Summary: The claim that evolutionary theory is unscientific or unfalsifiable [1] because it is a tautology, results from a focus on and a misunderstanding of the expression "Survival of the Fittest" (SoF) as well as a muddling of the various possible meanings of tautology.


Previous
Contents
Next

H istorical Introduction to the Tautology Argument

The tautology argument grew out of a change made by Darwin to the fifth edition (published Feb. 10 1869) of his "On the Origin of Species". He changed the title of the fourth chapter from "NATURAL SELECTION" to "NATURAL SELECTION; OR THE SURVIVAL OF THE FITTEST." (SoF) He wrote that he did this because "Several writers have misapprehended or objected to the term Natural Selection" on the basis that nature can't "select" in the way that man can "select".

Beginning from Samuel Butler's charge in 1879 that natural selection is a "truism", this led to a focus on the phrase SoF and the assertion that because survival rates define fitness, "'fittest' has no force" and thus natural selection and hence the whole of the theory of evolution explains nothing.

Here, note that an attack launched against the phrase SoF immediately incorporates its synonym "Natural Selection", despite the obvious fact that selection by nature is no more tautological than selection by man. Thereafter, as usual, the attack is widened to include all of evolution.

A Current version of the Tautology attack

A recent version of the tautology argument can be found in Ann Coulter's 2005 book "Godless: The Church of Liberalism" where she says:

"The second prong of Darwin's "theory" is generally nothing but a circular statement: Through the process of natural selection, the "fittest" survive. Who are the "fittest"? The ones who survive! Why look - it happens every time! The "survival of the fittest" would be a joke if it weren't part of the belief system of a fanatical cult infesting the Scientific Community. The beauty of having a scientific theory that's a tautology is that it can't be disproved." (pp. 212-213) [2].

Some basic reality checks

Note that the foregoing 'argument' characterizes as a "joke" the idea that it is the best adapted (fittest) parents who have most offspring. So what do creationists propose as an alternative? - that it is the worst adapted who are most fertile; that it is the arctic fox with the shortest fur who fathers the most pups generation after generation?

No, apparently not, Coulter like many other fundamentalists, accepts that such adaptation occurs, but denies that adaptation below the level of species (microevolution) is really evolution. She says:

"Evolution is not the capacity of bacteria to develop antibiotic resistance, but which never evolves into anything but more bacteria. Evolution is not the phenomenon of an existing species changing over the course of may years for example." (p. 202)

But this reveals a contradiction. SoF refers to only one element of evolution, not the entire theory. SoF doesn't refer to the creation of new species, or to the tree of common descent generated by speciation, or the nested hierarchy of characteristics within that tree, or even to mutations as the cause of variation. While supposing that Coulter, like many creationists, believes that all characteristics are preloaded into the genome by design she also clearly knows that species _do_ change in response to changes in their environment. But this is precisely the part of evolution that SoF describes! So why are creationists arguing that a position they already accept is a "joke"? Could it be they don't understand their own argument?

However, a more fundamental problem with Coulter's argument is that it sets up an equivalence between survival and fitness. "Who are the "fittest"? The ones who survive! Why look - it happens _every_ time!" (emphasis added) But if that is true, then all differential survival must necessarily be selection. But we have a name for differential survival that isn't selection; it is called drift (basically, changes in a population's gene pool due to chance). And in fact we can often perform tests that distinguish selection from drift [3]. We couldn't do that if selection were just "those that survive survive".

We now turn to a more detailed examination of tautologies.

Are all tautologies necessarily true?

Asking if tautologies are necessarily true may seem ridiculous, after all, isn't that the definition of a tautology? In fact there are a number of definitions. Two common formulations are:

Semantic: saying the same thing twice (which dates from 1579). Logic: a statement that is necessarily true.

Looking at these two definitions we immediately note a key difference. Saying the same thing twice, if the statement is false can't be necessarily true. Hence the semantic and logic forms are not equivalent.

The quote from Coulter begins by describing SoF as a "circular statement" an apparent reference to the semantic meaning, but ends by stating that tautologies "can't be disproved" presumably refering to the definition in logic. This makes about as much sense as saying that bats are nocturnal mammals that catch flying insects using echo location so I use mine to hit base balls. This is definition switching. Does Coulter believe bad puns can discredit evolutionary theory?

Due to this ambiguity, in the remainder of this FAQ, when we say SoF is not a tautology we mean not necessarily true. It is more difficult to make a categoric statement that it is not a semantic tautology.

Are tautological statements verifiable?

Since a semantic tautology is, in principle, just the same idea expressed in different formulations, if one can establish that one of the formulations is false, then the statement must be false. An examination of some examples shows that properties of commonly cited tautologies vary in this respect.

Here we consider six examples: "married husband" (or more formally "all husbands are married men") "free gift", "cold ice", "tuna fish", "horned unicorn" and "lubricating grease". All these cases are considered to be tautologies because an adjective (like lubricating) describes one of the necessary properties of the subject (here grease). Note that some can also be viewed as tautologies because the subject (grease again) is an obligate member of the class (lubricants) which modifies it. Thus we have moved from saying that a tautology is a repetition of the "same thing" to repeating the "thing" and one of its _necessary_ properties or one of the classes of which it is a member.

Looking more closely at our examples we note that they can, with a bit of force fitting, be sorted into two groups:

necessary: those whose meanings are necessary only from the words, including ,"all husbands are married men" "lubricating grease", "horned unicorn", and "free gift".

contingent: those whose meanings are contingent on an examination of associated real world facts including "all tuna are fish" and "cold ice".

We can see this difference clearly if we consider what would happen during a process of verification. If in a survey one found an unmarried husband, one would simply exclude him from the category of husband and move on. Hence, it is a _necessary_ truth that a husband is married. Similar reasoning applies to "lubricating grease" [4], "free gift" and "horned unicorn" since all are human constructs, not independent realities.

By contrast tuna and ice exist independently of human classification. Thus, even though the idea of a tuna which is not a fish may seem absurd, the fact that tuna are fish is _contingent_ on observation. Tuna are fish (and dolphins are not) because morphologic examination and genetic testing have confirmed their position in the tree of life. The case against "cold ice" as necessary is just as clear because it is possible, at least in principle, that we may one day see warm ice, as a result of manipulation of water with additives, or in cleverly designed magnetic fields or under extremely high pressures etc. Hence, it _can_ be (or could have been) false that ice is cold or tuna are fish.

Having established that tautologies that can be classified as contingent are verifiable the question is: Is SoF necessarily true or is its truth contingent on observation? To answer this question it is useful to look at SoF as a mathematical expression.

Survival of Characteristics, not Individuals

In population genetics SoF is represented mathematically. Therein we see a formal variable called fitness (W) which is measured by the proportion of a trait that survives into the next generation. The simplest form of the equation looks like this:

W_abs_ = N_after_/N_before_

where:
W_abs_ is absolute fitness
N_before_ is the Number of individuals with some genotype in a first generation (before selection)
N_after_ is the Number of individuals with an alternative genotype in the following generation (after selection)

Note that in the above statement there are absolutely no surviving individuals. Thus Coulter's question and answer "_Who_ are the "fittest"? The ones who survive!" fundamentally misrepresents the intended meaning of SoF. In the evolutionary timescale individuals never survive, hence fitness refers to survival rates of some specific heritable _characteristic_ in a population over time.

Specific heritable characteristics include things like:

  • visual sensitivity to yellow
  • densest fur
  • longest tail feathers
There is simply nothing tautological (in the circular sense) about survival of long tail feathers. That "fitness" is intended to refer to specific characteristics is the core to understanding that SoF is not in any sense a tautology, because by observation we can, for example, establish that those Arctic foxes with the densest fur survive to pass on that characteristic. Hence, instances of SoF are clearly observationally contingent, and repeated failures to see the obvious, would disprove SoF.

Mathematical Expressions of Scientific Laws as Tautologies

However, suppose that we continue by accepting that W_abs_ = N_after_/N_before_ is itself a tautology, perhaps because, as we saw above the equivalence of the same idea expressed in a different form is a tautology. If so, then all mathematical expressions containing a single equal sign between two sets of equivalent expressions must also be tautologies. This must therefore include Newton's F=ma and Einstein's E=MC^2.

F=ma is of particular interest in this respect because although we are here defining it as a tautology, it has in fact been, at least partially, falsified. That is, Newton's gravitational laws are true only outside relativistic frameworks, a fact which has been observationally verified. The fact that Newton's laws of motion have been at least partially falsified leaves no doubt that mathematical formulas that are observationally contigent are not empty circular tautologies.

As we saw above, _drift_ as an alternative mechanism to _fitness_ similarly partially falsifies SoF, thus SoF is likewise observationally contingent, not necessarily true.

SoF does not have a propositional form

An additional, problem with the Coulter's argument arises from the fact that tautologies must be formally expressible as _propositions_, thus:

  • "married husband" becomes "all husbands are married men"
  • "tuna fish", becomes "all tuna are fish"
  • "horned unicorn" becomes "all unicorns have horns"
What about "survival of the fittest"? As presented it is not a proposition. If this statement had a propositional form it would read something like: "all the fittest are survivors" or "all survivors are the fittest". But both these statements are transparently false since no one believes the fit always survive. In short SoF cannot be a tautology because it is not expressible as a proposition. It is more a descriptive label or suggestive summary intended to outline one part of Darwin's theory.

Summary Points:

  • In SoF fitness cannot be equivalent to survival because characteristics generated by drift also survive, thus SoF cannot be a tautology.
  • Survival of the Fittest (SoF) was intended to be a synonym for Natural Selection (NS), but NS is clearly not a tautology.
  • SoF does not include many key aspects of evolution, like common descent, the nested hierarchy, mutations, or speciation. What SoF does include is accepted by most (but not all) creationists as part of "microevolution". So why are they attacking the only part of evolution many of them agree with?
  • SoF is a descriptive outline of part of Darwin's theory. As such it cannot be a tautology because it isn't expressible as a proposition as in 'All survivors are the fittest'.
  • Semantic and logic tautologies are substantially different. Switching from the semantic to the logic definition in mid argument produces incoherent nonsense.
  • SoF (expressed as W_abs_ = N_after_/N_before_) is a tautology only in the same sense that F=ma and E=mc^2 are tautologies.
  • Since fitness refers _not_ to individuals but to _characteristics_ passed to the next generation, those characteristics can be plugged into mathematical formulas for testing against the real world in the same way m in F=ma can be checked. There is nothing tautological about survival of the longest fur, longest tail feathers, or most bark-like coloration.
  • Purely semantic tautologies like "All husbands are married men" are by definition necessarily true, but "tautologies" that make reference to verifiable facts in the real world such as "All tuna are fish", are observationally _contingent_ and thus can be verified. Since statements about fit characteristics can be checked against real world observations they cannot be _necessary_ truths.

[1] The word unfalsifiable will be linked to John's FAQ: http://www.talkorigins.org/faqs/evolphil/falsify.html

[2] Temporary Info: I got the Coulter quote here: http://www.csicop.org/specialarticles/show/is_natural_selection_a_tautology/

[3] The word "drift" will be linked to the drift FAQ. http://www.talkorigins.org/faqs/genetic-drift.html)

[4] Note: Since water and other unlikely materials can act as lubricants "lubricating grease" depends on human classification.


Previous
Contents
Next
begin trailer

Home Page | Browse | Search | Feedback | Links
The FAQ | Must-Read Files | Index | Creationism | Evolution | Age of the Earth | Flood Geology | Catastrophism | Debates
end trailer