Difference between revisions of "Talk:Falsifiability of evolution"

Revision as of 03:38, July 1, 2007

Misconceptions about the Theory of Evolution

Many people assume the "Theory of Evolution" is a single theory. It is not. Rather, it is a collection of 1,000,000+ theories, papers, experiments, observations and tests, all of which contribute to the overall theory of evolution". The Bible is similarly comprised of a collection of books which are a collection of verses.

A common false assumption made is that disproving one of the underlying theories, invalidates the entire theory of evolution. Unlike the Bible, evolutionary theory does not claim to be Inerrant. An error in a single theory does not disprove the larger body of knowledge. Falsifying the "Theory of Evolution" would require disproving thousands of supporting theories.PerpetualAngst 16:23, 20 June 2007 (EDT)

So you are agreeing that evolution per se is not falsifiable? Philip J. Rayment 09:43, 21 June 2007 (EDT)

With apologies to Mr. Rayment, I must take issue with Angst's claim that "The Bible claims to be inerrant." It makes no such claim. And even if it did claim to be inerrant, finding an error would only disprove the individual proposition of inerrancy -- it would not prove that everything in the Bible is wrong. Ungtss

Correction accepted. The Bible makes no claim to be inerrant. Instead, others make this claim of the Bible. Similarly, finding an error in the theory of evolution would only disprove the individual proposition -- it would not disprove evolutionary theory in it's entirety. PerpetualAngst

The Theory of Evolution is still falsifiable. Tens of thousands of papers and supporting theories would need to be refuted before one could claim the larger evolutionary theory has been refuted. PerpetualAngst

Okay, that's your position, but clearly it's not something that even all evolutionists agree on. And I did notice the error in claiming that the Bible claims to be inerrant, but whilst the Bible doesn't directly explicitly claim that, I believe that is a reasonable deduction from what it does claim, and furthermore it wasn't the key point that PerpetualAngst was making, so I decided to ignore that point (which is not to suggest that there was anything wrong with Ungtss responding to it). Philip J. Rayment 10:30, 21 June 2007 (EDT)

The Theory of Evolution is still falsifiable. Tens of thousands of papers and supporting theories would need to be refuted before one could claim the larger evolutionary theory has been refuted.

Could you be more specific in defining what "Theory of evolution" could only be falsified by disproving all the papers and supporting theories? Do you mean the idea that all life evolved from a single protocell? Or do you mean the idea that all life changes gradually as the result of variation and natural selection? If you mean the latter, I think you're right -- there are a lot of papers showing that life changes over time by variation and natural selection, and it would take a lot of work to disprove it. But if you mean the former, then explain what you mean: what papers are out there providing scientific evidence for universal common ancestry that would have to be disproven? How would you falsify the proposition that all life evolved from a single protocell? Ungtss 19:24, 21 June 2007 (EDT)

Without an answer to my question above, your argument fails. Creationists don't dispute the falsifiable, scientific, and obvious proposition that life changes over time by variation and natural selection. They dispute the unfalsifiable and unscientific proposition of universal common ancestry. Ungtss 08:10, 22 June 2007 (EDT)

See http://www.talkorigins.org/faqs/comdesc/section1.html for a much more detailed argument in favor of common descent.

According to Steven J Gould: "Darwin continually emphasized the difference between his two great and separate accomplishments: establishing the fact of evolution, and proposing a theory - natural selection - to explain the mechanism of evolution." - "Evolution as Fact and Theory", May 1981.

http://www.google.com/search?q=evolutionary.biology

http://www.biomedcentral.com/bmcevolbiol/ PerpetualAngst

But my friend, that wasn't the question:). I know there are arguments out there in favor of universal common ancestry. There are also arguments against it. But is the proposition falsifiable? What experiment could be performed to prove it wrong? That's the issue here. Ungtss 15:34, 22 June 2007 (EDT)

My original point was that there is no single potential falsification test for evolution, the same as there is no single falsification test for particle physics. For example, http://www.talkorigins.org/faqs/comdesc/section1.html gives a falsification test for each of its 29 predictions. There is no single test to falsify the proposition that all life evolved from a single protocell. Instead, you would need to refute the lines of evidence with lead to that conclusion. (See link).PerpetualAngst

I agree that evolutionists present "a number of lines of evidence that reasonably lead to a conclusion." But I submit that this doesn't comply with a meaningful falsificationism. When you have a number of different lines of evidence that "lead to a conclusion," you still have room for interpretation. For example, "I believe Joe murdered Jane because he was seen running away from the scene, covered in blood." A reasonable conclusion -- but not a falsifiable conclusion. And the reason is this: Somebody else might look at the same facts (Joe running away covered in blood), but interpret the facts differently, saying, "Joe was walking with Jane when she was murdered by somebody else, got covered in blood trying to do CPR, and was running for help when he was seen." The same facts can reasonably lead to either conclusion -- and that's why it's not science. Falsifiable science leaves no room for alternative explanations. The force of gravity is 32 ft/second squared. The earth revolves relative to the sun. The human heart pumps blood through blood vessels. There's no doubt. There's no interpretation. The answers are right up there for us to see and test. That's what falsifiable science is. The same problem applies to the "potential falsification" provided by the link you cited. Consider, for example, the section in the article you linked on "Common metabolism." They say that evolution predicts that "we will never find any modern species from known phyla on this Earth with a foreign, non-nucleic acid genetic material. We also make the strong prediction that all newly discovered species that belong to the known phyla will use the 'standard genetic code' or a close derivative thereof." But there are two problems with this. First, even if a phylum was found with a non-nucleic acid genetic material, the theory of evolution could be altered to say "Well, early in our evolution, life evolved two different types of genetic material by variation and natural selection." Second, a common genetic code is consistent with creationism -- a creationist could say "God just used a common genetic material in all life." In other words, falsifying that prediction would only alter, not discredit, evolution; and the facts as we observe them are consistent with both creationism and evolution. In sum, I submit that falsifiable science is not science where there are "lines of evidence that lead to a conclusion, but which still require interpretation of the evidence." Falsifiable science consists of specific propositions that can be specifically and experimentally tested. What do you think? Ungtss 13:09, 23 June 2007 (EDT)

falsifying that prediction would only alter, not discredit, evolution. Exactly my point: To discredit evolution would require refuting millions of data points. There is no single theorem which, if disproved, will invalidate the encompassing theory of evolution.PerpetualAngst

But even refuting millions of data points would only alter evolution. Find human bones in precambrian rock -- "Oh! Must not be precambrian rock, because there's a human in it!" No matter what findings are made, they can all be "fit into" evolution, because it's such a malleable concept that it cannot be held to account. "I cannot see your dragon!" "That's because it's invisible!" That's why I'm arguing that universal common ancestry is not a falsifiable proposition -- because it's too malleable to test. However, the experimental datapoints you mention ARE falsifiable science. Those datapoints can be interpretted any number of ways with respect to the origins of life, but the datapoints themselves require no interpretation. Science has nothing to say (yet) about how life originated. But it has a lot to say about how it works today. Ungtss 15:27, 23 June 2007 (EDT)

Falsifiable science leaves no room for alternative explanations. I disagree. No science is absolute, including the theory of gravity or the theory of evolution. Nor do they claim to be. Any competing theory must explain all the observations and data points, at least as well as the original theory, and in a manner which is consistent at every level. The objective of science it to reduce the margins of error as much as possible.PerpetualAngst

I think that your point of view is exactly what Popper (and earlier creationist proponents of falsifiability, like G.K. Chesteron) was arguing against. It's always possible to come up with alternative explanations that completely explain the data points as well as the original theory in a manner which is consistent at every level. "Gravity is caused by little invisible strings on every object, and a little gopher named George is pulling them down at exactly the same rate." It explains the facts. But it's not testable. So it's no better than any other nonsense we can concoct. Just because you can hypothesize an explanation doesn't mean we can TEST it. And until you can TEST an alternate explanation, it's no better than any other crazy explanation I can come up with. Ungtss 15:27, 23 June 2007 (EDT)

What is the force of gravity on the mars? How do you know?PerpetualAngst

Approx .38G at the equator. I know because if the gravity calcs were off, the spacecraft orbiting and landing on it would not have been successful. Also readings from Mars Global Surveyor. Those readings are testable. That's falsifiable science. Ungtss 17:57, 24 June 2007 (EDT)

As given, you answer is correct to within 5x10^-3G. That is what science seeks to do: make the error bars as small as possible. PerpetualAngst 17:36, 25 Jun 2007 (PDT)

Agreed. And the magic of falsifiable science is that you can move toward that improvement. But with unfalsifiable hypothesis, you can just pick one error you're emotionally attached to, and stick with it 'cause it can't be tested. Ungtss 20:56, 25 June 2007 (EDT)

Hubble Deep Field image showed light from galaxies 12 billion light years from earth. The estimation of the time it took the light to reach earth was based on the observed redshift values of galaxies in the image. Do you contend that this estimate is false because it is not falsifiable? The light was emitted 12 billion years ago. Is it wrong because it is simply an interpretation of the evidence?PerpetualAngst

No, I'm not saying it's false, I'm saying it's unfalsifiable. From a scientific perspective, it could be true and it could be false -- we don't know with scientific certainty, because we cannot test it. Ungtss 10:05, 26 June 2007 (EDT)

Actually we can create a falsification test for it. Disprove Hubbles Law, the doppler effect or the speed of light and my statement The light was emitted 12 billion years ago will be false. Falsification can be established by refuting principals used interpret the evidence. PerpetualAngst

Not necessarily. The Doppler effect (and by this I mean specifically the proposition that light emitted from receding objects is red-shifted) could be wrong and the light still 12 billion years old. The stars might still be 12 billion light-years away, even if they are not receding. Same with the speed of light. The speed of light might be twice what we think it is -- and the stars could still be 12 billion light years away (although that would be twice as many AUs). Just because the assumptions underlying a proposition are proven false does not necessarily disprove the proposition itself. Ungtss 21:25, 26 June 2007 (EDT)

Second, a common genetic code is consistent with creationism -- a creationist could say "God just used a common genetic material in all life." In other words, falsifying that prediction would only alter, not discredit, evolution; and the facts as we observe them are consistent with both creationism and evolution. In sum, I submit that falsifiable science is not science where there are "lines of evidence that lead to a conclusion, but which still require interpretation of the evidence." Falsifiable science consists of specific propositions that can be specifically and experimentally tested. Using your own definition of falsification, would you agree that creationism is not science?

I'd agree that parts of it aren't science ... Ungtss 10:05, 26 June 2007 (EDT)

Evolution is so malleable, it would accept creationism if creationism decreased the margins of error! Science rejects creationism because creationism widens, (not narrows), the error bars. Creationism tries to explain complexity by creating an even larger unexplainable complexity: (a creator or designer). Evolution explains complexity using observable facts. For Example: We know that retroviruses insert a copy of their DNA into their host's genome; if the DNA is inserted into a sperm or egg cell, the mutation will be passed to offspring. If the mutation enhances the viability of the recipient it will be passed to future generations,if it does not, the recipient will die before pro generating (also known as natural selection). The same mutations, across different species, indicate common descent. It is not just the commonality of the genetic code that supports common descent, it is also the differences. The differences in the genetic code indicate accumulated mutations over time. Common descent is not a single theory. It is a explanation of existing falsifiable theorems explaining observable facts. ...a common genetic code is consistent with creationism -- a creationist could say "God just used a common genetic material in all life." In so doing the creationist has introduced more complexity, (an unknown creator), without reducing existing complexity.

"Reducing complexity" is not a fair measure for the "scientificness" of a proposition. General relativity introduced more complexities than Newtonian physics, which introduced more complexities than Aristotle's Physics. Quantum physics introduces more complexities than ancient atomism. Modern biology introduces more complexity than the doctrine of humors. Science says, "if experimental data indicates that reality is more complicated than we thought, then reality is more complicated than we thought. same thing if the experimental data indicates simplicity. We just follow the experimental data." Just because creationism introduces more complexity to the equation does not make it "unscientific" any more than genetics makes biology "unscientific." Ungtss 10:05, 26 June 2007 (EDT)

I do not claim science "reduces complexity". I claim science reduces the margins of error. General relativity reduced the error bars provided by Newtonian physics. Creationism, by comparison, increases the Error Bars, (to infinity), by introducing new unknowable complexity, (creators). Creationism provides no new knowledge to science. Hence, creationism can make no new predictions about science. PerpetualAngst

It's important to note that I put evolutionism and creationism in the same category -- hypotheses that are not yet falsifiable. So I agree with you that creationism can make no falsifiable predictions, while applying the same conclusion to evolution. At the same time, I think both hypotheses are worthy of investigation, because they may somebody yield falsifiable results. Creationism may indeed add new knowledge to science, if we were created. Because then, common ancestry would be false. Were we created? I think so, but neither of us can prove it or disprove it with science. Yet. Hopefully someday science will get there. Ungtss 21:25, 26 June 2007 (EDT)

the theory of evolution could be altered to say "Well, early in our evolution, life evolved two different types of genetic material by variation and natural selection." Absolutely. Guess what? You just falsified an evolutionary biology theory supporting common descent! Because science can not deal in absolutes, (with the exception of mathematics) , it is forced to make self corrections. Disproving a theorem is just as valuable to science as proving a theorem. Richard Feynman said it best: "If you thought that science was certain — well, that is just an error on your part."

But the real point is that the point of contention, common descent itself remains unfalsifiable. Maybe someday it will be falsifiable. Then we can either call it science, or call it fiction. But for the moment, it remains steeped in the interpretation of evidence, which reasonable people can interpret differently. In my opinion, you may call it "obvious," but you cannot call it "scientific," because you cannot test it. Ungtss 10:05, 26 June 2007 (EDT)

One falsification test for common descent: find an example of an organism that fails to conform to the unified genetic code by using some method, other than DNA, to store genetic information. Common descent can be falsified by disproving its predictions. So far, every new species tested verifies a unified genetic code. This testing supports common descent because it continually reinforces the prediction of common descent. PerpetualAngst

That wouldn't falsify common descent. As I said above, it would only modify it to either "one form of genetic material evolved to the other," "both evolved separately," or "both diverged." Common descent might still be true. Only "All genetic information is contained in nucleic acid" would have been falsified. I can make the same sort of pseudo-falsification with creationism. "Finding the missing link between land mammals and whales will falsify the proposition that they were created separate." You find a fossil with characteristics of both and I say either "They're not related" or "Those two are related, but they're not related to insects." You haven't falsified creationism -- you've just modified it.

Copyright laws prevent me from including the text here, but I would urge you to read http://www.talkorigins.org/faqs/comdesc/ It presents 30 evidences supporting common descent including falsification tests! PerpetualAngst

I've read them -- but in my opinion, they do not provide experiments to test the point of contention -- common ancestry. They provide hypothetical findings that could alter evolution, and which are consistent with creation. That's not falsificationism in any meaningful sense. Ungtss 10:05, 26 June 2007 (EDT)

To give but one falsification test for common descent, show that a single-celled organism can not evolve into multi-cellular form. Common descent requires this to be true for evolution to occur. Proving it would refute common descent. PerpetualAngst

It's virtually impossible to prove something impossible -- one can only point to the absence of evidence that it did happen, and the myriad difficulties it would have to overcome. (e.g. prove it's impossible to fly at the speed of light -- it might be impossible now, but that might only be because we don't know how). A better example, I think, would be the evolution of sexual reproduction. Which came first, the male or the female? The egg or the sperm? One without the other is useless. Both had to appear at the same time. And both are enormously complex, even in their simplest forms. Nobody's provided any plausible explanation as to how it happened. All the efforts to provide such an explanation strike me as absurd. Does that mean it's impossible? Naw. It might have happened -- we just might not know how. But please don't call it scientific, because it's not. It's bald speculation, devoid of experimental or observational support. Ungtss 21:25, 26 June 2007 (EDT)

PerpetualAngst closing remarks:

The problem is not that common descent is potentially falsifiable, the problem is that the level of proof requested in this forum is impossible meet.

The first clue came with the discussion of the falsifiability of the age of galaxies in the Hubble Deep Space photograph. After reading irrefutable falsification criteria, Ungst made the argument that if two of the criteria were falsified, the proposition would still be theoretically possible. This level of falsification far beyond normal levels of proof. For the theoretical possibility to be true, two alternative theories would have to be proved.

When presented with the potential falsification of unified genetic code, Ungst argues that is invalid because evolution will adapt it's theories to accommodate the falsification. the theory of evolution could be altered to say... Falsification results invalidate the original theory. Arguing that an alternative theory might be the presented to explain the data points, does not negate the original falsification. That a new theory might be proposed is completely irrelevant.

Again, when faced with a link containing 29+ falsification tests, it was summarily dismissed because they do not provide experiments. Not all proofs are performed in a laboratory. For example, fossil evidence.

Again, when presented with the single-cell the multiple-cell falsification proof, which would unequivocally refute common descent, it was dismissed as virtually impossible to prove.

Why summarily dismiss perfectly valid proofs? Ungst answers this question with two statements. First, It's important to note that I put evolutionism and creationism in the same category -- hypotheses that are not yet falsifiable. Second, I agree with you that creationism can make no falsifiable predictions Given that creationism is not falsifiable, it removes itself from the realm of science. The only way left for creationism to compete with evolution is to try to also remove common descent from the realm of science by claiming there are no valid falsification arguments for evolution.

As demonstrated, there are multiple valid potential falsification proofs for common descent.

Quod erat demonstrandum PerpetualAngst 10:00, 27 Jun 2007 (PDT)
p.s. Which came first the egg or the sperm?

The first clue came with the discussion of the falsifiability of the age of galaxies in the Hubble Deep Space photograph. After reading irrefutable falsification criteria, Ungst made the argument that if two of the criteria were falsified, the proposition would still be theoretically possible. This level of falsification far beyond normal levels of proof. For the theoretical possibility to be true, two alternative theories would have to be proved.

You're stating without showing that the falsification criteria are irrefutable. I believe I have refuted them. Let me reiterate: Even if light didn't redshit, the stars we observe might still be 12 billion light years away -- that would just be a lot farther in kilometers. Why? Because our measurement of distant stars has nothing to do with redshift. You failed to address my point. Similarly, alternative forms of genetic information were found to exist, both common descent and creationism might be true. Evolution could be true and either both forms arose separately, one arose form the other, or both diverged. Creationism could be true and either both forms were created separately, or one descended from the other. The proposed falsifications would not falsify the core propositions. Ungtss 16:23, 27 June 2007 (EDT)

When presented with the potential falsification of unified genetic code, Ungst argues that is invalid because evolution will adapt it's theories to accommodate the falsification. the theory of evolution could be altered to say... Falsification results invalidate the original theory. Arguing that an alternative theory might be the presented to explain the data points, does not negate the original falsification. That a new theory might be proposed is completely irrelevant.

The original theory was not falsified. The original theory was "common descent." A different proposition, "All life shares a common genetic structure," was falsified and replaced with "All life does not share a common genetic structure." The original theory, "common descent" was not touched. Ungtss 16:23, 27 June 2007 (EDT)

Again, when faced with a link containing 29+ falsification tests, it was summarily dismissed because they do not provide experiments. Not all proofs are performed in a laboratory. For example, fossil evidence.

First, I did not dismiss them that way. Second, fossil evidence is not "proof" -- it is evidence that can be interpretted in a number of different ways, consistent with both evolution and creation. You haven't grasped falsification, my friend. Falsification involves testing. You can't "test" fossil evidence to determine whether it evolved or was created. You can look at it, and decide what you think reasonably explains the facts. Ungtss 16:23, 27 June 2007 (EDT)

Again, when presented with the single-cell the multiple-cell falsification proof, which would unequivocally refute common descent, it was dismissed as virtually impossible to prove.

You are consistently misrepresenting what I am writing. I did not say it was virtually impossible to prove -- I said "it is virtually impossible to prove something is impossible, because someone can always say "Maybe we just don't know how yet!" I provided you with an example. You didn't respond. Ungtss 16:23, 27 June 2007 (EDT)

Why summarily dismiss perfectly valid proofs? Ungst answers this question with two statements. First, It's important to note that I put evolutionism and creationism in the same category -- hypotheses that are not yet falsifiable. Second, I agree with you that creationism can make no falsifiable predictions Given that creationism is not falsifiable, it removes itself from the realm of science. The only way left for creationism to compete with evolution is to try to also remove common descent from the realm of science by claiming there are no valid falsification arguments for evolution.

You are stating, without showing, that your "proofs" are perfectly valid. I am telling you they are not valid, because they do not meet the criteria of falsificationism. I have explained why above. You are ignoring my arguments. Ungtss 16:23, 27 June 2007 (EDT)

p.s. Which came first the egg or the sperm?

This doesn't answer the question. Parthenogenic species have the capacity to produce both sexually and asexually. The question of how they developed the capacity for sexual reproduction remains unanswered. Ungtss 16:23, 27 June 2007 (EDT)

As demonstrated, there are multiple valid potential falsification proofs for common descent. Quod erat demonstrandum PerpetualAngst 10:00, 27 Jun 2007 (PDT)

Not yet. But a valiant effort. Ungtss 16:23, 27 June 2007 (EDT)

Evolution Is False

For the sake of argument, assume that all the theories contained by evolution are false. What now? PerpetualAngst 17:36, 25 Jun 2007 (PDT)

Your hypo is too theoretical to be useful. Nobody's arguing that it is "totally false." Evolution contains a lot of theories that are obviously true. Variation and natural selection is demonstrable and obvious. If every theory contained by evolution were false, we'd have to reexamine everything we know about life. But nobody's arguing that "everything contained by evolution is false." Creationists are arguing that some of the things are true, some things are false, and some things are unfalsifiable. "Evolution" is too broad a label. We need to be more precise in our analysis. The key question to me is, "What if common ancestry is false?" Ungtss 20:55, 25 June 2007 (EDT)

The entire article is about how evolution could be proved wrong (which it can't), however the argument is never turned around. Creationism is not falsifiable. If you are to debate a subject, you must give BOTH sides of the argument. —The preceding unsigned comment was added by DrCody (talk)

No, this article is not about how evolution can be proved wrong; it's about how it can't be proved wrong (as you agree), i.e. how it is unfalsifiable.

And you are incorrect to claim that the argument is never turned around. On the contrary, the argument is usually how creation is unfalsifiable, but rarely is that argument turned around to see if evolution itself is also unfalsifiable.

I agree that the creation model itself is also unfalsifiable in principle (although specific claims about it are falsifiable). So creation and evolution are therefore both in the same boat. Agreed?

Philip J. Rayment 23:38, 30 June 2007 (EDT)