Atheism Stack Exchange Archive

Evolution as a mathematical and physical process: theory of evolution vs theory of gravity?

I often see people try to stand on “Evolution as a Mathematical Process.” This often plays into the hands of a common theological argument about math and analytical truths. This can be won, but it requires work. It goes like this:

If evolution is a certain or as fruitful as physics, why hasn’t it yielded an equation of the sort E = MC^2 ?

There are all kinds of variations of this argument, and it is most often seen when evolutionists (excuse the use of this word, it will never happen again) make the claim that theory of evolution is on par with the theory of gravity. It’s becoming rather pervasive.

If evolution and gravity are both a theory, and I’m to believe they both have an equal amount of evidence, then what can you tell me about the human species in 15,000 years? I’ll happily tell you the chances of a tracked asteroid colliding with the moon on that date, and the force the moon will exert on the moon..

I’d like to know:

Answer 1761

I find the question

If evolution is a certain or as fruitful as physics, why hasn't it yielded an equation of the sort E = MC^2 ?

to be deeply flawed (which, no doubt, is why it comes up so damned often). The correctness or utility of a theory is not determined by whether or not it gives us a simple equation. The premise is invalid and should be rejected out of hand.

Edit

The other big problem with the question as posed is that it conflates one particular sub-discipline within biology with the entirety of another branch of science. It would be better posed as “if evolution is as certain or fruitful as {quantum mechanics general relativity special relativity …}”, but even then the original objection applies.

IOW, it’s an attack based on sloppy terminology and misdirection, and again the only appropriate response is to reject the question entirely.

Edit2

I’m choosing to answer Evan’s comment here.

Here are the reasons, in order, why the question as it is currently written should be rejected:

  1. It begs the question that in order to be certain or fruitful, a scientific theory must yield a simple equation (this isn't true);
  2. It conflates a specific theory in biology (evolution) with an entire branch of science (physics); there are plenty of theories in physics (string theory, for example) that have considerably less certainty associated with them than evolution does. </ol> IOW, it's a logical fallacy wrapped in sloppy terminology. ## Answer 1862 - posted by: [Konrad Rudolph](https://stackexchange.com/users/-1/82-konrad-rudolph) on 2011-01-02 - score: 6

    If evolution is a certain or as fruitful as physics, why hasn't it yielded an equation of the sort E = MC^2 ?

    Well, what about Hardy-Weinberg?

    f(AA) = p^2
    f(Aa) = 2pq
    f(aa) = q^2

    This is from population genomics and predicts the ratio of two alleles a and A in a given diploid population.

    This is similar in its descriptiveness to Einstein’s famous formula. And even this simple formula makes a surprisingly accurate prediction – to answer another of your creationists’ questions:

    what can you tell me about the human species in 15,000 years?

    Hardy-Weinberg tells us that no matter how many male humans we kill off in future wars (or females, provided at least a few remain), the male/female ratio will still be at equilibrium in 15,000 years. This is a pretty strong prediction considering that common sense tells us that this can’t happen, and yet we can be absolutely sure that it will.

    What kind of quantification can be done for biological evolution?

    Too many to name. For further reading, look (for example) at the Wikipedia articles on quantitative genomics or evolutionary game theory.

    Even more impressive is bioinformatics which in a way is purely quantitative. For example, sequence comparison uses similarity matrices to determine the probability of certain mutations (e.g. the PAM matrix) which are based off quantitative analysis of evolutionary changes in proteins.

    What is the most powerful rebuttal to the argument that gravity is a more certain theory?

    To anyone familiar with the subject, this is obvious. To others, not so much. The problem is that gravity, in a sense, is familiar: we experience it every minute of our life. Evolution is hard to grasp because it operates on extreme scales: the microscopic scale of protein machinery and the gigantic scale of evolutionary time, which is measured in eons. The human mind is not equipped to cope well with either scale (ironically, this is due to evolutionary adaptation of the brain).

    But how about this: if evolutionary theory were less certain than gravity, modern medicine wouldn’t work. It fundamentally relies on the truth of evolution, and nothing would make sense without it. Consider something as simple as animal testing: the whole premise here is that animals are not only incidentally similar to us, but that they are similar because we have a common ancestor and that the machinery which we are testing the substances against works in the same way.

    So do paternity tests, AIDS tests, immunizations (flu shot), penicillin production, chemotherapy (cancer medication) and drug design. All of them need reliable predictions, which evolutionary theory provides.

    But for me, the most impressive prediction is this (adapted from Richard Dawkins): if we ever find life on another planet anywhere in the vastness of the Universe, we can be certain that its evolution obeys the same basic principles as here. They may not share the same information storage mechanism (e.g. DNA) but there will be some storage, there will be (lossy) copying from one generation to the other, there will be some kind of interpreter of the information and there will be an interplay between environment and “genome” – in summary, there will be mutation and selection. We know this just as certain as we know that that planet will obey the laws of gravity.

    Is it wrong to say that the theory of evolution can only ever make qualitative predictions?

    I guess that one is answered, then.

    Is it dangerous to concede the point that biology can not yet live up to the level of quantification of physics, and it never will, thus affirming it is less certain?

    It’s not just dangerous, it’s wrong. See above.

    ## Answer 1726 - posted by: [Peter](https://stackexchange.com/users/-1/168-peter) on 2010-12-30 - score: 4 Evolution has a mathematical basis for some areas of balance between behaviors. For instance, the balance between cooperators and defectors can be determined when it comes to altruistic behavior. Richard Dawkins wrote an article for Edge.org about changing his mind after a colleague produced a mathematical model explaining the concept of prestige in sexual selection. http://www.edge.org/q2008/q08_15.html#dawkins When it comes to gravity we have better evidence to support evolution than we have for gravity. We don't know where gravity comes from or what causes it. It is related to mass but we don't know if mass produces gravity or gravity produces mass. We know how evolution works and the processes that drive it are well understood. ## Answer 1714 - posted by: [John](https://stackexchange.com/users/-1/627-john) on 2010-12-30 - score: 1 - *What kind of quantification can be done for biological evolution?* Start with what you've seen evolve. Perform measurements on it. - *What is the most powerful rebuttal to the argument that gravity is a more certain theory?* I know of none. The theory of gravitation stands up to test far better has been tested far more thoroughly than evolution. - *Is it wrong to say that the theory of evolution can only ever make qualitative predictions?* It's unlikely to even make qualitative predictions. How can you know how a species will mutate? - *Is it dangerous to concede the point that biology can not yet live up to the level of quantification of physics, and it never will, thus everything is less certain?* No, it's completely reasonable to concede this. With a powerful enough computer, you can apply physical principles to explain biological ones. It's all protons, neutrons, and electrons. ## Answer 1716 - posted by: [bubu](https://stackexchange.com/users/-1/398-bubu) on 2010-12-30 - score: 1 - What kind of quantification can be done for biological evolution? the best you could do is a complete sequencing, sampling organisms introduced to new environment with time lapse between each complete sequencing. The quantification of change of gene is base-pairs. Note of course that there are changes in form of gene activation and deactivation by various mechanisms, changes like methylation do take place. In addition, you will measure the fitness (i.e. number of offsprings) of the species as a whole across generations. - What is the most powerful rebuttal to the argument that gravity is a more certain theory? The theory of gravitation, to me, is more well-tested in experimental settings and thus to most of us, is of course of a higher certainty. There is, however, nothing that is totally certain as far as deities go. The flying spaghetti monster deity may as well be using thy mighty power from thy noodly appendages to manipulate all experiments concerning gravitation so that it appears true whereas the forces binding things on earth are just telescopic power derived from the meatballs. [1] - Is it wrong to say that the theory of evolution can only ever make qualitative predictions? You can design an experiment and prove it wrong. Collect enough data and then you can make more qualitative predictions. I actually must say that we should encourage all theist to work in quantitative experiment on evolution so that while they try to disprove the theory, more knowledge is yielded. Their time is better spent on disproving (or proving - in fact, with scientific method, disproving something with experiments is basically the same thing as proving something) the theory of evolution than introducing their creationism into science class. - Is it dangerous to concede the point that biology can not yet live up to the level of quantification of physics, and it never will, thus affirming it is less certain? You can (almost) always quantify something. It just depends on technology. 100 years ago we can't even manipulate atoms. Nowadays we're working on subatomic particles. [1] google flying spaghetti monster for more on this. check http://en.wikipedia.org/wiki/Flying_Spaghetti_Monster if you are too lazy :) ## Answer 1730 - posted by: [Raskolnikov](https://stackexchange.com/users/-1/144-raskolnikov) on 2010-12-30 - score: 1 - *What kind of quantification can be done for biological evolution?* It is true that biology is harder to pour into a mathematically precise theory, but that is because it has an incredibly higher complexity than most of the topics that are treated in physics, like gravity. Nevertheless, it can be done to some extent. In fact, evolution and genetics have been tied to the development of the mathematical theory of statistics. One of the pioneers was [Ronald Fisher](http://en.wikipedia.org/wiki/Ronald_Fisher). I also invite you to view this [BBC program with Jim Al-Khalili](http://www.youtube.com/watch?v=HACkykFlIus), point in case, the discovery of Alan Turing about a mathematical model for embryology. - *What is the most powerful rebuttal to the argument that gravity is a more certain theory?* There's no question that the theory of gravitation is the most well established one, although it's hard to put a number on it. Let's say that the effects of General Relativity have been measured in several tests to astonishing accuracies and they have passed the tests with brio. There are of course problems, like singularities and such, but there are problems in evolutionary theory as well. I'm not sure the comparison is entirely fair though, since the theory of gravitation is older than evolution. GR is recent, but it's actually an improvement upon the older Newtonian theory. - *Is it wrong to say that the theory of evolution can only ever make qualitative predictions?* Yes, it is. Several quantitative predictions have been made. Predicting where you have to look for a particular fossil is in my opinion very quantitative. So are several results in genetics. Evolution also serves as an inspiration for the development of computer algorithms. So that is an example where evolution has provided tools for mathematics. - *Is it dangerous to concede the point that biology can not yet live up to the level of quantification of physics, and it never will, thus everything is less certain?* I'm happy to concede that biology has not yet attained the same level of quantification of physics. Wether that will always be the case is not so clear to me. My hunch would be yes, but with the improvement of computational power and techniques, the gap might become smaller at a fast tempo. I've seen how data mining can be used to find relationships between different genes, it's really impressive. ## Answer 1813 - posted by: [aaronasterling](https://stackexchange.com/users/-1/84-aaronasterling) on 2010-12-30 - score: 1

    Science is based upon quantitative predictions. This is what it excels at and where it derives it's power from. There is some of this to be had from evolutionary theory, e.g. the Hardy-Weinberg Principle.

    On the other hand, the principle of natural selection has a property that sets it above any particular theory of gravity. In a logically consistent universe, a population of replicating entities subject to a selection pressure has to evolve. This is not to say that natural selection is the only way in which evolution can occur but it must happen. Essentially, there are seven requirements for a population to not evolve. Failure to meet any one of these requirements will result in a population undergoing evolution. This is a far higher level of certainty than can ever be placed in a conventional scientific theory such as General Relativity or Quantum Mechanics.

    Ultimately, the questions all seem to be conflating certainty with quantitative predictions. These are two orthogonal concerns. No amount of satisfied quantitative predictions should ever result in certainty of a scientific theory but, even if we could not derive a single quantitative prediction from natural selection, we would still have to be certain that it occurs.

    One of my favorite quotes is from Darwin's brother in law who saw his theory before it was published. His response to Darwin's concerns was "If the facts don't support this theory, then so much the worse for the facts." This is absolutely correct. We no more need to look for evidence of natural selection than we need to cut out little paper triangles to verify Pythagoras Theorem.

    ## Answer 1723 - posted by: [Nomæd](https://stackexchange.com/users/-1/27-nom-d) on 2010-12-30 - score: 0 Biology in general is a very complex subject in the mathematical sense. In the core of it, it is one of the layers that are built on top of The Standard Model of Particle Physics. The preciseness (with uncertainty in mind) of this basic scientific theory is in itself mathematically very complex. And when you start building up the levels from this basic level of fundamental particles and forces to hadrons and atoms, atoms to molecules, molecules to life (cells) and so on, by the time we want to analyze how evolution works on a molecular level, how DNA makes life, there are unimaginably numerous levels of complexity that needs to be calculated, that it's extremely hard to do in practice. But in principle, it is possible to calculate probabilities for certain changes, for example. Gravity is more certain in the sense that it is a very basic theory, with very little variables (F = mg), it always works (at least for objects that aren't moving near the speed of light) and also is easier to grasp. Both of them certainly work in retrospect, but unlike physics and chemistry, biology's complexity makes it extremely hard to predict certain outcomes. We can make good predictions in biology if we describe it from a level that's higher than the molecular level, and that's really what Darwin described in the theory of evolution. We can learn from experience, judge adaptations that were often selected for, characterize types of adaptations that work well in different environments and so on, and to build a model of likely possibilities that can occur, if we consider different surroundings and selective agents. But in a sense it's like weather forecasting. You have so many variables that you can predict only so much with ever increasing error margin once you try to see further to the future. ## Answer 1782 - posted by: [mfg](https://stackexchange.com/users/-1/135-mfg) on 2010-12-30 - score: 0 > Is it dangerous to concede the point that biology can not yet live up to the level of quantification of physics, and it never will, thus affirming it is less certain? With respect to the other queries, I am honest enough to concede that I am unqualified to venture a guess or cite a particular biologist. In terms of the danger in conceding this ground in a debate, one's entire platform, or a majority plank of it, would need to rely on the predictions of evolution. Since there are no predictions, no one is in any real danger. For the common evolutionist or atheist, though, basically it can be used as a trade: "your God explains everything and predicts nothing", and so does evolution. To return to the standard rebuttal from there, one need only posit the burden of proof back on the theist, where it was all along. The only cautionary note to add is that any atheist who wants to rely heavily on evolution had be prepared to argue without the haven of "science" at any time. The inherent red herring fallacy of questioning science in order to prove the existence of [God] is pretty obvious (as long as the atheist doesn't assert that science is proof of [God]'s inexistence). ## Answer 1818 - posted by: [Scott Mitchell](https://stackexchange.com/users/-1/336-scott-mitchell) on 2010-12-31 - score: 0

    I'd contend that both the orbit of an asteroid and the evolutionary lineage of a species could be expressed mathematically. The problem is there are too many variables and inputs to quantify and qualify such processes.

    Consider this...

    I'll happily tell you the chances of a tracked asteroid colliding with the moon on that date...

    How can you factor in, let alone even know, of all of the other trillions of objects floating through our solar system? How can you assess with any certainty what probability the asteroid in question will collide with or be impacted by the gravitational effects of other asteroids in, say, the asteroid belt between Mars and Jupiter? Ignoring chaos theory, there may be a formula you could arrive at that would give you a very good idea, but it would probably need to take into effect trillions of independent variables. Good luck with that.

    I surmise that the same could be said about evolution. If you somehow knew the precise populations, what pressures would be exerted, what mutations would happen and which ones wouldn't, how the environment would change in its many interconnected ways, and so on, then you'd probably be able to give a good guesstimate at the evolutionary path of a species over time. (Although 15,000 years is the blink of an eye in evolutionary terms.)

    --- All content is licensed under [CC BY-SA 3.0](https://creativecommons.org/licenses/by-sa/3.0/).