Skip to content

Tautology 5a: The issues

So where are we? The tautology problem has laid open some deep concerns and confusions about evolution. In this post (5a) and the next (5b), which I will return to afterwards and add and amend (in other words it’s a work in progress), I aim to lay out what I think the issues are:

1. Prediction and testing: how does the theory achieve this?

2. The nature of fitness: is it just a mathematical variable or a physical property?

3. What is a function? Is it in the mind/theory, or in the world?

4. Is natural selection a mechanism? If so, what kind?

5. Is the principle of natural selection a law?

1. Prediction and testing

Is the principle of natural selection (PNS) is a tautology, then it follows that it will always be true. This means that any outcome must be the result of natural selection. Since, if it is a tautology, it is true in each case, nothing can falsify it, and it predicts every outcome, so goes the argument.

I think this makes sense only of you treat the fitness of every variant as the same. Obviously if you have two variants, and they have different fitnesses, the outcomes will be different. If the difference in fitness is less at one time than another then the theory predicts a different outcome. In other words, when you start applying some figures to the PNS, you start to get predictions. If the argument is purely verbal, then the tautology seems inevitable. This leads us to the question of what fitness is, however.

2 The nature of fitness

If fitness is a key term in the PNS, what can we do with it to prevent tautology? There are four approaches, so far as I can tell.

One is to treat fitness as a propensity. In this way it is a physical disposition, and not a logical truism. This is the approach taken by Mills and Beatty (1979, see Rosenberg 1982).

Another is to treat fitness as a multiply realised property, that is, a supervenient property (Rosenberg 1983, Rosenberg and Williams 1986, see Sober 1984. Rosenberg had argued that most biological concepts are supervenient elsewhere, 1978). Here fitness is something that emerges out of the physical/biological properties of organisms and genes.

A third is to distinguish between expected fitness and actual fitness. Here the idea is that actual fitness is based on reproductive success, and is therefore tautological, while expected fitness is soemthing that measures the adaptedness of an organism in its environment. Bylerly and Michod (Byerly 1981, and Byerly and Michod 1991) had distinguished between the r-fitness or rate of increase of a genotype in a population from the F-fitness or the causal factors acting on the genotype (see de Jong 1994 for a summary).

A final proposal is to eliminate fitness entirely from the real world and to treat it as a useful computational aid (Krimbas 2004, cf. Brandon 1990).

Each of these has its defenders and critics. I will return to this later.


Brandon, Robert N. 1990. Adaptation and environment. Princeton, N.J.: Princeton University Press.

de Jong, G. 1994. The Fitness of Fitness Concepts and the Description of Natural Selection. The Quarterly Review of Biology 69 (1):3-29.

Krimbas, Costas B. 2004. On fitness. Biology and Philosophy 19 (2):185-203.

Mills, Susan K., and John H. Beatty. 1979. The propensity interpretation of fitness. Philosophy of Science 46:263-286.

Rosenberg, Alexander. 1978. The Supervenience of Biological Concepts. Philosophy of Science 45 (3):368-386.

Rosenberg, Alexander. 1982. On the Propensity Definition of Fitness. Philosophy of Science 49 (2):268-273.

Rosenberg, Alexander. 1983. Fitness. The Journal of Philosophy 80 (8):457-473.

Rosenberg, Alexander, and Mary Williams. 1986. Fitness as Primitive and Propensity. Philosophy of Science 53 (3):412-418.

Sober, Elliott. 1984. Fact, Fiction, and Fitness: A Reply to Rosenberg. The Journal of Philosophy 81 (7):372-383.


  1. jeff jeff

    I was almost ready to say that #1, propensity (probability) solves everything. But then I realized, that just like actual fitness, you can’t really know it until post hoc – even worse, in this case it’s multiple post hocs to calculate a probability that’s still won’t be 100% accurate under ever-changing conditions.

    So that leaves #4 for me. One criticism would be that it denies even post hoc fitness. But if actual fitness is contingent upon random events, is it a meaningful word? On the other hand, one can observe a batch of newborn animals, observe that one is very sick, and conclude that it’s siblings are more fit, so fitness seems quite real. This is confusing.

  2. For me it’s a combination of 2-4. Organisms don’t have a fitness, it’s something we ascribe to them (4). The number we ascribe is determined by the properties of the individual, and the environment it is in (2). But this is the expected fitness (3): the actual reproductive success is stochastic. If you look at the stochastic models of population genetics, it’s clear that this is how fitness (i.e. s or w) is defined.

    In empirical work a lot of people call reproductive output fitness, which annoys the heck out of me, simply because they’re conflating selection and drift.

  3. Paul Armstrong Paul Armstrong

    Jason Rosenhouse questions whether fitness really is tautological. It is not merely the result of surviving and passing on heritable traits that gives an organism fitness. The traits themselves confer fitness because they confer a selective advantage in a particular environment. He writes, “[t]here is nothing tautological about saying, for example, that moths possessing dark coloration will be less visible than light colored moths to predatory birds when resting on dark-colored trees.” Rosenhouse also quotes S.J. Gould’s response to the tautology question, “In nature, Darwinian evolution is also a response to changing environments. Now, the key point: certain morphological, physiological and behvioral traits should be superior a priori as designs for living in new environments. These traits confer fitness by an engineer’s criterion of good design, not by the empirical fact of their survival and spread. It got colder before the wooly mammoth evolved its shaggy coat.”

Comments are closed.