Testing Theories of Origins

Darwin is long since dead and buried. Exactly what do you mean by this statement?

Darwin doesn't accept Darwin - how interesting.But never mind, I understand what you meant. So reading your other posts about Dawkins I now see that you think he continues to accept LUCA because he is maintaining Darwinian orthodoxy rather than basing his views upon the evidence.What Venter said that Dawkins took issue with was this: When Paul Davies, who was also at the conference, responded that we have the same genetic code Venter went on: Dawkins later says: Venter has an interesting opinion and even seems to offer some evidence in support, but if you look up the Mycoplasmas over at Wikipedia you won't find anything supporting Venter's claim that they "use a different genetic code" and I think most biologists would share Davies and Dawkins skepticism at this point in time. I also think that most biologists would be terribly interested and excited if it were ever discovered that there are actually multiple trees of life. Imagine the opportunities to expand our knowledge that would develop out of the discovery that life originated not once but at least twice.You can find the above cut-n-pastes here: Venter vs. Dawkins on the Tree of Life -- and Another Dawkins WhopperAnd you can find the video here: The Greate Debate - What is Life?And you can find the request that you stop throwing out unsupported claims and accusations faster than people can catch up with them in this sentence.--Percy

While I disagree with Dr. Adequate a bit about terminology, I submit that there is no difference at all between prediction and explanation in a properly formulated theory, and that this distinction of factors is total crap.Let's consider general relativity's ability to predict the entire amount of the precession of the orbit of mercury.In simple terms, the general theory of relativity is a consequence of equivalence principle, but it is a consequence that was extremely difficult to work out because of the mathematics involved. However Einstein managed to master and use the mathematics to logically derive the theory of general relativity essentially from the geometric considerations surrounding the equivalence principle.The only calibration of the theory involved was that the general theory of relativity had to reduce to Newton's law of universal gravitation in the domain of slow velocities and weak gravitation fields. Note though that Newton's theory was incapable of accounting for the orbit of mercury. This calibration could not have introduced the advancing phenomena into the general theory.In other words, the fact of that Mercury's orbit precessed about 43 seconds of arc per century more than predicted by Newton's theory of gravitation was not used to construct Einstein's theory. Thus, the fact that 43 arc second per second value fell out of Einstein's equation was no less impressive than if the value had been measured later. It is of historical interest only that the extra 43 seconds of arc was already known.Yes it is true that the Eddington confirmation of the amount of bending of star light by the sun coming after the fact of Einstein's calculation made Einstein publically famous in non- scientific circles, but that timing is irrelevant to a determination of whether general relativity is actually valid.The order of confirmation provides some evidence that the theory has not been manipulated to provide a confirming result of existing data, but we could also have determined that Einstein had done no such manipulation by reviewing his work. Despite the claims made on Conservapedia, it is beyond question that Einstein did not perform such manipulations.So this "future prediction" stuff is a possible heuristic test, but it clearly is not correct in all cases, so it is not a main test. The real test would concern the independence of the prediction from the observations and not whether the prediction followed or preceded the observations in time. Also important is having a solid enough formulation such that calculations of reality can actually be made at all.Edited by NoNukes, : No reason given.Edited by NoNukes, : No reason given.

Hi NoNukes,I have to disagree with your interpretation of prediction. By your definition string theory is making successful predictions every time it comes up with answers consistent with the standard model.Wikipedia's introductory paragraph on Scientific Theories seems pretty clear about the separate roles of explanation and prediction: --Percy

I don't think I've advanced such a definition.If I have suggested anything other than that predictions must match reality then I was wrong. But I don't believe I've said anything about matching one theory to another theory rather than to reality as being a test.I did speak of Newton's law of gravitation. Perhaps that is a point of confusion. First, Newton's law of gravitation is indistinguishable from reality where low speeds and weak gravitational fields are involved. But I did not cite Newton as a prediction. Instead I spoke of the technique of assuring that general relativity matched Newtonian gravity in the domain in which Newton was correct in the process of developing Einstein's theory and not in verifying it.And I spoke of Newton only to point out the danger of incorporating a phenomenon into a theory and then claiming that the theory predicted the phenomenon. That would not be a true prediction. It's also not an error that Einstein made.I'll also note that at in Einstein's day it was impossible to measure the anomalous advance of the perhileon of Venus and Earth because the orbits involved have very low eccentricities, and because the anomaly is much smaller. But I reject the idea that there is any qualitative difference in verifying general relativity against those values and that of Mercury which was known prior to Einstein's work.If that does not address your point, then I don't see what your point is. I agree that falsifiability is a proper requirement for a theory, but that is not one of designtheorist's tests. I did not discuss it but I certainly don't reject it. We should add falsifiability to the list of real tests.Edited by NoNukes, : No reason given.Edited by NoNukes, : No reason given.

To get us started, here are some links.Explanatory PowerPredictive PowerPerhaps these are not great articles but they get us started. I think a theory or model has the greatest explanatory power when it can explain the greatest number of important facts from a variety of scientific disciplines and shows the greatest number of causal relationships.Dr. Ross writes:"Predictions must be detailed, distinctive and comprehensive to be of any use in evaluating a particular model. When predictions are so vague that the proponent of a particular model runs no risk of being wrong, they are virtually worthless. P. 233Designing predictions to show a difference with respect to competing models permits comparisons. Predictions unique to one model and contrary to all other models hold the greatest promise for furthering understanding of specific creation/evolution issues. P. 233Finally, a set of predictions must be comprehensive enough to address all (or nearly all) the major relevant issues. While no model can hope to explain everything (human knowledge will always remain finite), a good creation/evolution model needs to provide explanations for already observed relevant phenomena. As such, the model should produce predictions about what researchers will discover as they continue to study the broad array of creation/evolution disciplines. P. 233.

Oops. I meant Dawkins does not accept the recent rejection of Darwin's LUCA and the tree of life.Thank you for finding the video clip for me.Koonin has written more than one article on the topic. His view is catching on because it has good science behind it. Yes, there are some who have not accepted the new view yet but most people are not even aware of it.

No, no, no, not what I meant, my fault. By "standard model" I only meant what we already know. When I said "consistent with the standard model" I only meant "consistent with what we already know."Einstein's theory of general relativity had explanatory power because it was consistent with what we already knew, for example, about the precession of Mercury. And it had predictive power because it predicted things we didn't know, such as how much gravity would bend light, and how much rotating mass would bend space-time.--Percy

Dawkins' skepticism of Venter at the conference seems fully justified. Is there something Dawkins has said about Koonin's work that justifies your characterization of denial?--Percy

I don't see how the precession and gravity bending predictions prove your point. I don't see a distinction between knowing the values before and after theory formulation. Being wrong about either would falsify the theory. In fact, Einstein actually made a bad prediction regarding light bending that might have gotten exposed if Eddington's pre-war expedition had gotten off the ground.However, 'how much rotating mass would bend space-time'... Yes I can see that Einstein predicted an entire phenomena that was unexpected in that case.Okay. I yield.From the link designtheorist provided.

quote:

---

Explanatory power is the ability of a theory to effectively explain the subject matter it pertains to. One theory is sometimes said to have more explanatory power than another theory about the same subject matter if it offers greater predictive power. That is, if it offers more details about what we should expect to see, and what we should not.

---

This definition seems to show a rather tight relationship between explanatory power and predictive power. Perhaps predictive power can be distinguished from explanatory power, but I simply don't see how differences in timing between phenomenon and observation make a theory more or less viable.

They don't. As a theory is being formulated, contradiction by known data is fatal. After a theory is formulated, contradiction by furute data is fatal.A simple analogy. A career in show business can be killed before it ever gets off the ground (no talent), or after it gets off the ground (scandal, perhaps). In the same way a promising theory can be killed before it gets off the ground (contradicted by known data), or after it gets off the ground (contradicted by future data).Maybe some are wondering that if contradiction by real-world data is fatal to a theory then why should it matter whether it happens before or after the theory is introduced. Why distinguish between explanatory and predictive?I guess there are two answers. First, a theory will never be introduced if it is contradicted by known data (unless the theory is being proposed by a creationist). Second, the predictive power of a theory is incredibly valuable because it leads us into realms not previously understood. It is one of the important ways that we expand our knowledge of the universe.--Percy

Well, I'd say it is.But perhaps we are spending too much time choosing our vocabulary. I think we can agree that the proper way to test a theory is the comparison of its logical consequences to the data; and that the historical order of the invention of the theory and the observation of the data is irrelevant to the goodness of the theory.Edited by Dr Adequate, : No reason given.

This is absolute nonsense - pure drivel! How many Hollywood films about 'the future' have you seen to believe that prediction is about things that are 'yet to happen'?Meanwhile on planet Earth and the Scientific Method the word 'prediction' has a completely different meaning. It means, given a data set and a particular hypothesis you can predict the outcome - not in the future but here NOW. This is what is meant by 'prediction'.To use the Theory of Evolution as a familiar well trodden example:Descent by modification of existing species (the hypothesis) makes the following PREDICTIONS:1.Organisms if descended from earlier organisms should be arranged in a 'nested hierarchy' (which is what we find)
2. Organisms should show clear signs of predecessor physiology - evolution has to build on what went before - no clean sheets - no wiping of the slate clean (again which is what we see)Neither of the two examples above are predications about what WILL happen in the near future (on the contrary they predict what should have happened in the past based on what the hypothesis can allow)But predictions within the scientific method should do much more than this. More importantly than what a hypothesis predicts WILL happen, is what it predicts CAN'T happen (falsifiability). This is crucial as there are far more things a theory can't do (given the fact there are billions of things that 'can happen' out there in this world) than what it can do.For the ToE the following are examples of falsifying predictions:1.No animals should appear in the fossil record earlier than the time it evolved (no rabbits in the Precambrian)
2. No features that appear in one specific line should suddenly 'jump' into another line (the 'correctly wired' eyes of cephalopods should not suddenly appear in the mammal line (incorrectly wired eyes).Again, the two examples above of falsifiability have nothing to do with predicting future actions but checking out what is already there.I think you have a serious misconception about the word "prediction". You seem to have linked it in an almost Pavlovian way to "what will happen in the future" and this is absolutely not the case at all.In the scientific method all 'prediction' means is "does the hypothesis make general statements about what should and shouldn't happen if this hypothesis is correct?"By the way: if you can't falsify a hypothesis then it is NFG (no fucking good). If a rocket engineer was asked what his rocket propulsion formula predicted and he said "You name it my formula does it" what he is actually saying is "it does nothing at all". An engineer (in common with scientists) need to be very specific about what something can and CANNOT do. And the 'cannot' part is even more important than the 'can' part.A dead giveaway for unscientific proposals is the lack of an ability to properly falsify a hypothesis. If you try and find you can't falsify then you seriously haven't got a scientific proposal up for discussion....try this simple test - it will sieve out a lot of nonsense before you get to try it out here.

Regarding theory, no one's saying that prediction is about things that are yet to happen. It's about things that are yet to be discovered.And no one's saying that prediction is only about what will be discovered. It can as easily be about what won't be discovered.--Percy

It is a criterion which in science more often prevents blunders than corrects them, because scientists generally aren't fools. (It could be applied with profit by pseudoscientists who are fools, but being fools they're not going to.) For example, the principle of parsimony would have prevented a chemist from going around saying: "I've discovered a new element with exactly the same properties as tungsten" --- but it would, in fact, have prevented him from doing so; the situation would never have developed so far that other people were obliged to use the principle of parsimony as a critique of his claim.