Colin Patterson, Evolution, 2nd Ed. (1999)Ch. 14 - Proof and disproof; science and politics emphases mine, [paraphrases in square brackets], my comments [in small print].
Turning now to the second or special theory of mechanism, many critics have held that natural selection, as the cause of evolution, is not scientific because the expression 'survival of the fittest' makes no predictions except 'what survives is fit' and so is tautologous.... Indeed, natural selection theory can be presented in the form of a deductive argument, for example:
In this sense, natural selection is not a scientific theory but a truism, something that is proven to be true, like one of Euclid's theorems... [Genetic drift & neutral drift can also protect natural selection from the possibility of falsification.]
- All organisms must reproduce.
- All organisms exhibit hereditary variations.
- Hereditary variations differ in their effect on reproduction.
- Therefore variations with favourable effects on reproduction will succeed, those with unfavourable effects will fail, and organisms will change.
[Patterson seems to be saying that natural selection is actually more than a tautology - it's axiomatic - proven true like Euclid's theorems!]
...But the essence of scientific method ... is to test two (or more) rival theories, like Newton's and Einstein's, and to accept the one that passes more or stricter tests until a better theory turns up. We must look at evolution theory and natural selection theory in terms of performance against the competition.
[The general theory of evolution] has only one main competitor, creation theory.... All creation theories are purely metaphysical. They make no predictions about the activities of the Creator, except that life as we know it is the result of His plan. Since we do not know the plan, no observation can be inconsistent with it. ...
In 1978 Popper wrote that the Darwinian theory of common descent 'has been well tested' (he did not say how) and 'That 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. ... Yet in every particular case it is a challenging research programme to show how far natural selection can possibly be held responsible for the evolution of a particular organ or behavioural programme.'
Any particular non-tautological version is a proposal for a "theory of fitness" (at least in a restricted setting), and thus exactly the kind of thing I insist is needed, and the kind of thing which, when provided in general will supply more order than will support the argument-from-no-design.