The opening of the current version of the Wikipedia article, “Evolution as theory and fact,” (with links and references removed) reads:
The statement “evolution is both a theory and a fact” is often seen in biological literature. Evolution is a “theory” in the scientific sense of the term “theory”; it is an established scientific model that explains observations and makes predictions through mechanisms such as natural selection.
When scientists say “evolution is a fact”, they are using one of two meanings of the word “fact”. One meaning is empirical: evolution can be observed through changes in allele frequencies or traits of a population over successive generations. Another way “fact” is used is to refer to a certain kind of theory, one that has been so powerful and productive for such a long time that it is universally accepted by scientists. When scientists say evolution is a fact in this sense, they mean it is a fact that all living organisms have descended from a common ancestor (or ancestral gene pool) even though this cannot be directly observed. [Emphases added.]
In explaining this, they cite the US National Academy of Sciences:
Scientists most often use the word “fact” to describe an observation. But scientists can also use fact to mean something that has been tested or observed so many times that there is no longer a compelling reason to keep testing or looking for examples. The occurrence of evolution in this sense is fact. Scientists no longer question whether descent with modification occurred because the evidence is so strong.
[Science and Creationism: A View from the National Academy of Sciences, Second Edition (1999), National Academy Press, Washington DC, 2006.]
{UPDATE, Jan 14, 2011: An objector, below, points out that NAS has somewhat corrected this blunder in the 2008 update to their pamphlet; which Wikipedia did not cite — NB: above, I am noting on how Wiki supported its claim as at Jan 12, 2011. That, first, leaves unexplained how from at least 1999 – 2006, such a body could make the basic error of confusing observed facts with inferred explanatory models, i.e. theories. That, surely, is a basic distinction that should be clear from Grade School mnemonics on the scientific method — tracing to Newton’s Opticks, Query 31 — that distinguish between observations and hypotheses and again experimental tests. Moreover, in the 2008 adjustment, NAS tries (using context) to indirectly compare favourably the operations science theory of gravity — something subject to direct observational tests — with macro-evolutionary theory, which is about origins issues on a deep and unobservable past that we can only infer to. In short, we have here a subtler form of the same blunder. A fairer approach would be to recognise the inescapable difference between what is observable and what is not, and so acknowledge that origins science theories are simply not as capable of empirical support as are operations science theories such as gravity. We use the principle of uniformity to infer from the present to the deep past of origins, but we do not directly observe that past. And, as we do so, that raises the point that, for instance, the only reliably observed source of digitally coded, functionally specific, complex information, is intelligently directed configuration, i.e. design. So, on the uniformity principle [roughly: like causes like, where we see characteristic signs], we have excellent reason to infer that DNA — which manifests just such dFSCI — is designed. Which, if acknowledged, would immediately devastate the whole Darwinian theoretical account of the origin of major body plans on undirected chance variation, natural selection and similar culling mechanisms, thence descent with modification deemed powerful enough to account for biodiversity from pond scum to us.}
First, we see how Wikipedia resorts to a No True Scotsman fallacy.
Regardless of qualifications, if one is not in agreement with the above asserted “consensus,” one is disenfranchised as a scientist. This is little more than name calling, and shows us an example of why it is fair to caution the user that Wikipedia is so often marred by bias. And, the appeal to consensus in science ignores the basic fact that science is inherently provisional, so it must be open to correction and progress, whether by theory refinement or by theory replacement. AKA, scientific revolutions.
Next, a slippery definition acts: observed minor changes in populations, sometimes called micro-evolution, are “evolution.” It is indeed reasonable to call such observed changes a fact: something we directly know per observation, is so and/or has occurred.
But, we are being told as well — and not only by so humble a source as Wikipedia but by the US National Academy of Science — that so is the UN-observed hypothesis that all forms of life and all body plans derive from a common ancestor, through descent with modification on chance variation, differential reproductive success and the like.
However, this is an assertion, not a demonstrated or observed reality. We have no right to infer or assume that the one simply accumulates into the other. Nor, are we even remotely capable of directly observing the remote past of origins, so we cannot know the proposed universal common descent for a fact. Instead, we can only observe evidence in the present, and infer and debate about alternative explanations and proposed timelines and mechanisms. But, plainly, an inferred explanation — however strongly we may wish to believe it to be true — is not and cannot be a fact.
Sadly, even more unsupportable claims have been made in recent textbooks. For example, in a recent blog post, Dr Cornelius Hunter pointed out on the discussion of Fig. 17.3 in the 5th Edn of the Johnson and Losos text, The Living World (McGraw Hill, 2008), that the authors claim:
It is important not to miss the key point of the result you see illustrated in figure 17.3: evolution is an observation, not a conclusion. Because the dating of the samples is independent of what the samples are like, successive change through time is a data statement. While the statement that evolution is the result of natural selection is a theory advanced by Darwin, the statement that macroevolution has occurred is a factual observation. [Emphases added.]
Hunter aptly rebuts:
A sequence of fossils is an observation of macroevolution? It would be difficult to imagine a more misleading statement than this. And it is not as though this was an unintended mistake that just happened to elude the 100+ reviewers. Johnson and Losos went out of their way to make and elaborate this message, and the army of evolutionist reviewers all nodded their heads. [Emphasis added.]
For, first, events claimed to have happened 50 – 35 million years in the deep past are simply not open to direct observation; as, we were not there to see for ourselves, nor do we have generally acceptable and credible record of the true facts from those who were.
Notwithstanding, the claim is being made that the dating is “independent” of the reconstructions and artistic photo-paintings made based on fossils recovered in certain layers of rocks. Not quite.
For, as Science writer Richard Milton has summarised , such dating schemes face several challenges [U/D, 01:23, link added], and again, are simply not direct observations of the remote, unobserved — and, credibly, unobservable — past:
[2 Ballpark thinking:] Any dating scientist who suggested looking outside of [the standard] ballpark . . . would be looked on as a crackpot by his colleagues. More significantly, he would not be able to get any funding for his research . . . .
[3 Intellectual phase-locking:] . . . all scientists make experimental errors that they have to correct. They naturally prefer to correct them in the direction of the currently accepted value thus giving an unconscious trend to measured values . . . .
[4 Conformity to consensus:] Take for example a rock sample from the late Cretaceous, a period which is universally believed to date from some 65 million years ago. Any dating scientist who obtained a date from the sample of, say, 10 million years or 150 million years, would not publish such a result because he or she will, quite sincerely, assume it was in error. On the other hand, any dating scientist who did obtain a date of 65 million years would hasten to publish it . . . [Shattering the Myths of Darwinism (Park Street Press, 1997), pp. 50 – 51. {UPDATE, 11:01:13: I here cite Milton as having made a cogent summary of the challenges faced by dating science, not as an endorsement of either his wider argument as a Neo-Lamarckian, or of his general views as an alternative science journalist. One may accept geo-dating results on the preponderance of evidence and argument [note the standard of warrant applied], but equally, one should also reckon with the sort of concerns on strengths and limitations as are cited. We should not mistake inherently and inescapably provisional results of a long chain of inferences within a school of thought for indisputable observed fact. HT: Bevets.} ]
Now, we may argue that, notwithstanding such concerns, there is a general consensus that the dating scientists are dating something real. But, that is also an inference, not a direct observation or record of it by a competent and credible eyewitness. Radioactive dates, index fossil dates and stratigraphic dates — however plausible they may seem to be — form a model timeline within the general origins science theoretical framework; they are not an extrinsic, independent cross-check on it. (And, a case like this one should give us pause before dismissing such a concern out of hand.)
Worse, as we have discussed recently here and here, the underlying context for all of this is the sort of imposed a priori evolutionary materialism that Lewontin so notably summarised in his 1997 NYRB article:
Such blatant question-begging should give us serious pause when we hear the ever so confident assertion that “Evolution is a FACT!”