The story, as usually told, goes something like this. Geology, for several centuries, had languished under the thrall of Archbishop Ussher and his biblical chronology of but a few thousand years for the earth’s age. This restriction of time led to the unscientific doctrine of catastrophism—the notion that miraculous upheavals and paroxysms must characterize our earth’s history if its entire geological story must be compressed into the Mosaic chronology. After long struggle, Hutton and Lyell won the day for science with their alternative idea of uniformitarianism—the claim that current rates of change, extrapolated over limitless time, can explain all our history from a scientific standpoint by direct observation of present processes and their results. Uniformity, so the story goes, rests on two propositions: essentially unlimited time (so that slow processes can achieve their accumulated effect), and an earth that does not alter its basic form and style of change throughout this vast time. Uniformity in geology led to evolution in biology and the scientific revolution spread. If we deny uniformity, the homily continues, we undermine science itself and plunge geology back into its own dark ages.
Yet Kelvin, perhaps unaware, attempted to undo this triumph of scientific geology. Arguing that the earth began as a molten body, and basing his calculation upon loss of heat from the earth’s interior (as measured, for example, in mines), Kelvin recognized that the earth’s solid surface could not be very old—probably 100 million years, and 400 million at most (although he later revised the estimate downward, possibly to only 20 million years). With so little time to harbor all of evolution—not to mention the physical history of solid rocks—what recourse did geology have except to its discredited idea of catastrophes? Kelvin had plunged geology into an inextricable dilemma while clothing it with all the prestige of quantitative physics, queen of the sciences. One popular geological textbook writes (C.W. Barnes, in bibliography), for example:
Geologic time, freed from the constraints of literal biblical interpretation, had become unlimited; the concepts of uniform change first suggested by Hutton now embraced the concept of the origin and evolution of life. Kelvin single-handedly destroyed, for a time, uniformitarian and evolutionary thought. Geologic time was still restricted because the laws of physics bound as tightly as biblical literalism ever had.
Fortunately for a scientific geology, Kelvin’s argument rested on a false premise—the assumption that the earth’s current heat is a residue of its original molten state and not a quantity constantly renewed. For if the earth continues to generate heat, then the current rate of loss cannot be used to infer an ancient condition. In fact, unbeknown to Kelvin, most of the earth’s internal heat is newly generated by the process of radioactive decay. However elegant his calculations, they were based on a false premise, and Kelvin’s argument collapsed with the discovery of radioactivity early in our century. Geologists should have trusted their own intuitions from the start and not bowed before the false lure of physics. In any case, uniformity finally won and scientific geology was restored. This transient episode teaches us that we must trust the careful empirical data of a profession and not rely too heavily on theoretical interventions from outside, whatever their apparent credentials.
So much for the heroic mythology. The actual story is by no means so simple or as easily given an evident moral interpretation. First of all, Kelvin’s arguments, although fatally flawed as outlined above, were neither so coarse nor as unacceptable to geologists as the usual story goes. Most geologists were inclined to treat them as a genuine reform of their profession until Kelvin got carried away with further restrictions upon his original estimate of 100 million years. Darwin’s strong opposition was a personal campaign based on his own extreme gradualism, not a consensus. Both Wallace and Huxley accepted Kelvin’s age and pronounced it consonant with evolution. Secondly, Kelvin’s reform did not plunge geology into an unscientific past, but presented instead a different scientific account based on another concept of history that may be more valid than the strict uniformitarianism preached by Lyell. Uniformitarianism, as advocated by Lyell, was a specific and restrictive theory of history, not (as often misunderstood) a general account of how science must operate. Kelvin had attacked a legitimate target.
KELVIN’S ARGUMENTS AND THE REACTION OF GEOLOGISTS
As codiscoverer of the second law of thermodynamics, Lord Kelvin based his arguments for the earth’s minimum age on the dissipation of the solar system’s original energy as heat. He advanced three distinct claims and tried to form a single quantitative estimate for the earth’s age by seeking agreement among them (see Joe D. Burchfield’s Lord Kelvin and the Age of the Earth, the source for most of the technical information reported here).
Kelvin based his first argument on the age of the sun. He imagined that the sun had formed through the falling together of smaller meteoric masses. As these meteors were drawn together by their mutual gravitational attraction, their potential energy was transformed into kinetic energy, which, upon collision, was finally converted into heat, causing the sun to shine. Kelvin felt that he could calculate the total potential energy in a mass of meteors equal to the sun’s bulk and, from this, obtain an estimate of the sun’s original heat. From this estimate, he could calculate a minimum age for the sun, assuming that it has been shining at its present intensity since the beginning. But this calculation was crucially dependent on a set of factors that Kelvin could not really estimate—including the original number of meteors and their original distance from each other—and he never ventured a precise figure for the sun’s age. He settled on a number between 100 and 500 million years as a best estimate, probably closer to the younger age.
Kelvin based his second argument on the probable age of the earth’s solid crust. He assumed that the earth had cooled from an originally molten state and that the heat now issuing from its mines recorded the same process of cooling that had caused the crust to solidify. If he could measure the rate of heat loss from the earth’s interior, he could reason back to a time when the earth must have contained enough heat to keep its globe entirely molten—assuming that this rate of dissipation had not changed through time. (This is the argument for his “brief” refutation of uniformity, cited at the beginning of this essay.) This argument sounds more “solid” than the first claim based on a hypothesis about how the sun formed. At least one can hope to measure directly its primary ingredient—the earth’s current loss of heat. But Kelvin’s second argument still depends upon several crucial and unprovable assumptions about the earth’s composition. To make his calculation work, Kelvin had to treat the earth as a body of virtually uniform composition that had solidified from the center outward and had been, at the time its crust formed, a solid sphere of similar temperature throughout. These restrictions also prevented Kelvin from assigning a definite age for the solidification of the earth’s crust. He ventured between 100 and 400 million years, again with a stated preference for the smaller figure.
Kelvin based his third argument on the earth’s shape as a spheroid flattened at the poles. He felt that he could relate this degree of polar shortening to the speed of the earth’s rotation when it formed in a molten state amenable to flattening. Now we know—and Kelvin knew also—that the earth’s rotation has been slowing down continually as a result of tidal friction. The earth rotated more rapidly when it first formed. Its current shape should therefore indicate its age. If the earth formed a long time ago, when rotation was quite rapid, it should now be very flat. If the earth is not so ancient, then it formed at a rate of rotation not so different from its current pace, and flattening should be less. Kelvin felt that the small degree of actual flattening indicated a relatively young age for the earth. Again, and for the third time, Kelvin based his argument upon so many improvable assumptions (about the earth’s uniform composition, for example) that he could not calculate a precise figure for the earth’s age.
Thus, although all three arguments had a quantitative patina, none was precise. All depended upon simplifying assumption
s that Kelvin could not justify. All therefore yielded only vague estimates with large margins of error. During most of Kelvin’s forty-year campaign, he usually cited a figure of 100 million years for the earth’s age—plenty of time, as it turned out, to satisfy nearly all geologists and biologists.
Darwin’s strenuous opposition to Kelvin is well recorded, and later commentators have assumed that he spoke for a troubled consensus. In fact, Darwin’s antipathy to Kelvin was idiosyncratic and based on the strong personal commitment to gradualism so characteristic of his world view. So wedded was Darwin to the virtual necessity of unlimited time as a prerequisite for evolution by natural selection that he invited readers to abandon The Origin of Species if they could not accept this premise: “He who can read Sir Charles Lyell’s grand work on the Principles of Geology, and yet does not admit how incomprehensively vast have been the past periods of time, may at once close this volume.” Here Darwin commits a fallacy of reasoning—the confusion of gradualism with natural selection—that characterized all his work and that inspired Huxley’s major criticism of the Origin: “You load yourself with an unnecessary difficulty in adopting Natura non facit saltum [Nature does not proceed by leaps] so unreservedly.” Still, Darwin cannot be entirely blamed, for Kelvin made the same error in arguing explicitly that his young age for the earth cast grave doubt upon natural selection as an evolutionary mechanism (while not arguing against evolution itself). Kelvin wrote:
The limitations of geological periods, imposed by physical science, cannot, of course, disprove the hypothesis of transmutation of species; but it does seem sufficient to disprove the doctrine that transmutation has taken place through “descent with modification by natural selection.”
Thus, Darwin continued to regard Kelvin’s calculation of the earth’s age as perhaps the gravest objection to his theory. He wrote to Wallace in 1869 that “Thomson’s [Lord Kelvin’s] views on the recent age of the world have been for some time one of my sorest troubles.” And, in 1871, in striking metaphor, “But then comes Sir W. Thomson like an odious spectre.” Although Darwin generally stuck to his guns and felt in his heart of hearts that something must be wrong with Kelvin’s calculations, he did finally compromise in the last edition of the Origin (1872), writing that more rapid changes on the early earth would have accelerated the pace of evolution, perhaps permitting all the changes we observe in Kelvin’s limited time:
It is, however, probable, as Sir William Thompson [sic] insists, that the world at a very early period was subjected to more rapid and violent changes in its physical conditions than those now occurring; and such changes would have tended to induce changes at a corresponding rate in the organisms which then existed.
Darwin’s distress was not shared by his two leading supporters in England, Wallace and Huxley. Wallace did not tie the action of natural selection to Darwin’s glacially slow time scale; he simply argued that if Kelvin limited the earth to 100 million years, then natural selection must operate at generally higher rates than we had previously imagined. “It is within that time [Kelvin’s 100 million years], therefore, that the whole series of geological changes, the origin and development of all forms of life, must be compressed.” In 1870, Wallace even proclaimed his happiness with a time scale of but 24 million years since the inception of our fossil record in the Cambrian explosion.
Huxley was even less troubled, especially since he had long argued that evolution might occur by saltation, as well as by slow natural selection. Huxley maintained that our conviction about the slothfulness of evolutionary change had been based on false and circular logic in the first place. We have no independent evidence for regarding evolution as slow; this impression was only an inference based on the assumed vast duration of fossil strata. If Kelvin now tells us that these strata were deposited in far less time, then our estimate of evolutionary rate must be revised correspondingly.
Biology takes her time from geology. The only reason we have for believing in the slow rate of the change in living forms is the fact that they persist through a series of deposits which, geology informs us, have taken a long while to make. If the geological clock is wrong, all the naturalist will have to do is to modify his notions of the rapidity of change accordingly.
Britain’s leading geologists tended to follow Wallace and Huxley rather than Darwin. They stated that Kelvin had performed a service for geology in challenging the virtual eternity of Lyell’s world and in “restraining the reckless drafts” that geologists so rashly make on the “bank of time,” in T.C. Chamberlin’s apt metaphor. Only late in his campaign, when Kelvin began to restrict his estimate from a vague and comfortable 100 million years (or perhaps a good deal more) to a more rigidly circumscribed 20 million years or so did geologists finally rebel. A. Geikie, who had been a staunch supporter of Kelvin, then wrote:
Geologists have not been slow to admit that they were in error in assuming that they had an eternity of past time for the evolution of the earth’s history. They have frankly acknowledged the validity of the physical arguments which go to place more or less definite limits to the antiquity of the earth. They were, on the whole, disposed to acquiesce in the allowance of 100 millions of years granted them by Lord Kelvin, for the transaction of the long cycles of geological history. But the physicists have been insatiable and inexorable. As remorseless as Lear’s daughters, they have cut down their grant of years by successive slices, until some of them have brought the number to something less than ten millions. In vain have geologists protested that there must be somewhere a flaw in a line of argument which tends to results so entirely at variance with the strong evidence for a higher antiquity.
KELVIN’S SCIENTIFIC CHALLENGE AND THE MULTIPLE MEANINGS OF UNIFORMITY
As a master of rhetoric, Charles Lyell did charge that anyone who challenged his uniformity might herald a reaction that would send geology back to its prescientific age of catastrophes. One meaning of uniformity did uphold the integrity of science in this sense—the claim that nature’s laws are constant in space and time, and that miraculous intervention to suspend these laws cannot be permitted as an agent of geological change. But uniformity, in this methodological meaning, was no longer an issue in Kelvin’s time, or even (at least in scientific circles) when Lyell first published his Principles of Geology in 1830. The scientific catastrophists (see essay 7) were not miracle mongers, but men who fully accepted the uniformity of natural law and sought to render earth history as a tale of natural calamities occurring infrequently on an ancient earth.
But uniformity also had a more restricted, substantive meaning for Lyell. He also used the term for a particular theory of earth history based on two questionable postulates: first, that rates of change did not vary much throughout time and that slow and current processes could therefore account for all geological phenomena in their accumulated impact; second, that the earth had always been about the same, and that its history had no direction, but represented a steady state of dynamically constant conditions.
Lyell, probably unconsciously, then performed a clever and invalid trick of argument. Uniformity had two distinct meanings—a methodological postulate about uniform laws, which all scientists had to accept in order to practice their profession, and a substantive claim of dubious validity about the actual history of the earth. By calling them both uniformity, and by showing that all scientists were uniformitarians in the first sense, Lyell also cleverly implied that, to be a scientist, one had to accept uniformity in its substantive meaning as well. Thus, the myth developed that any opposition to uniformity could only be a rearguard action against science itself—and the impression arose that if Kelvin was attacking the “doctrine of uniformity” in geology, he must represent the forces of reaction.
In fact, Kelvin fully accepted the uniformity of law and even based his calculations about heat loss upon it. He directed his attack against uniformity only upon the substantive (and dubious) side of Lyell’s vision. Kelvin advanced two complaints about this substantive meaning of u
niformity. First, on the question of rates. If the earth were substantially younger than Lyell and the strict uniformitarians believed, then modern, slow rates of change would not be sufficient to render its history. Early in its history, when the earth was hotter, causes must have been more energetic and intense. (This is the “compromise” position that Darwin finally adopted to explain faster rates of change early in the history of life.) Second, on the question of direction. If the earth began as a molten sphere and lost heat continually through time, then its history had a definite pattern and path of change. The earth had not been perennially the same, merely changing the position of its lands and seas in a never-ending dance leading nowhere. Its history followed a definite road, from a hot, energetic sphere to a cold, listless world that, eventually, would sustain life no longer. Kelvin fought, within a scientific context, for a short-term, directional history against Lyell’s vision of an essentially eternal steady-state. Our current view represents the triumph of neither vision, but a creative synthesis of both. Kelvin was both as right and as wrong as Lyell.
RADIOACTIVITY AND KELVIN’S DOWNFALL
Kelvin was surely correct in labeling as extreme Lyell’s vision of an earth in steady-state, going nowhere over untold ages. Yet, our modern time scale stands closer to Lyell’s concept of no appreciable limit than to Kelvin’s 100 million years and its consequent constraint on rates of change. The earth is 4.5 billion years old.
Lyell won this round of a complicated battle because Kelvin’s argument contained a fatal flaw. In this respect, the story as conventionally told has validity. Kelvin’s argument was not an inevitable and mathematically necessary set of claims. It rested upon a crucial and untested assumption that underlay all Kelvin’s calculations. Kelvin’s figures for heat loss could measure the earth’s age only if that heat represented an original quantity gradually dissipated through time—a clock ticking at a steady rate from its initial reservoir until its final exhaustion. But suppose that new heat is constantly created and that its current radiation from the earth reflects no original quantity, but a modern process of generation. Heat then ceases to be a gauge of age.