Creationist brain surgeon Michael Egnor has been busy over the last couple of days, posting first a "response" to Orac's challenge then a "response" to Mark ChuCarroll's repeated attempts to explain the concept of tautology to him. There have been several responses to these two posts over at various of the Scienceblogs already - PZ, Orac, Mark, and Kevin have all addressed one or both of Egnor's latest claims, and all of their responses are worth reading. I'm actually feeling a little left out right now - after all, Egnor still hasn't deigned to address the two specific examples I presented of cases where natural selection has and is playing a role in public health decisions.
I'm not going to address Egnor's claims about the role of the design inference in medicine at the moment. The stupidity really does burn, and Orac did a superb job of working through the brain spasms to deal with that post. Kevin and Mark have also done a good job addressing some of Egnor's claims in the more recent "tautology" post, but I think that there is something that I can add to their responses. An introduction - Dr. Egnor, I'd like to introduce you to Sir Ronald Aylmer Fisher, FRS.
Egnor, in his most recent attempt at demonstrating his complete and utter ignorance of evolutionary biology, argues that natural selection is a tautology, but other scientific theories are not:
Scientific theories generally cannot be reduced to tautologies. Newton's law cannot be reduced to 'If you let go of something, it will fall-- therefore, if you let go of something, it will fall'. Newton's law of gravitation, in its most 'reduced' form, states that the gravitational force acting between two masses is proportional to the product of the masses and inversely proportional to the square of the distance between the centers of the masses, and that the constant of proportionality is the gravitational constant. Newton's law is not a tautology, and it can't be reduced to a tautology. It is not logically true. It's empirically true, but it could have been false. The force of gravity could have been proportional to the inverse cube, not the inverse square. Neither is Einstein's theory of relativity a tautology. The curvature of space-time is given by Einstein's tensor equations. They've been confirmed experimentally, and they're not the least bit tautological.
. . .
All permutations of "Natural Selection"-- "survival of the fittest," "reproductive success," "mechanisms that contribute to the selection of individuals that reproduce," "sexual selection," "gametic selection," "compatibility selection"--reduce quite readily to "successful reproducers successfully reproduce," or colloquially, "survivors survive." The truth of Natural Selection is its logical structure. It can't be false. "Natural Selection" is logically true, but it's a weak theory because it's merely a tautology.
Egnor, it would appear, thinks that Newton's Law and Einstein's Theory are not tautologies because they don't just specify what happens ("things fall") but also the manner in which the things happen ("e = mc^2"). Natural selection, he believes, simply says what happens ("survivors survive"). Egnor is obviously unfamiliar with the work of Fisher (and many other evolutionary biologists).
Actually, I should probably rephrase that. Egnor is probably somewhat familiar with some of Fisher's work - the man invented a lot of the statistical procedures that are used in scientific and medical research. (He did the basic work behind ANOVA, for starters, and any test that uses an F-distribution owes something to him.) Egnor has probably seen that stuff regularly, but is not familiar with the work Fisher did on evolutionary biology. In particular, he is clearly unfamiliar to Fisher's mathematical description of the effect of natural selection on a gene pool.
Fisher's Fundamental Theorem of Natural Selection states that: "The rate of increase in fitness of any organism at any time is equal to its genetic variance in fitness at that time." It's incomplete - it only considers the rate of increase ascribable to natural selection, and it has been built on by many workers since then. I'm not going to go into all of that now, though, since this should be enough to illustrate the basic problem with Egnor's latest attempt: there really is a lot more to our understanding of natural selection than just knowing that survivors survive.
- Log in to post comments
With regard to Eggnor's brilliant quote about materialism and truth, it sounds to me like he was having a seizure, or perhaps he has been operating on himself. Goodness, I aspirated my coffee when I read his comment. Hm ...if matter and energy are all that exist, then falsity doesn't exist, and so materialism must be true!
Did you catch Egnor's Sunday contribution to intellectual discourse?
Mike - Fisher's fundamental theorem doesn't get round the criticism, because it still just plugs in the strength of selection.
What does answer the question is what precedes it in The Genetical Theory of Natural Selection: a formal definition of fitness, in terms of survival and reproduction. From this one can work out the strength of selection, and hence avoid the tautology.
Actually, Fisher's definition wasn't totally general, but the more recent formulations have a similar form.
Bob
Materialism is nonsense, because if matter and energy are all that exist, then truth doesn't exist (it's neither matter nor energy). If truth doesn't exist, then materialism can't be true.
Pardon my French, but what an imbecile.
I was also unaware that 'natural selection' in and of itself was a theory.
And I too feel left out - when the Egnor wars first began, I asked him at Orac's for some examples of morphological features that humans have that chimps do not that cannot be explained by evolution (one of his claims) and he never replied. I'm so hurt...
Nice post. Perhaps more directly addressing the tautology issue, however, is the very well-known fact (among biologists anyway, if not surgeons) is that natural selection is not equivalent to "survivors survive". Sometimes the survivors in a population (especially a small population) survive by mechanisms other than natural selection. Survivors can survive, not because they are they most fit, but rather because they are lucky. In sufficiently small populations, natural selection does not work, and genetic drift takes over. And if the fittest do not always survive (they do not), then the statement that "the fittest survive" cannot be a tautology.
As a prominent critic of evolutionary theory at the Discovery Institute, Egnor is of course not ignorant of genetic drift, which is extremely basic, simple, and fundamental populations genetics. Egnor is, however, nothing more than a well-paid deciever and clown.
Bob -- Fisher's fundamental theorem was misunderstood by geneticists for decades. It turns out that, properly understood in Fisher's exact terms (and as carefully stated by Mike in his post), it is completely general.
The FTNS is probably the most misunderstood concept in all of pop-gen, and the literature and textbooks are filled with errors on the subject. But don't just take my word for it -- here are some good refs on the issue. Frank 1997 is very good, the relevant section is found in pp. 1717-1719. Frank and Slatkin is an easy short read. If you can get Edwards 1994, that is pretty much the final word on the history of the misunderstanding. Price 1972 and Ewens 1989 are the key papers that (finally) explained what Fisher meant by the FTNS, and both prove that the FTNS is valid when stated as partial mean fitness increase due to natural selection alone.
Edwards, A.W.F. (1994) "The fundamental theorem of natural selection." Biological Reviews 69:443-474.
Ewens, W.J. (1989) "An interpretation and proof of the fundamental theorem of natural selection." Theor Popul Biol 36:167-180.
Frank, S.A. (1995) "George Price's Contributions to Evolutionary Genetics." J Theor Biol 175:373-388.
Frank, S.A. (1997) "The Price Equation, Fisher's fundamental theorem, kin selection, and causal analysis." Evolution 51:1712-1729.
Frank, S.A. (1998) _Foundations of Social Evolution_. Princeton: Princeton University Press.
Frank, S.A. and Slatkin, M. (1992) "Fisher's fundamental theorem of natural selection." Trends in Ecology and Evolution 7:92-95.
Price, G.R. (1972). Fisher's "fundamental theorem" made clear. Annals of Human Genetics 36:129-140.
Douglas - read what I wrote again, please. Your post isn't relevant to what I wrote.
I assume you haven't done any work in quantitative genetics, otherwise you'd be aware that Mike's verbal description of the FTNS is wrong (there's an "additive" missing). I'm not going to take Mike to task for that, because explaining additive genetic variance is a bit of a pain.
Incidentally, I wouldn't describe the FTNS as completely general: a couple of years ago a colleague showed me a (contrived) counterexample where fitness actually went down over time.
Bob
...natural selection is not equivalent to "survivors survive". Sometimes the survivors in a population (especially a small population) survive by mechanisms other than natural selection.
It is even worse than that. Natural selection is not about surviving, it is about passing on one's genetic material to the next generation. After all, survival is always temporary. Natural selection could favor a weak, slow, short-lived, unintelligent life form over a strong, fast, long-lived, intelligent one if the former has a sufficiently higher fertility.
So "survivors survive" simply isn't even close to an accurate summary of what evolutionary theory is all about.
Bob,
My post is indeed relevant to what you wrote. You said "Fisher's definition wasn't totally general", but that is incorrect; the FTNS is totally general. Please just read the references I cited if you don't believe me -- the result has been proven and reproven and is not really debatable; it certainly isn't something I'll be able to convince you of in a blog comment section. Of course mean fitness can decrease over time -- the simplest models of frequency dependent selection show that. But that is not what FTNS is about, contrary to popular misconception. The FTNS concerns only fitness changes directly attributable to selection alone, which always increases (e.g., frequency dependent selection also includes changes in fitness which are not wholly attributable to selection, but rather are due to changes in allelic frequency). Again, please just read the references. The FTNS holds universally, absolutely, in all circumstances. Furthermore, the qualifier "additive" is unnecessary and is in fact misleading here; see Frank 1997, p. 1716 and following.
Douglas: read the previous paragraph in my post. The next paragraph follows that. Just to make sure there couldn't be any chance of mis-understanding, I even mentioned Fisher's definition.
You also seem to concede my point that the FTNS is not general: you accept that it doesn't apply with frequency dependent selection. I'm not impressed by "it's completely general: it applies except when it doesn't", which seems to be the only way to maintain your position.
Bob
Bob: The FTNS is absolutely general. There is a difference between partial fitness change and total fitness change. If you don't understand that then I give up. (It looks like you refuse to read the scientific lit too, so this is a senseless conversation. Bye.)
Douglas - THE COMMENT WAS ABOUT FISHER'S DEFINITION OF FITNESS, NOT ABOUT THE FTNS.
Got that?
And his definition (of fitness, not the FTNS) isn't completely general: it only applies for a constant population size.
Bob