of the DNA are functional and some are non-functional; the functional parts are similar. Why not expect the non-functional parts to be similar? I don't know whether this is "bootless" speculation since I don't know what that means. It *is* extrapolative prediction, of course, whether valid or not. We're talking about a very basic process here: seeing a pattern, expecting that pattern to continue even where we can't see it. We see the sun rise every morning: we assume that it rose every morning before we will born, will rise in the future we haven't seen yet, and continues to rise every morning (on the surface) if we are in a prison far below the earth. It certainly is "extrapolative prediction," if you want to put it that way, but a more basic term will suffice: it's called induction.

Suppose you are shown two pieces of paper through a lion-shaped stencil. On each piece of paper, you can see that portion of the paper that is revealed by the lion-shaped hold in the stencil: the rest of the paper is hidden, but you are told and allowed to feel with your hand that each paper is a standard 8.5-by-11 sheet of paper. You observe that in the part of the paper revealed by the lion-shaped stencil, the pattern of colors is almost exactly the same, with only slight details differing. Isn't it natural to guess that probably the entire sheets have similar color patterns? Of course, it might not be so. Maybe the two sheets have similarly-colored lion shapes against totally different backgrounds. But for the sheets to be similarly colored would be a reasonable expectation.

Nato is so enthralled with his grand-theory-of-everything that he's blinded himself to this kind of simple induction. But hey, why not? Who wants to rely on poor, fallible induction when you can apply deduction, starting from the premise of a grand-theory-of-everything?

Nato

In the lion-shaped stencil example, the two pieces of paper have an apparently common source - the researchers. Even if the researchers told us they had picked them at random from the street, we would still have reason to suppose the similar patterns owed to common origins (however obscured), and thus the rest of the page would be likely to be the same.

Say researchers conducted a similar experiment with plates onto which ball-bearings had been dropped from great height, leaving marks. The researchers looked for areas to frame that looked significantly similar before presenting them to us. Would we make the same prediction about the pattern of impact on the hidden portions? I might if I suspected that the researchers' method was insufficiently randomizing, but overall I wouldn't.

We expect similarities in cases where we presume or suspect common organizing principles, not where distal facts happen to be the same. We don't expect to find the same or similar distributions of change stuck in the seat cushions of the same model of couch across the country except insofar as the shape of the couch would tend to make it the case. If we found three such couches with a yellow stain in the same spot, we'd naturally conclude that the stain occurred at the factory, and that we would find the same stain on other such couches. Without an expectation of common origin, however, we'd have no reason to hypothesize a generalized mustard stain.

Researchers chase after regularities so as to find the underlying organizing principles. Hypothesizing that some variable will vary with another with which we know of only an incidental connection is no sort of theoretical prediction. Heck taxonomy by itself is just classification, and I don't see how there can even *be* a fact of the matter* short of some organizing principle that goes beyond taxonomy itself. Why should taxonomy care about junk DNA at all? We don't classify mammals by how many hairs they have and we don't have to classify based on how similar their junk DNA is.

*And if there's no underlying right and wrong, there are no predictions to be made. I suppose one could say it's more efficient or helpful to classify based on one set of characteristics than another, and one could hypothesize in some way that some system of classification would be most parsimonious even as more features of an organism are revealed, but this is still pretty bootless if its founded on no proposed effective organizational principles.

Nathan Smith

Nato gets to the heart of the matter...

"Heck taxonomy by itself is just classification, and I don't see how there can even *be* a fact of the matter* short of some organizing principle that goes beyond taxonomy itself. Why should taxonomy care about junk DNA at all? We don't classify mammals by how many hairs they have and we don't have to classify based on how similar their junk DNA is."

Taxonomy was around before evolution, but what did it mean? Taxonomists love evolution because they get to say that the biological "families" are literally families. And they have a reason to privilege their judgments above popular alternatives: they can "prove" that a whale isn't-- *really* isn't-- a fish. This is the way to interpret the "evidence" of comparative genomics: its evidence of the validity of taxonomy, but since there's something ontologically odd about taxonomy without the evolutionary account behind it, taxonomy has gotten bound up with evolution to the point where evidence for taxonomy is taken as evidence for taxonomy-cum-evolution.

Nato suggests that we'll make assumptions about the similarity of the pieces of paper by guessing the story about how the pieces of paper came to be there. He postulates researchers whom I never even mentioned. Of course if we know the story, so much the better. But we can, and regularly do, and indeed must, infer from patterns without knowing, or perhaps even trying to guess, the stories behind the patterns. "The two pieces of paper seem to be very similar" is more fundamental than "The two pieces of paper must be very similar because..."

Nato

"...we can, and regularly do, and indeed must, infer from patterns without knowing, or perhaps even trying to guess, the stories behind the patterns. "The two pieces of paper seem to be very similar" is more fundamental than "The two pieces of paper must be very similar because...""

All true, but it's still speculation, not prediction (in the scientific sense).

Val Larsen

Nathaniel, what is your stance with respect to similarities in non-functional DNA? Do you have a hypothesis on why the similarities exist, e.g., the various animals have, in God, a common creator? Are you simply incurious about the phenomenon? It strikes me that one should generally have a working hypothesis on such matters so that normal science can proceed and either develop support for this or that working hypothesis or falsify them. If we are merely incurious about patterns we notice, we can never hope to have deeper understanding. It is rather in the nature of things that the working hypothesis in this case would propose a causal mechanism for how the non-functional DNA came to be similar. Of course, I am using "normal science" in a broader sense than Kuhn since I would want the term to cover (would not rule out ex ante as most scientists would) systematic investigation of the hypothesis that the underlying cause for commonalities in non-functional DNA might be special creation by God--though, of course, showing that special creation or the seeding of Earth with preformed life was the best explanation wouldn't necessarily mean God had done it. Perhaps all one could hope to show would be that at certain points, complex forms of life suddenly appear without requisite time spans or intermidiate forms which could make evolution a possible explanation. IF this supposition were to pan out, it would probably mean that one reaches a dead end in giving a causal account of the reality around us. But as I undertand, the Big Bang sometimes plays this dead end role in traditional materialist science, so perhaps dead ends are unavoidable. If so, we should at least be able to give a plausible account of why our explanations can go just this far an no further. Anyway, I am curious as to what your favored working hypothesis is on the DNA similarities.

Nathan Smith

"Nathaniel, what is your stance with respect to similarities in non-functional DNA? Do you have a hypothesis on why the similarities exist, e.g., the various animals have, in God, a common creator? Are you simply incurious about the phenomenon? It strikes me that one should generally have a working hypothesis on such matters so that normal science can proceed and either develop support for this or that working hypothesis or falsify them."

I don't think I'm *exactly* incurious: if there were adequate evidence in nature or experimental opportunity to rigorously test hypotheses about this, I would be happy to find out the reason. But I don't think there is adequate evidence. My basic response to this question would be: "Whatever is the explanation for similarities in functional DNA, probably the similarities in non-functional DNA have the same explanation; but both are buried in the mysterious past, and occur far too slow for experiment to throw light, so we can probably never know."

If you want a working hypothesis the evolutionist account is surely it. But I'm not sure it's always better to have a working hypothesis. If the evidence isn't there you can only fill in the gap with speculation and myth.

I don't understand the distinction Nato is trying to draw between speculation and prediction, but for my part I think he has it backwards. To reason that "pattern X holds for the data observed so far... therefore I expect it to hold for the data not yet observed..." this seems like the paradigm case of scientific prediction. Spinning elaborate, largely untestable stories abaout why the pattern should exist-- *that's* "speculation."

Nato

""pattern X holds for the data observed so far...""

The pattern is the theory. Before we could sequence DNA, there was no pattern to observe qua DNA series. That we had "junk" DNA was clear, but we couldn't yet compare its contents, yet evolutionary theory suggested the hypothesis that the DNA was vestigial. That's a prediction, based on evolutionary processes (patterns).

What would pure taxonomy predict? "These arbitrary animals that we grouped together will also happen to be similar in their non-coding DNA" - why? What pattern is this following? The pattern of human taxonomists grouping them together? This is pure speculation because it's not based on any principle or non-arbitrary pattern. If turns out to be untrue, it doesn't disconfirm anything except the hypothesis. The taxon is not invalidated.

Nato

"...evolutionary theory suggested the hypothesis that..."

Really, "...evolutionary theory pretty much required that.." The reason it's only "pretty much" is because if it had turned out otherwise, evolutionists could have still held out for a mysterious function for all that non-coding DNA that was powerful enough for evolutionary pressure to erase inter-species similarities. One would have to be a true dogmatist to hold out for more than a couple dozen genera-worth of unrelated junk DNA results, though.

Mark

Evolution has a mountain of evidence supporting it. Aside from the considerable second-hand evidence we have, people have observed evolution happening in multiple different short-lived organisms. In scientific circles at least, this debate ended long ago.

If you're going to ignore pretty much the entirety of modern biology, you might as well just believe in the flying spaghetti monster and cast aside the other sciences, too.

http://cosmicvariance.com/2005/08/04/116/

Nathan Smith

It's a bit superfluous to argue with someone who makes a fool of himself by talking about flying spaghetti monsters but...

No one here is denying evolution. What I'm skeptical of is evolution-as-a-complete-account-of-the-existence-of-life. One doesn't even NEED to observe evolution to know that mutation and natural selection will cause some modifications in organisms over time. How much work this mechanism of change can do, and whether it is sufficient to explain all life on earth: that is the question. As for the "mountain of evidence" supporting evolution-as-a-total-theory-of-life, I beg to differ. For it is not, after all, even meaningful to talk about evidence for a non-falsifiable theory. Hence the debate about the falsifiability of Darwinism, which Mark has apparently not understood.

Nato

Mark, you are being tricked by His Noodly Appendages. Don't fall for it or no beer volcanoes for you!

Thomas

RAmen!

(For future reference, Mark is my old college roommate. He's currently teaching English in Taiwan.)

Mark

Well, if talking about the flying spaghetti monster makes me a fool, then I'm in good company. Richard Dawkins, one of the world's most eminent biologists, has been known to talk about the flying spaghetti monster. So have Matt and Trey- two very amusing guys. Perhaps you meant Shakespeare's Motley fool?

In any case, I'm familiar with the creationist idea that "evolution isn't falsifiable". That claim is ridiculous. Evolutionists (known as "scientists" to most people), have addressed it long ago:
http://www.talkorigins.org/indexcc/CA/CA211.html

For a more in depth explanation of each step in evolution from RNA to complex animals, I suggest Nobel Prize-winning biochemist Christian De Duve's book Vital Dust.
http://www.amazon.com/Vital-Dust-Life-Cosmic-Imperative/dp/0465090451

Just as the theory of gravity has broken down and required refinement in some extreme cases, it wouldn't be surprising if the same happened to the theory of evolution. Intelligent Design nonsense that Kern embraces, on the other hand, really isn't any more falsifiable than the flying spaghetti monster. Once you believe that a giant cosmic being who rampages around the universe, jacking with it as he pleases and has no need for science, is the most logical explanation for life on Earth, you're far beyond the limits of falsifiability.

Mark

BTW, I met you at Whizzer's, Nathaniel. I was the poor guy that got stuck with France in that game of Diplomacy. Chris Larsen back-stabbed me.

The comments to this entry are closed.

My Photo

Only use a payday cash advance as a last resort.

Categories

Blog powered by Typepad