Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Predictions from evolution

  1. Aug 28, 2010 #1


    User Avatar
    Gold Member

    I'm currently reading a book that presents the different kinds of evidence for evolution. Throughout the book, the author claims that evolution can make certain testable predictions. For example, he writes about biologists who examined birds and reptiles and predicted that they'd find a common ancestor in a certain level of rock. When they looked for fossils in that level they actually found the common ancestor.

    Now, since the fossil record is rather sparse (less than 0.1 percent of living animals are fossilized), there are many common ancestors that we cant find. So even when biologists don't find the proposed ancestor, they can always say that this doesn't disprove the theory because the fossil record is missing so many evolutionary steps anyway. With this in mind, why are the fulfilled predictions of evolution more impressive that the predictions of a false prophet? If only a small percentage of evolution's predictions come true, and there's a ready explanation for all the times that they don't (which is most of the time), then how can the true predictions be proof of anything?
  2. jcsd
  3. Aug 28, 2010 #2


    User Avatar

    Staff: Mentor

    How can finding the fossils of a common ancester prove anything? They prove that the scientist's guess about the common ancestor was correct. Sometimes they find a common ancestor that surprises them. We're finding more and more fossil information all of the time.

    You can't seriously be trying to compare scientists who are studying the past to cranks predicting the future? One is factual science, the other is not.
    Last edited: Aug 28, 2010
  4. Aug 28, 2010 #3


    User Avatar

    Staff: Mentor

    I don't think Daniel compares scientists and cranks, basically he is asking about testifiable predictions of evolution theory.
  5. Aug 28, 2010 #4


    User Avatar
    Gold Member


    Obviously, I'm not fully comparing the them, there's a lot of evidence for evolution regardless of the predictions I mentioned. I'm only comparing one aspect of the two.

    Don't you agree that in order for predictions to prove anything, they have to come true consistently? You say that the fulfilled predictions prove that the biologist's guess was correct and I agree. But the author of the book was trying to say that the fulfilled predictions add proof to the theory of evolution in general, not just to one particular common ancestor. In that sense, how is it different from the cranks
    that try to prove the fact that they have superpowers by making many predictions and highlighting the ones that happen to come true?

    If for example the crank says that the winning lottery number tomorrow will be 1111 and it turns out that he was right then that proves that his particular guess was correct. But it doesn't prove that he has supernatural powers, especially if he also makes many mistakes. Similarly, if the biologist uses evolutionary theory to predict where a certain fossil will be found, then finding the fossil shows that his "guess" was correct. But that alone doesn't show that his theory is correct, especially if not all of his predictions work out.

    I'm not debating the truth of evolution itself, but merely questioning one particular method of proof. To sum it up, I'm aware of the vast quantities of evidence in favor of evolution, but I fail to see how the fulfilled predictions add proof for anything other than the specific ancestor that was found.
  6. Aug 28, 2010 #5


    User Avatar

    Staff: Mentor

    What book are you reading?
  7. Aug 29, 2010 #6


    User Avatar
    Gold Member

    "[URL [Broken]
    Why Evolution is True[/URL] by Jerry Coyne
    Last edited by a moderator: May 4, 2017
  8. Aug 29, 2010 #7
    Your claim, daniel _i_l, is that you accept the evidence for evolution and that your question is genuine. Your problem is that your argument bears a strong resemblance to that commonly used by deniers of evolution. It is an old joke that whenever such people have been scornfully pointing out absence of a particular fossil to provide the evolutionary link between two known species, and palaeontologists then find that ‘missing link’ fossil, the naysayers then immediately point out that there are now two missing links.

    But taking your claims at face value, the truth is that your statement about how sparse the fossil record is, is not accurate. The rate of palaeontological finds has been growing steadily, and there is now a substantial body of evidence from palaeontology alone. But even more significantly, palaeontology has started to improve its work with other areas of evolutionary study. Once was the time, not so long ago, when university palaeontology departments and microbiology departments were at best oblivious to each other and at worst in open conflict. These days they work together and regularly uncover evidence that corroborates each other’s work. I know all this from a couple of books I have read, which I have already mentioned elsewhere on this forum. They are ‘Endless Forms Most Beautiful’ and ‘The Making of the Fittest’ by Sean B. Carroll. He talks very tellingly about this ‘modern synthesis’ is evolutionary biology. I think you would find his books very interesting and very stimulating. I certainly did.
  9. Aug 29, 2010 #8
    Id also recomend this book and author in general: Life Ascending: The Ten Great Inventions of Evolution by Nick Lane. As for predictions a person must be very careful about that. People can proclaimate all kinds of predictions but it dosnt mean they are on the right path. Scientist's can be as full of themselves as any proper fool. And there is no fossil that holds some place in the record that will make it the perfect piece of evidence or proof. Its just one indicator that among many will let you know that your on the right or wrong path.

    Anyways I do hope you will continue ask questions in this thread so that it can be discussed!
  10. Aug 30, 2010 #9


    User Avatar
    Homework Helper
    Gold Member

    It has been said that the only thing that has ever been rigorously proved is that nothing can be rigorously proved.

    I think one has to take with salt the simplistic model of a mill of theory -> prediction -> experiment or observation -> falsification or not.

    It is quite difficult for the many sciences where experiment is not possible. Although certain experiments will be useful to them. Fieldwork sciences. Astronomy, earth science, bioevolution, even ordinary history. They are an eclectic combination of detective work, modelling, deduction etc. Although we cannot do experiments we are to an extent saved by the great number of "natural experiments" there have been - that we are not dependent on studying our one sun but can see myriad other suns in various stages of their lives (biology has more stories, fewer examples of most stories).

    It is fairly obvious that e.g. mammals all have rather related body plans, then vertebrates a bit more distantly related. Then if you start to discover fossils which look also related, if also you can extrapolate known geological processes to get estimates of their ages, then if you are without prejudices you begin to frame certain ideas... Anyone with prejudices can object you haven't "rigorously" satisfied a criterion of scientificity (which they have invented).

    Links fill in - not always. Reasonable people will believe the fact of evolution. This does not easily prove the Darwinian mechanism of evolution, whose epistemological status has been much debated. It was to an extent almost a theorem deduced from the facts of heredity known to Darwin, now much better understood. Why one animal was less well adapted than another descended from might be difficult to say, but with greater understanding of climate and geological change and seeing the evolution together of e.g. predators and prey, a greater understanding of physiology, biomechanics etc. one can make increasing sense in terms of fitness.

    A far finer trace of evolution is contained in DNA sequences, now massively analysed. This has confirmed and much refined many of the previous conclusions, e.g. all the primates related, dogs foxes and wolves closely related, etc. but has changed some of the previously thought routes of evolution and relations of groups. It will take longer to relate DNA sequence to function, but in bits this has already been done.

    We now have this incomplete but huge body of an understanding called modern science. Belief in most of it is less risky maybe than for earlier pioneers. But anyway this existing mass is a product of an evolution. Sometimes, I think, it is more instructive to look at how it has evolved than do what you seem to accuse the evolutionists of doing, fitting it into a preconceived plan of what ought to be?
    Last edited: Aug 31, 2010
  11. Sep 1, 2010 #10


    User Avatar
    Gold Member

    I'm not "accusing" the evolutionists of anything. As I mentioned above, I wasn't trying to start a debate regarding the veracity of evolution. Because of the many types of evidence, I'm convinced that evolution happened. I'm just trying to ask a question about one fine point of the theory. For some reason some posters in this thread got all worked up about the fact that I might be casting evolutionary evidence in doubt, while in fact, I just want to have an objective and scientific discussion.

    Anyway, I probably didn't explain myself properly either so I'll reformulate my question.

    In physics, one major way to support a theory is to use it to devise an experiment that has two possible outcomes, one that's predicted by the theory and one that's not. For example, using Newtons laws of motion we can predict the time it will take for a ball to roll down a ramp up to a certain degree of measuring error. So we can measure the time and check if it falls within the error margin predicted by the theory. If the prediction was wrong then we'd know that there was a problem with the theory.

    My question is, is there an analogous experiment for the theory of evolution? In other words, is it possible to come up with some experiment such that one set of outcomes are predicted by the theory and the other ones are not? The point I was trying to make before is that the prediction "If we dig in place X then we should find fossil Y" isn't enough because neither finding the fossil nor not finding it would go against the theory.

    From what I understand, epenguin basically said that the type of experiment I asked about only exists in certain kinds of sciences and may not be applicable to evolution. Is that correct?

  12. Sep 1, 2010 #11


    User Avatar
    Gold Member

    It has been noted by many, yes, that evolution is not easily falsifiable. Falsifiability is a requirement for a good theory.
  13. Sep 1, 2010 #12
    Haha, damn man that was harsh, I don't think he is questioning evolution, if anything it is simply an interesting debate for the Philosophy of Science and the methods of human inquiry. Simply that within Physics and similar sciences we have ideas about the nature of reality that are most times pitched within the language of mathematics, thus allowing it to have a level of quantitative predictivity that can be falsified and used in a somewhat more straightforward way. The question is simply about the methodology of science as practiced in evolution theory, not so much comparing Physics and Biology and saying that Biology doesn't predict in the same way and as a result is inferior, rather pointing out that Biology has a different methodology yet it still retains a degree of effectiveness and illuminates our understanding quite a bit. In light of this, how can we discuss Evolutionary methodology? Kind of like accepting a Humean critique of Newtonian Mechanics, but simply seeing that Newtonian Mechanics "works" as knowledge and what does this say about knowledge as a whole?...then we can discuss more specific aspects of Evolutionary inquiry, how it has proceeded, and how it should proceed.
  14. Sep 1, 2010 #13
    The suggestion that evolutionary biology cannot work by experimentation is not accurate. Neither would it be fair to say that non of its assertions are verifiable. That man I’ve mentioned already, Sean B. Carroll is an experimental microbiologist, and he gives some pretty vivid accounts of some of the experiments he has been involved in. The very reason that Drosophila melanogaster (the humble fruit fly) has become such a closely studies species is because of the experimental opportunities it offers to evolutionary biologists, because of its short breeding cycle and great fecundity in favourable conditions. And much of the history of evolution is written into DNA. The accurate dating of exactly when we became the naked ape, and when we started to wear clothes, for example, is written into the DNA of lice.
  15. Sep 1, 2010 #14
    Ok, my "harsh" reply got deleted, so maybe I'll put it more gently...

    The OP asked us to compare evidence for evolution with the predictions of a "false prophet" in post #1, then in a later post says "I'm not accusing the evolutionists of anything"

    These are not the words of someone looking for a rational debate, at least not on a science forum. The term "evolutionists" sounds like something out of 19th century theological crusades against Darwin.

    There is obviously no experiment which will help "prove" evolution since evolution occurred over a few billion years. The best you can do in that respect is examine the historical evidence, fossil records and dna comparisons. But you can show that genetic mutation and selection works in fruit fly breeding experiments for example, or with plants subjected to varying environments.

    But that's pretty trivial, and hardly sounds worthy of debate.

    Ultimately we will probably discover other planets with life in various stages of evolution, which would be nice evidence but still probably wouldn't convince those who don't want to be convinced.
  16. Sep 1, 2010 #15


    User Avatar
    Gold Member

    A point of order: it is not merely naysayers of evolution who are asking this question. It is a valid and concerning question for any scientist.

    Your attempts at refutation are ad hominems. Refute the argument, not the arguer.

    The same rhing could be said about the Big Bang. That event too happened long in the past.

    Yet we have models that we can test, they can fail the tests, the models are falsifiable.

    What predictions does evolution make that we can test for falsifiability?
  17. Sep 1, 2010 #16
    Evolution predicts that complex and varied life can evolve on a planet from simpler origins.

    That happened.
  18. Sep 1, 2010 #17


    User Avatar
    Science Advisor
    Gold Member

    "Evolutionary theory" really is a unifying theory, as such its huge! There are literally hundreds of avenues one could explore on it's falsifiability.

    To name a few; homology and phylogenetics, homology in cladistics, modes of speciation, fossil record complexity, etc

    (Basically the advent of all of molecular biology has been one giant test of evolutionary biology, where time and again it could have been toppled by molecular biology).
  19. Sep 1, 2010 #18


    User Avatar
    Gold Member

    I think that this is the difference between most evidence for evolution and evidence for physical theories. Of course it's possible to come up with ways that evolution could have been falsified [the famous rabbit in the Cambrian for example], but that isn't as powerful as an experiment designed such that one of two possible outcomes is incongruous with evolutionary theory. Has such an experiment ever been done?
  20. Sep 1, 2010 #19
    Yeah it seems that evolution can only be studied in the rear view mirror. Setting up an expiriment with one right outcome dosnt take into account the way evolution takes the easiest path possible. Rewind the story of earth a hundred times and youll get one hundred different movies. But mostly youll get movies that are 4.5 billion years long and involve bacteria munching minerals in a planet wide desert. You can prove it takes place but you cant predict by what pathway it will arrive at an end point. And it might not ever reach your end point.

    This is as close as Ive ever seen and its merely an observational experiment. http://myxo.css.msu.edu/ecoli/
  21. Sep 1, 2010 #20
    In the spirit of "know thy enemy", I read http://books.google.com.au/books?id=BG9EAAAAYAAJ" (beware, most of the authors just fail to understand statistical physics, and half consciously fore-go reason). An example therein was a test of the hemoglobin of some related (I think semi-aquatic) animals. One hypothesis is that the precise details of the molecule should correlate to how closely related the different species were overall (phylogeny). The competing hypothesis was that the precise details of the molecule should correlate with how similar the habits of the animals (as though designed for the purpose). This was presented as your binary test, in which the evolutionists made one prediction and the creationists made the other. Except that in this case the outcome favoured the latter. In hindsight we of course say that neither prediction would be inconsistent with evolution. As already said, the theory isn't terribly conducive to making simple hard predictions, and really a lot of evidence needs to be weighed.
    Last edited by a moderator: Apr 25, 2017
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook