by Paul Braterman
Last November, creationist objectors in Texas tried yet again to sabotage the state's textbook adoption process. One of the objections concerned the age of the Earth, using the long refuted cooling argument that goes back to Kelvin in the 1860s. An online conversation about the matter directed me to the real flaw in Kelvin's reasoning, which is different from what I had believed (see my earlier posting). Further digging led me to the oldest formulation I know of Intelligent Design (ID) creationism (of course, it was not called that, but “Unsolved Problems of Science”. Now over a century old, it already shows the key features of “modern” ID, even down to the link with conservative politics, and the despicable misuse of fraudulently edited quotations.
Kelvin's reasoning was based on a very simple physical model, heat flow from a solid sphere initially at uniform high temperature. This model, and estimates of the rate of heat flow and temperature gradient, led him to assign a maximum age of a mere hundred million years, with the most probable age around a quarter of that. And yet the argument from radiometric dating, something with which Kelvin himself was never happy, gives overwhelming evidence to the contrary. Rutherford, and everyone else for decades afterwards, thought that Kelvin's error lay in the neglect of the heat generated within the earth by radioactive decay itself. Actually, it is a mistake to imagine that radioactive heating has all that much to do with it, and I must confess to having repeated this mistake many times in my own teaching and writing.
The real error (details here and in Pt I) had been pointed out a decade before Rutherford confronted Kelvin at the Royal Institution, and three years before radioactivity had even been discovered. Kelvin's calculation only considered heat transfer by conduction, whereas convection from depth is far more important. Convection can efficiently transport heat over long distances. It would have brought far more to the surface than Kelvin's model allowed for, meaning that it must have taken far longer to get rid of it. John Perry, one of Kelvin's own former students, was sure that Kelvin's estimate of the earth's age was far too low, suggested that Kelvin could have drastically underestimated the efficiency of heat transfer, and even suggested that the Earth's interior could be in a partly molten state, making convection possible.
Christ Church College. The future Lord Salisbury studied here, but did not sit his final examination.
In this piece, I want to talk about two things, how I learned the error of my ways, and exactly what it was that goaded Perry into an uncharacteristic public quarrel with his former mentor. I will also very briefly discuss the enormous importance of mantle convection for the present-day science of geology.
First, the question of John Perry's timing. If Kelvin had been promoting his cooling argument since 1862, why did no one query his physical assumptions until Perry did so in 1894? The answer, according to the historian Brian Shipley (now Canadian consul in Minneapolis), lies more in the domain of politics than of science.
In the late 19th century, the annual meeting of the British Association for the Advancement of Science was an event of major importance. Presidential addresses had been given by Kelvin, by T. H. Huxley, and in 1892 by Sir Archibald Geikie, the most eminent British geologist of the time. In his address, Geikie tactfully thanked Kelvin for bringing physics to bear on the problems of geology, and for pointing out that thismeant that the earth could not be indefintley old as the geologists of half a century earlier had imagined. Nonetheless, he insisted that the Earth had to be much older than Kelvin would admit, and that therefore there must be some flaw, yet to be discovered, in Kelvin's reasoning. In 1894, the Association met in Oxford. The Presidential Address was in part a response to Geikie. It was given by Lord Salisbury, a senior Conservative politician, who would be responsible for the Boer War among other things, and was the last Prime Minister to run his administrations from the House of Lords. Salisbury was extremely pessimistic by temperament, believing that “Whatever happens will be for the worse, and therefore it is in our interest that as little should happen as possible.” It is not surprising, then, that he was unsympathetic to the idea of evolution.
Some of Salisbury's remarks make one nostalgic for an earlier age, before the mid–20th century phenomenon of “scientific creationism”. “Few men”, he says, “are now influenced by the strange idea that questions of religious belief depend on the issues of physical research. Few men, whatever their creed, would now seek their geology in the books of their religion…”. Alas, this is exactly the way proponents of “creation science” would have us proceed, rejecting all evidence that cannot be fitted into a biblical framework. Salisbury also praises Darwin, and refers to the “lasting and unquestioned effect” of his work in disposing of the doctrine of the immutability of species, so that “Few are now found to doubt that animals separated by differences far exceeding those that distinguished what we know as species have yet descended from common ancestors.”
But it is all downhill from there.
Salisbury, caricatured by Spy for Vanity Fair, 1900
Salisbury accepts Kelvin's estimates for the rate at which the Earth is cooling, asserts that according to these estimates organic life could not have existed on the Earth as it was 100 million years ago, and contrasts this with the requirements of the geologists and biologists, and “the prodigality of ciphers [zeroes] which they put at the end of the earth's hypothetical life.” Natural selection demands well over a hundred million years; but according to the physicists, if the Earth even existed that far back in time it would have been so hot that all living things would have been vaporized. Faced with this disagreement, Salisbury proclaims himself neutral, but there is no doubt who he is neutral for.
He puts forward various arguments from ignorance, which had more weight then than when creationists repeat them now. He also puts forward plausible arguments based on the improbability of matings between animals carrying favoured new variations. Those arguments were wrong, but this was not shown convincingly until the 1920s, with the development of statistical genetics, so we can forgive him. He also links the popularity of evolutionary thinking to Darwin's own personality, and to disputes outside the domain of science itself, and predicts its decline. These themes, also, continue to resonate in the ID and other creationist literature. But, to be fair, they may have had more force at the time than they do now, more than a century later.
What cannot be forgiven, however, is Salisbury's editing, truncating, and reassembling quotations in such a way as to completely distort the meaning. This is a serious charge, and therefore requires detailed evidence. I can only apologise for the amount of space necessary for this.
Distortion provides the basis for Salisbury's most seemingly conclusive argument, which is philosophical, and hinges on an alleged quotation from a paper by the evolutionary scientist Weismann. The same Weismann to whom we owe our clear distinction between inheritance and development, genotype and phenotype. The relevant article, Contemporary Review, 1893, 64, 309-338, is titled “The All-Sufficiency of Natural Selection”, and Salisbury quotes from it as follows:
We accept natural selection, not because we are able to demonstrate the process in detail, not even because we can read more or less easily imagine it, but simply because we must – because it is the only possible explanation that we can conceive. We must assume natural selection to be the principle of the explanation of the metamorphoses, because all other apparent principles of explanation fail us, and it is inconceivable that there could yet be another capable of explaining the adaptation of organisms without assuming the help of a principle of design.
This, Salisbury would have us believe, demonstrates the ultimate bankruptcy of evolutionary theory. Invoking a principle of design is simply declared illegitimate, by arbitrary decree, so that natural selection can be accepted on faith. Wrenched from their context, Weismann's words are paraded as evidence that evolution rests on assumptions as arbitrary as those of any religion.
We are in the quote mine, territory all too familiar to students of creationism. Weismann is rebuked for dismissing “the help of a principle of design,” and thereby excluding the mood of explanation that now goes by the name of “Intelligent Design theory.” But that has absolutely nothing to do with what Weismann is actually talking about. Salisbury is neglecting context. Weismann is arguing, very specifically, with Herbert Spencer. What is at stake is not whether evolution occurs buthow it occurs, whether adaptations emerge within individuals through usage as organisms develop and live out their lives, or whether it occurs, as Weismann is correctly arguing, through selection between individuals.
Actually, it's worse. The two sentences that Salisbury runs together do not even belong together. Both sentences in the purported quotation come from Weismann's article. The text of the second sentence, slightly longer than Salisbury's version, is on p. 328. The first comes from considerably later on in the article, p.336, and the passage reads in full:
We accept it [natural selection], not because we are able to demonstrate the process in detail, not even because we can read more or less easily imagine it, but simply because we must – because it is the only possible explanation that we can conceive. For there are only two possible a priori explanations of adaptations for the naturalist – namely, the transmission of functional adaptations [as Weismann termed acquired characters] and natural selection; but as the first of these can be excluded, only the second remains.
In other words, Weismann and Spencer are already agreed on seeking naturalistic explanations, so that the question of a “principle of design” is not even under active consideration. Weismann, in the course of some 30 densely argued pages, is considering the competing explanations of evolved adaptedness, and makes his point by discussing how sterile worker insects come to be adapted to their way of life. He argues that only two such explanations are possible, namely the inheritance of acquired characters, and natural selection. But sterile workers cannot transmit acquired characters to their offspring, because they have none. Therefore the correct explanation must be natural selection; in this case selection for the ability to give rise to effective worker offspring. Weismann is not invoking extra-scientific arguments to exclude design; he is talking about something different altogether. Salisbury's reading of Weismann is a contrived and unnatural distortion, made possible only by juxtaposing sentences that do not even belong on the same page, and in the wrong order to boot.
Having misrepresented Weismann as an unwitting ally, Salisbury concludes his attack on evolution by natural selection with a quotation from Kelvin's presidential address of 1871:
I have always felt that the hypothesis of natural selection does not contain the true theory of evolution, if evolution has been in biology…. I feel profoundly convinced that the argument of design has been greatly too much lost sight of in recent zoological speculations. Overpoweringly strong proofs of intelligent and benevolent design lie around us, and if ever perplexities, whether metaphysical or scientific, turn us away from them for a time, they come back upon us with irresistible force, showing us through Nature the influence of a free will, and teaching us that all living things depend on one everlasting [ever-acting?] Creator and Ruler.
I very much doubt if Perry had been following the lengthy and complex debate between Spencer and Weismann on the mechanism of evolution, but he would certainly have known of the various methods, such as the thickness of sediments, that geologists at that time were using to estimate the age of the Earth. To him, Kelvin's much less generous estimate was the anomaly, and, when this estimate was invoked by someone of Salisbury's eminence for semi-religious or even political purposes, that was enough to goad him into action.
I found out about this all rather indirectly. As I admitted in my last post, one of my many vices is commenting in online forums. So when Jerry Coyne's website, Why Evolution Is True, asked what cooling had to do with last November's Texas textbook adoption drama, I wrote a comment that mentioned Kelvin and Rutherford. I also gave there, as an example of creationist absurdity, an alleged repudiation of Rutherford in a 1978 Institute for Creation Research pamphlet by Harold Slusher and T. P. Gamwell, cited by Bob Jones University. Just to be safe, I did a Google search on Slusher and Gamwell, and the only appraisal I could find dismissed their work, on the grounds that it had considered diffusion of heat across a plane surface, rather than the surface of a sphere.
Reality, as always, is more interesting. My comment got a rejoinder from one Robert Seidel, who pointed out to me the paper by Englander and colleagues that I discussed on my last post. This, based on current estimates of radioactive heating, actually confirms Slusher and Gamwell's conclusion, and in correspondence Englander told me that the difference between a plane and a sphere is quite unimportant in this context, since conductive cooling would only have penetrated in 5 billion years to one tenth of the Earth's radius. But none of this really matters in the great scheme of things, because Kelvin's argument, as we saw before, is swept aside by mantle convection.
Conclusions: never take a quotation from a real scientist in a creationist text at face value. Strange things happen in the quote mine, as we have seen. But on the other hand, just because an argument refutes creationism, that doesn't always mean it's right. Just because an argument is used to support creationism, that doesn't necessarily mean it's wrong; quite probably, as in the Slusher and Gamwell case, it's simply irrelevant. And just because something is said by creationists, that doesn't mean there's nothing to be learned from it. “Learn even from your enemies”, said Ovid, and he was right.
A quick afterword (a complete afterword would invoke a large part of all geological studies for the past 40 years): by around 1930 it had become clear, from how the shockwaves caused by earthquakes travel, that the solid crust of the Earth floats on a viscous fluid, the mantle. Arthur Holmes, the most farsighted physical geologist of his generation, realised that this would imply mantle convection. He spelt out some of the implications of this in a remarkable paper in Transactions of the Geological Society of Glasgow. Heat would build up beneath over-large continents, attracting upward convective flow which would eventually tear them apart. New basaltic crust would appear at such separation zones, while old crust would disappear in regions of downward flow. Thus for the continents to move it it would not be necessary for them to force their way through basement rock, an obvious physical impossibility, but merely to ride on that moving basement like luggage on a conveyor belt. To use the language adopted a full generation later, when these ideas at last gained general acceptance, Holmes was describing rifting, subduction, and plate tectonics.
I thank Philip Englander, Philip Kitcher, Peter Molnar, Brian Shipley, and one Robert Seidel (if you see this, Robert, please identify yourself so I'll know just who to thank) for helpful discussions. I wrote to Prof Dan Olinger of Bob Jones University regarding his University's portrayal of the Earth's age in December 2013. He assured me that his colleagues would reply to my questions, but they have not yet done so.