Survival of the Fittest: Tautologous and Unfalsifiable?

Survival of the Fittest: Tautologous and Unfalsifiable? March 30, 2021

Yesterday, I ran a piece on the idea that you often hear from creationists that “survival of the fittest” (SoF) is a tautology that carries no meaning. I hopefully dealt with this in some way in that previous post. However, I would like to take the opportunity to add a few ideas and to discuss a connected notion, one that survival of the fittest is an unfalsifiable scientific concept, thereby invalidating it as being a scientific concept.

First of all, let us discuss the idea of falsifiability.


Philosopher of science Karl Popper constructed the idea that if a scientific theory is unfalsifiable and it is not a scientific theory. His Falsification Principle can be summarised as follows:

Summary of Popper’s Theory

  • Karl Popper believed that scientific knowledge is provisional – the best we can do at the moment.

  • Popper is known for his attempt to refute the classical positivist account of the scientific method, by replacing induction with the falsification principle.

  • The Falsification Principle, proposed by Karl Popper, is a way of demarcating science from non-science. It suggests that for a theory to be considered scientific it must be able to be tested and conceivably proven false.

  • For example, the hypothesis that “all swans are white,” can be falsified by observing a black swan.

  • For Popper, science should attempt to disprove a theory, rather than attempt to continually support theoretical hypotheses.

Of course, the nature of falsifiability (and testability) is problematic in certain areas of science, particularly those theories that involve predictions (postdictions) about the past. How the universe began, evolution itself and all sorts of processes that might necessarily take billions of years or have taken place billions of years ago and are not in any meaningful sense testable seem to cause problems for such a principle. As David Bailey states in “Is Evolution Falsifiable?“:

Evolution, when viewed in the modern context of a huge volume of convincing empirical data, entirely qualifies as a rigorously testable theory. And in fact it has survived decades of withering testing. This is precisely why evolution is taken so seriously as the governing paradigm of modern biology.

It should also be pointed out that strict adherence to “falsifiability” is not an accurate description of the process of modern science. For one thing, major theories are seldom falsified by a single experimental result. There are always questions regarding the underlying experimental design, measurement procedures, and data analysis techniques, as well as questions of whether the underlying theories have been properly applied. For example, if we were to strictly apply Popper’s principle, Copernicus’ heliocentric theory was falsified from the start and should not have been further considered, because it could not predict planetary motions as accurately as the traditional Ptolemaic system. It only after Kepler modified the theory to include elliptical orbits with time-varying speeds, and when Newton showed that this behavior could be mathematically derived from his laws of motion, that it gained widespread acceptance. It must also be kept in mind that in most cases, “falsified” theories continue to be extremely accurate models of reality within appropriate domains. Even today, over 100 years after Newton’s mechanics and Maxwell’s electromagnetic equations

Applying the principle to survival of the fittest

Some people, including philosophers and scientists themselves (i.e., not evolution-deniers), have sought to apply this principle to the “scientific theory” of survival of the fittest. Indeed, even Karl Popper earlier in his career did this. Frank J. Sonleitner (in “What Did Karl Popper Really Say About Evolution?”[1]) observes:

But he did make one mistake—for which we should forgive him; some well-known biologists (who should know better) have made the same mistake. Popper takes “survival of the fittest” as the definition of natural selection (Popper, 1972, p. 241). This catchy phrase was an invention of Herbert Spencer, which Darwin, in a rare example of bad judgment, interpolated into later editions of On the Origin of Species: “This preservation of favorable individual differences and variations and the destruction of those which are injurious I have called Natural Selection or the Survival of the Fittest” (p. 64). Clearly it is an alternate name (and not a very apt one) for the process in question but not a definition.

The argument regarding “survival of the fittest” is that the only way one can usually tell who the fittest are is to see who survives. But then survival of the fittest becomes “almost a tautology” and hence untestable (Popper, 1972, p. 69; 1963a, p. 964).

I have come to the conclusion that Darwinism is not a testable scientific theory, but a metaphysical research programme—a possible framework for testable scientific theories. [Popper, 1976, p. 168]

I would advise reading the paper “Popper’s Shifting Appraisal of Evolutionary Theory” that details how Popper confuses appraisal of evolutionary theory with that of physical theories, and this somewhat explains where he went wrong.

In my previous article, this was something that Christian commenter Luke Breuer found to be a problem. Indeed, many commenters gave Luke and Enoch Arden (a denialist) a hard time over their claims but these are genuinely interesting ideas that have often been debated in good faith. However, I don’t think Enoch Arden comes from a position of good faith; on the other hand, I am confident that Luke is genuinely striving towards a coherent truth (that is not mean to be nearly as condescending as it sounds!).

In a Philosophy Now article (“Survival of the Fittest”) Fred Leavitt adds another couple of quotes to the mix:

For example, Karl Popper wrote:

“The trouble about evolutionary theory is its tautological, or almost tautological character; there does not seem to be much difference, if any, between the assertion ‘those that survive are the fittest’ and the tautology ‘those that survive are those that survive.’ For we have no other criterion of fitness than survival.” (Objective Knowledge, revised. ed., 1979, pp.69)

In the same vein, C.H. Waddington claimed that:

“Natural selection turns out on closer inspection to be a tautology, a statement of an inevitable although previously unrecognized relation. It states that the fittest individuals in a population will leave most offspring. Once the statement is made, its truth is apparent.” (in Evolution After Darwin, ed. Sol Tax, 1960, p.385)

Popper changed his mind

Sonleitner points out that, later in his career, Popper recanted his original claims and position on SoF and evolution in general:

The fact that the theory of natural selection is difficult to test has led some people, anti-Darwinists and even some great Darwinists, to claim that it is a tautology. . . . I mention this problem because I too belong among the culprits. Influenced by what these authorities say, I have in the past described the theory as “almost tautological,” and I have tried to explain how the theory of natural selection could be untestable (as is a tautology) and yet of great scientific interest. My solution was that the doctrine of natural selection is a most successful metaphysical research programme. . . . [Popper, 1978, p. 344]

I have changed my mind about the testability and logical status of the theory of natural selection; and I am glad to have an opportunity to make a recantation. . . . [p. 345]

The theory of natural selection may be so formulated that it is far from tautological. In this case it is not only testable, but it turns out to be not strictly universally true. There seem to be exceptions, as with so many biological theories; and considering the random character of the variations on which natural selection operates, the occurrence of exceptions is not surprising. [p. 346]

There will be more on this last point later (referring to mechanisms, as I did yesterday, such as genetic drift).

Tautology revisited

I gave a whole series of arguments about this in yesterday’s article; however, it is worth adding something from the aforementioned Philosophy Now article:

… This the concept of fitness is not needed. The theory can be rephrased as follows:

1. Individuals vary in many ways. (This is empirical.)
2. Individuals survive and reproduce with varying degrees of success. (This is empirical.)
3. Some of the variations in characteristics that affect survival and reproductive success are transmitted to offspring. (This is empirical.)
4. Populations will come to resemble successful individuals more than unsuccessful ones. (This is both empirical and a logical deduction from the preceding.)

Omitted from the new formulation is any discussion of the particular qualities that enhance or impair reproductive success. These differ from species to species and in different environments, and they may change over time. For example, Bernard Kettlewell’s classic book (The Evolution of Melanism, 1977) documents how dark-colored moths evolved from lightcolored ones in industrial areas of England. Whereas lightcolored moths had been better camouflaged in unpolluted areas, dark-colored ones are less conspicuous when resting on tree trunks blackened by pollution. Kettlewell’s work notwithstanding, evolutionists are frequently uncertain about the adaptive value of specific characteristics. Still, all living organisms are descendants of, and thus inherited characteristics from, ancestors who were fit enough to mature and reproduce. So evolutionists are justified in presuming that the vast majority of present-day organisms are fit.

It would be tautologous to argue that certain cheetahs outreproduce others because they are more fit. However, to suggest that the success of cheetahs depends partially on their leg size is not tautologous; it is a falsifiable scientific hypothesis. Scientists can use their previous experiences (observations that fast cheetahs are more successful than slow ones and that long legs facilitate speed) plus the principle of induction to predict future reproductive outcomes based on leg size. Alternatively, they can use their prior observations plus the inductive principle to hypothesize that present day cheetahs have long legs because ancient cheetahs with long legs left more descendants.

In a philosophical paper, “A Philosophical Analysis of the Original Version of Natural Selection“, Xiang-rui Luo states that:

The original version of Darwin’s natural selection theory, or the idea “survival of the fittest”, is unfalsifiable because of its circular definition of “the survival” and “the fittest”. In order for a theory to state the relationship between two ideas, such as “those who survived” and “those who are the fittest to the environment”, these very ideas should be defined independently.

But then goes on to almost immediately admit: “Darwin himself also positively agreed with and adopted this phrase [SoF] to generalize this idea of natural selection theory.” Which is to admit this statement of SoF is a generalisation and not the theory in and of itself. This is the key!

“Survival of the fittest” is not a scientific theory

The Philosophy Now quote leads me to my next point. Namely, that SoF is not actually a scientific theory and need not be beholden to the falsification principle. I can’t really put this any better than the following quote:

Here’s the funny thing about that aspect of Evolution: it isn’t the theoretical part. It’s the mathematical part (statistics, to be exact). And math is just logic.

The thing with statistical models is that they cannot make hard and fast predictions. We can perform an experiment where the statistical prediction never happens 100 trials in a row; it’s just unlikely. And we know that, so we can’t make any predictions that can be falsified. Instead, we use a notion of confidence.

Anyway, that was more background than anything. The point is that “survival of the fittest” is not the part of the Theory of Evolution that should be falsified because selection isn’t a special thing; it just logically follows. The portion that would falsify the theory would be if genetics was wrong (offspring did not inherit traits, random mutation did not occur, etc.) or if an outside force was actively preventing selection.

TL;DR It doesn’t need to be defended because it’s the externally consistent part of the theory.

This is a hugely important point. SoF is merely a statistical summary of the theory of evolution by natural selection. In order to falsify that theory you would need to look at the component parts of that theory. When we say that finding a rabbit in the Precambrian rock strata would falsify the theory of evolution, we should actually be more granular than that (and therefore more accurate), but we could say that the theory of evolution can be falsified. Again, though, survival of the fittest is not a synonym for evolution.

Wikipedia details some of the debate:

However, the expression “survival of the fittest” (taken on its own and out of context) gives a very incomplete account of the mechanism of natural selection. The reason is that it does not mention a key requirement for natural selection, namely the requirement of heritability. It is true that the phrase “survival of the fittest”, in and by itself, is a tautology if fitness is defined by survival and reproduction. Natural selection is the portion of variation in reproductive success that is caused by heritable characters (see the article on natural selection).[19]

If certain heritable characters increase or decrease the chances of survival and reproduction of their bearers, then it follows mechanically (by definition of “heritable”) that those characters that improve survival and reproduction will increase in frequency over generations. This is precisely what is called “evolution by natural selection“. On the other hand, if the characters which lead to differential reproductive success are not heritable, then no meaningful evolution will occur, “survival of the fittest” or not: if improvement in reproductive success is caused by traits that are not heritable, then there is no reason why these traits should increase in frequency over generations. In other words, natural selection does not simply state that “survivors survive” or “reproducers reproduce”; rather, it states that “survivors survive, reproduce and therefore propagate any heritable characters which have affected their survival and reproductive success”. This statement is not tautological: it hinges on the testable hypothesis that such fitness-impacting heritable variations actually exist (a hypothesis that has been amply confirmed.)[19]

Even if SoF was a scientific theory, it arguably has been, indeed, falsified. As mentioned in my previous article, genetic drift (whereby random neutral mutations spread throughout the population) shows that SoF does not hold all of the time. The fact that we have dominance of certain eye colour throughout populations is arguably a reality of this. Having brown eyes as a characteristic is not a fitness-imbuing genetic characteristic. It may be more dominant due to some kind of correlation, but it does not itself carry some kind of advantage, it seems. But, you could argue, this is a single characteristic that does not represent overall fitness.

It depends at what point you are looking at something, and what reference set you decide upon.

But what we can do from this is look at certain neutral characteristics and try to compare to a control. For example, imagine a very simple organism in a lab that has a hundred different genes. It is hypothesised that one gene (A7) is neutral to fitness. You could flip this to another neutral variant B7 in an experiment and see which organism outcompetes the other in a fair test. We could put the same amount of each micro-organism (A7 and B7) in a petri dish and see which variety gains dominance over some time. If, over a massive number of simulations, we have a 50-50 split (so that half time A7 dominates and the other half of the time B7 dominates), then we can see that this genetic variance has no bearing on the fitness of the organism.

However, on the 36th running of the simulation, B7 is the dominant micro-organism and is the only one to survive by the end of that simulation. If you just look at this simulation, then the fittest hasn’t survived. This is an example of genetic drift, of course. I would imagine it outcompetes by some kind of random chance – perhaps reproducing more quickly over one generation – thus conveying a statistical advantage that eventually reaches a mathematical tipping point. But it falsifies the idea that survival of the fittest is some kind of entirely pervasive mechanism. It isn’t; it is a generalisation. Organisms tend towards survival of the fittest, but sometimes, out in the real world, particular organisms will dominate on account of another mechanism, like this. And this can be shown in the lab, and probably has been in experiments like Lenski’s (which itself looks to verify SoF).

In other words, where survival of the fittest is used as a shorthand for evolution by natural selection, it is falsified every time another mechanism for evolution holds. Thus, when we see genetic drift work, it falsifies evolution by natural selection. Except, it doesn’t falsify it because when you have various mechanisms working towards the same end (the difference in allele ratios over time) then when one mechanism takes hold, it doesn’t mean that the other isn’t true at all. It just means that it wasn’t true in that instance.

This perhaps exemplifies the difference between a physical law and a phenomenon created by various different mechanisms and constrained in particular ways. A ball falling up when we predict it will fall down due to gravity, ceteris paribus, arguably falsifies gravity. However, a group of organisms having a different allele ratio in a particular group at some point in place time, victim to huge numbers of variables and laws, can be arrived at in various different ways.

The reality is that in real life, various evolutionary mechanisms interact with each other to produce spciation and variation. For example, see “Natural Selection, Genetic Drift, and Gene Flow Do Not Act in Isolation in Natural Populations“.

I will leave you now with a long quote from the philosopher Rebecca Millstein, whose essay on this appears in the Blackwell Guide to the Philosophy of Science (chapter available here). This should do the job:

No summary of the philosophy of evolution would be complete without a discussion of the “tautology problem,” given the amount of space that has been devoted to it. Yet given a proper understanding of tautology and evolutionary theory, there is neither a prima facie tautology, nor is there a problem. Nonetheless, much interesting philosophical discussion about the nature of fitness has arisen as a result of the misunderstanding.

The “problem” isn’t new, either. According to Hull (1969), evolutionists as far back as Darwin have been defending the theory of natural selection against the criticism that it is tautologous. Nonetheless, the criticism refuses to die, kept alive in large part by creationists who love to quote Popper’s claim that “Darwinism is not a testable scientific theory but a metaphysical research programme” (Popper 1974: 134; italics in original), but who ignore his subsequent recantation (Popper 1978).

The standard criticism goes as follows: the principle of natural selection is “the survival of the fittest,” but who are the “fittest”? Those that survive. The principle then becomes “the survival of those that survive.” Thus, the critics charge, the theory of natural selection is a tautology, and is therefore circular and empty; it says nothing about the way the world is, since it is true regardless of the empirical reality. This claim is often conjoined with the claim that the theory of natural selection is unfalsifiable – a tautology cannot be proven false.

A few technical points regarding the standard criticism – if there is anything wrong with 20 the phrase “the survival of the fittest,” it is that it is an analytic statement, not that it is a tautology, as Sober (1984) points out. A tautology is a statement that is true by virtue of its logical form alone, such as “Either it is raining or it is not raining.” An analytic statement, on the other hand, is a statement that is true by virtue of the meaning of its constituent words (i.e., true by definition), with the classic example being “a bachelor is an unmarried man.” If the phrase “the survival of the fittest” were to be worded as a statement that could be true or false – it isn’t a statement in its current form – then it would be characterized as an analytic statement rather than a tautology. Still, even as an analytic statement, the critics’ charge that the phrase is circular, empty, and/or unfalsifiable lingers.

One possible line of response to the standard criticism involves a reexamination of the concept of “fittest”. If the only thing that makes one group of organisms fitter than another is that the first group in fact survived when the second did not, then this seems to be the source of the circularity. In response to this concern, Mills and Beatty (1979) and Brandon (1978) independently developed the propensity interpretation of fitness (although Brandon prefers the term “adaptedness”). On this view, fitness is not defined in terms of an organism’s actual survival or reproductive success. Instead, fitness is an organism’s propensity, or ability, to survive and reproduce in a particular environment. (Fitness is never defined absolutely, but always relative to a given environment; what enhances survival or reproductive success in one environment may not do so in a different environment.) Thus, “the survival of the fittest” is not “the survival of the survivors,” but rather “the survival of those who have the greatest propensity to survive.” The organisms that have the greatest propensity to survive may not in fact survive; consider, for example, two identical twins, one of which is struck by lightening and dies, the other which survives and leaves offspring. Both are equally fit (have the same propensity to 21 survive and reproduce), yet one has greater actual reproductive success. In this manner, the propensity interpretation of fitness attempts to break the purported circularity of the theory of natural selection.

The propensity interpretation of fitness is not without its critics (see, for example Rosenberg 1982; Rosenberg and Williams 1986). Even Beatty and Finsen (née Mills) return to point out some technical difficulties with their own position (Beatty and Finsen 1989). Nonetheless, the view enjoys widespread acceptance among philosophers of biology (see, for example, Burian 1983; Brandon and Beatty 1984; Sober 1984; Richardson and Burian 1992). Sober responds to Beatty and Finsen’s self-criticisms and points out that whereas the criticisms apply to the particular mathematical implementation of the propensity interpretation, they do not challenge the nonmathematical heart of the propensity interpretation (Sober 2000).

In spite of the popularity of the propensity interpretation as an account of the concept of fitness, some philosophers – including Beatty (1992), who has changed his position on this issue – have argued that it does not actually solve the tautology problem. Rather, Waters suggests, if we spell out the principle of the survival of the fittest as “Organisms with greater higher fitnesses in (environment) E will probably have greater reproductive success in E than (conspecific) organisms with lower fitnesses” (1986: 211), there are two basic ways of interpreting the term “probably”: the propensity interpretation and the frequentist interpretation. Waters argues that if one chooses the propensity interpretation, the principle is true by definition; if one chooses the frequentist interpretation, the principle is not analytic, but it is untestable.

If this argument is correct, does that mean that the theory of evolution is circular and unfalsifiable? It might, if the phrase “the survival of the fittest” actually described the theory – but it does not. The real problem with the standard criticism is that it misrepresents evolutionary 22 theory, as Hull (1969) and Waters (1986) note. As discussed above, the present-day theory of evolution includes not only natural selection as a possible mechanism leading to the differential survival and reproduction of types; random drift is a possible mechanism, as are migration and mutation. In other words, in any particular case survival may not be “the survival of the fittest.”

Even Darwin’s theory of natural selection alone is not captured by this phrase;9 as previously mentioned, Darwin described natural selection as a process requiring 1) a struggle for existence where not all organisms that are born can survive, 2) heritable variations between organisms in the population, and 3) variations that confer a differential ability to survive and reproduce. Whether any or all of these conditions obtain in a particular population is an empirical question, not a matter of definition, and thus we can test the population for the presence or absence of the three conditions. The theory of natural selection is neither circular nor vacuous.

The tautology problem ought to be a dead issue, even if there are those who refuse to let it go. [p. 238-40]

I hope that clears some things up.

Finally, you can always grab my book that sparked off this debate here… Please do…

Survival of the Fittest: Metamorphosis [UK]


[1] Sonleitner, Frank J. (1986),  “What Did Karl Popper Really Say About Evolution?”, Creation/Evolution Journal, Volume 6, No. 2, Summer 1986.

Browse Our Archives

error: Content is protected !!