Towards A Good Samaritan World

Wednesday, September 13, 2006


Tom engages my line of argument but makes an elementary mistake:

I'd just like to point out a sort of post-modernist corner that you've painted yourself into that undergirds your entire argument.

The "post-modernist" corner Tom is talking about simply IS the Popperian criterion that was the basis for the discussion:

You wrote: "If I mix carbon, hydrogen, nitrogen and oxygen in a totally sterile test tube, leave it in the sun for a while, and lo, we find living things in there a few months later, that surely shows that "living organisms can occur naturally." But if there are not living things there, does that show that living organisms can't occur naturally? Of course not."

Using your gravity example, if I drop a rock and it does obey our current understanding of gravity, does that mean that it can't ever *not* obey our understanding of gravity?

Now, what Tom is noticing is that there's a "heads I win, tails we flip again" aspect to both arguments, evolution and gravity. BUT...

In the case of gravity, it's the gravity-skeptic who gets to say "heads I win, tails we flip again." I drop the rock. If it floats, gravity is disproven. If it falls, the theory of gravity is not proven, it merely fails to be disproven for the time being. The gravity-skeptic can say: "Okay, so that data point is consistent with the theory, but what about next time?" I drop the rock again; he can say that again. Ad infinitum.

That was Karl Popper's whole point. That's why he talks about "the conjectural nature of human knowledge." Karl Popper accepts Hume's claim that induction is not valid, and he actually argues that we don't use induction, that it's a kind of "optical illusion." What we are doing, rather, is framing hypotheses which can then be proven or disproven by experience. Whether or not that's true as a description of cognitive psychology, I don't know, but it does seem to describe the "hypothetico-deductive" methods of science. A scientist concedes, as it were, the right of the skeptic to play the "heads I win, tails we flip again" game. Go ahead, repeat the experiment as often as you like he says, in effect. If ever my theory does not hold, even once, it stands disproven. Yet no matter how many times it does hold, it will still not be, strictly speaking, it will remain conjectural, though our degree of confidence in it will, of course, increase.

Now, the evolutionist who claims that life can occur "naturally" is in exactly the opposite position from the gravity-theorist. He does not permit the evolution-skeptic to play "heads I win, tails we flip again." He insists on his right to play "heads I win, tails we flip again." If life doesn't occur naturally, it proves nothing. If life does occur naturally, it proves the theory correct.

Tom sums up:

By being skeptical of inductive reasoning, you call into question any and all scientific claims, because they inherently rely on induction. In fact, I might go so far as to say that science is tautologically induction. So, you're right: if we can't use induction, then Darwinism isn't falsifiable; it wouldn't be confirmable or disconfirmable either way in this case.

But I'm bracketing my skepticism about inductive reasoning for purposes of this argument, or at any rate adopting a Popperian version that permits hypothetico-deductive reasoning as a means to conjectural knowledge. Under those criteria, many scientific theories, most scientific theories, stand up perfectly well, making clearly falsifiable predictions, and evolution, I have been saying until now, does not.

However, Nato beats the challenge. Here's the winning comment:

If we found a fossil (or better yet, a set of fossils, to avoid the anomaly issue) that was laid down in an age incontrovertably too early for its lineage. Here I'm thinking of something like cetacean fossils in rock dated solidly prior to the evolution of vertibrates. That would either disprove evolutionary history or prove time travel (anomaly?), anyway.

That's the best answer I've heard anyone give to this challenge so far. It actually succeeds in being comparable to the floating rock for gravity: a single radically anachronistic fossil could notionally disprove the whole theory, although, as with the floating rock, we'd look at a few other explanations first.

I'm not necessarily convinced still that evolution qualifies as science, because it seems to me that this (set of) disconfirmation(s) is too special, as it were, far narrower relative to all conceivable events than the set of disconfirmations of gravity. It seems to me that the set of possible worlds with anachronistic fossils is Vanishingly small (as Dennett would say) compared to the Vast set of possible worlds without anachronistic fossils, whereas the set of possible worlds with floating rocks and other gravity-disconfirmations is Vast compared to the Vanishingly small world of gravity-confirming falling rocks...

Nonetheless, I accept that Nato has beaten the challenge of coming up with a possible disconfirmation of Darwinism. Well done!


  • The possible worlds discussion deserves comment (and some requests for elucidation) but for now all I have time to note is that while anachronistic fossils are somewhat of "cheap" in that they're a sort of special case*. Systems theoretical sciences are still new compared to more concrete sciences like chemistry and physics, but it seems that there are bound to become a few stable points in evolution's predictive machinery that may lend themselves to identifying capabilities finely enough to rule out certain (probably smallish) classes of design features.

    There are a lot of genetic analogs to the anachronistic fossil case, though there's still several types of genetic exchange that are poorly enough understood that counterexamples would be regarded as evidence of a heretofor undiscovered exchange mechanism that evidence against evolution. Of course, the reality so far is that genetic exchange mechanisms have usually been hypothesized prior to confirmatory evidence being found in the wild - viral splicing to multicellular germ lines are the main example that spings to mind here.

    *Though not THAT special - after all, it's the progressive aspects of the fossil record that many (though not me) would name as the most convincing argument in favor of evolution. One might think that if the classic young-Earth creationist line were true, then more or less all fossils would be about as likely to be contemporary.

    By Blogger Nato, at 8:41 AM  

  • Pardon the editing mistakes in the above.

    By Blogger Nato, at 8:41 AM  

  • Experiments with Darwinism and gravity are perfectly analogous. Falsification is about refuting predictions (through induction, of course). Gravity predicts the rock will fall, so if the rock doesn't fall, then gravity *may* be disconfirmed. But what if we tried the same experiment using a helium atom as opposed to a rock? It would be impossible for us to decide one way or another the veracity of the theory of gravity due to the poor design of the experiment. A more instructive example than the rock falling in Newtonian gravity is the light bending through gravitational lensing predicted by Einstein (or you really could use any of Einstein's predictions). Einstein published his theory in 1905, but the light bending effect wasn't experimentally verified until 1919 due to the difficulty in formulating a proper experiment. Until that point, you could use the same argument against his theory that you're now using against Darwinism. Does that mean that his theory was not falsifiable at some point, and then it magically became falsifiable later? No, of course not. His prediction required an appropriate experiment to confirm or disconfirm it. So we don't have a perfect experiment to test parts of Darwinist theory, just like most of Einstein's theories back in the day, but that doesn't mean those parts aren't falsifiable. It just means we need to formulate clever experiments to test the predictions. Now, I think you're claiming that the kind of experiments required would be computationally intractible, but that's an easy and lazy claim to make against any theory unless you have hard theoretical evidence indicating that that's the case. If you made a strong theoretical case indicating that computationally tractible experiments could not be devised to test Darwinist theory, then I would concede at that point that Darwinism is not falsifiable in a practical sense (though it still would be theoretically falsifiable). But really, Nato's fossil example and my young Earth example are sufficient to falsify a great many of Darwinst predictions, and so you really have your work cut out for you if you want to prove otherwise.

    By Blogger Thomas Reasoner, at 12:02 PM  

  • This is driving me crazy. Why is the Popperian argument so hard to understand? Tom writes:

    "Experiments with Darwinism and gravity are perfectly analogous."

    Even putting to one side the fact that the fossil record is not an "experiment," I am baffled as to how Tom thinks that he has supported anything like this claim, but perhaps the nature of his confusion is illuminated by this statement:

    "Gravity predicts the rock will fall, so if the rock doesn't fall, then gravity *may* be disconfirmed."

    Please try to understand this. In Karl Popper's description of science, the nature of a truly scientific theory is such that a single counter-example disproves it. Of course a single floating rock would not disprove the theory of gravity because there might be magnetic fields involved or it might be full of helium, or hanging from a very slender invisible thread, or supported by air currents, etc. This is splitting hairs. There are only a few potential explanations of why a rock would not fall, compatibly with the theory of gravity, and if an actual floating rock were observed, it would be a fairly quick and straightforward matter to exhaust them. If the anomaly remained, the theory of gravity would be refuted. Of course the response to refutation would probably be refinement rather than abandonment of the theory: that's standard, and irrelevant.

    I don't think the young-world argument works, precisely because the Darwinian narrative itself is taken to be one of the strongest arguments for the older world. There probably couldn't be any direct evidence of a young world; certainly no straightforward observational evidence (absent the invention of time travel, but then, how could you really know you had gone back in time, as opposed to some other travel to an alternate possible world which happened to resemble an earlier period of Earth's?) Even if such evidence could be found, this would merely present people with the choice of choosing whether to trust it, or the Darwinian theory.

    Nato's disconfirmation example-- an anachronistic fossil-- seems at least notionally valid, and as Nato comments later, it illuminates a general category of possible disconfirmation scenarios. This suggests that Darwinian theory may be more defensible in Popperian terms than I had realized, but it comes nowhere close to showing that "experiments with Darwinism and gravity are perfectly analogous." Tom doesn't seem to have grasped the epistemological dilemma that he is in.

    By Blogger Lancelot, at 12:36 PM  

  • Between disconfirmation qua Popperian concept and disconfirmation in pratical science goes, I think there's a large gap - a gap not uncommon between a philosopher's idealized case and the scientist's real-world heuristic evidence weighing. A philosopher can just state a scenario and it's ceteris paribus true for the purposes of argument. In real experiments, however, there are usually so many sources of error and misinterpretation that scientists maintain a fairly strong skepticism if an experiment appears to disconfirm a theory considered well-established. It is in this practical realm where I think that evolution and gravity are very similar. When one moves to the pure propositional world of philosophy, however, the two are shorn of their messy empirical sides. Since gravity is exhaustively described by a bit of compact mathematics, it's easy to describe simple counterexamples. Evolution, however, is not only algorithmic design as a concept, but a large body of dynamical systems that intantiate that concept in different ways. The fossil history of our own Earth, since it is a result of this aggregation of processes, works as a tractible synecdoche thereof but the expansiveness of "evolution" even in propositional terms means it cannot be directly dealt with as simply as we do gravity. I think that this is at least part of what generates the disconnect between Tom and Lancelot, though by no means do I regard it as a complete exegesis.

    As for "Darwinism" being the chief argument against a young Earth - this is simply incorrect. Geology and astrophysics (primarily) establish the age of our planet, which protobiologists like Christian de Duve take as an input.

    By Blogger Nato, at 4:11 PM  

  • Some day I will learn to edit before I post.

    By Blogger Nato, at 4:11 PM  

  • There's plenty of evidence for an old earth. The fact that evolutionary theory takes this as an "input" is another reason why it's less compelling as a potential disconfirmation of Darwinism; it's not really a prediction of evolutionary theory, and Popper was particularly interested in predictions. But the point stands: if someone came up with evidence (somehow) for a young earth, it would without doubt be objected: "But how could all the life-forms we observe have evolved in that time?" Hence a certain circularity.

    However, this point is non-essential...

    By Blogger Lancelot, at 5:29 PM  

  • The precise (such as it is) time is taken as an input to tune the constraints on more speculative areas of evolutionary hypothesis, but I don't think anyone would regard an evolutionary account of Terran life remotely plausible if the Earth (somehow) turned out to be 10k years old. That is to say, it's taken as an input, but only within limits. In fact, there remain those who think the ~200 million years available between the condensation of the oceans and the first known life on Earth is too short to get from abiotic sludge to primitive protists, and so conjecture that life on earth decends from some space protist that settled down on the surface at some point.

    By Blogger Nato, at 7:47 PM  

Post a Comment

<< Home