|Jet in Carina WFC3 IR [Courtesy NASA]
Is evolution just a "theory"?
David H. Bailey
Updated 1 January 2020 (c) 2020
What is a "theory"?
Merriam-Webster's Dictionary lists several definitions for the word "theory" [Merriam2009]. The two that are the most relevant here are: (a) "a plausible or scientifically acceptable general principle or body of principles offered to explain phenomena, e.g., the wave theory of light" and (b) "a hypothesis assumed for the sake of argument or investigation; an unproved assumption." In most scientific discourse, scientists use definition (a), while in popular public discourse, definition (b) is more widely assumed. This distinction is the root of the widespread misunderstanding of the phrase "theory of evolution." Evolution is not termed a "theory" because it is a sketchy conjecture that has never been seriously tested. To the contrary, evolution has passed more than a full century of withering empirical tests. It is termed a "theory" in the same sense that one refers to "atomic theory" or "theory of relativity" or "theory of equations" -- not because of any credible doubt of the correctness of the main points of the theory, but instead because it is a general principle with substantial explanatory power (and falsifiability) that has withstood very rigorous scrutiny.
On the other hand, many scientists are content to live with the double meaning for evolution as well as other firmly established theories, as a form of self-imposed humility and resistance against taking any theory as absolutely unchangeable truth. It is undeniably true that science is "forever tentative," as philosopher Karl Popper once wrote [Popper1959]. The tentative nature of scientific theories was impressed on scientists most vividly in the early 20th century, when Newton's classical laws of motion and gravitation, which had dominated scientific research for more than three centuries, were displaced by Einstein's relativity (for objects traveling at very high speeds) and by quantum mechanics (for very small objects, such as atoms and subatomic particles). But in mentioning these examples, it must be kept firmly in mind that Newton's laws to this day remain very accurate approximations of reality -- for ordinary speeds and ordinary-sized objects, the differences between the predictions of Newton's laws and the more modern theories are so small that only the most precise laboratory equipment can measure any difference whatsoever. Certainly it is true that Newton's laws remain the mainstay of almost all practical science and engineering calculations. In other words, even "falsified" theories often remain extremely accurate approximations that are valid in a wide range of appropriate arenas.
For more discussion on the nature of science and postmodern critiques of science, see
What is science? and
Does evolution qualify as a scientific theory?
Merriam-Webster defines "evolution" as "a theory that the various types of animals and plants have their origin in other preexisting types and that the distinguishable differences are due to modifications in successive generations." [Merriam2009]. Many biologists would further assert that all life forms in fact have a single common ancestor, since there is very strong evidence to support this assertion, but we need not formally include this in our definition of evolution.
Some critics have asserted that evolution, as defined above, is not a truly falsifiable theory. But there any number of findings that could upset the theory. For example, if scientists found rabbit fossils in the Cambrian (550 million years ago, approximately 500 million years before the emergence of modern mammals), this would be a major crisis for the field. An equally serious crisis would be the discovery of some set of genes in one organism that is so utterly distinct from corresponding genes in a related organism that they are not plausibly connected by a reasonable number of adaptations and mutations. Another crisis would occur if scientists identified some fundamental principle of physics or information theory that forbids the natural, unassisted formation of highly complex biological structures. But no such discoveries have been made. In other words, evolution definitely qualifies as a falsifiable and rigorously tested hypothesis. For additional discussion, see Falsifiable.
Evidence in support of evolution
The principal lines of evidence supporting evolution can be briefly summarized as follows:
For additional details, see
- Geological evidence: The fact that the various geological eras, as identified the fossils they contain, always appear in the same order (except in a few cases where there is clear evidence of overthrusting) and yield the same geological dates (as measured by highly refined and reliable radiometric techniques), no matter where they are unearthed.
- Fossil evidence: The record of many prehistoric species, preserved in stone, that have populated Earth over eons of geologic time.
- Morphological evidence: Similarities that have been found between the physical structure and function of existing biological species.
- Novel structures: Examples of novel biological structures formed by mutations and related evolutionary processes, documented in laboratory experiments and field studies.
- Fast evolution: Examples of rapid evolution, wherein species have changed or have subdivided into separate species over short periods of time (a few days, months, years or decades).
- Protein sequences: Analyses of common protein chains, such as alpha globin or cyctochrome C, across many species.
- DNA evidence: Direct analysis of DNA sequences (either whole or parts) made possible by the recent dramatic advances in DNA sequencing technology.
Any one of these lines of evidence constitutes very strong evidence for evolution; together they are as unassailable as any theory of science. It might be mentioned that several of these lines of evidence, notably radiometric dating, comparisons of protein sequences and analysis of DNA, were completely unforeseen at the time of Charles Darwin. Doubtless he would have been astounded that his theory could be analyzed by such powerful tools.
Does the fact that evolution occurred in the past affect its status as a scientific theory?
Some creationists and others have argued that evolution is a "theory" because no one was around to make an eyewitness record when the Earth was created, and thus any scientific reconstruction of those events and processes can never be "proven." Creationist Ken Ham, for instance, argues that there is no proof of the Big Bang theory, for instance, because scientists were not present to observe it [Ham2011]. But it must be kept in mind that many other aspects of our physical world are truly beyond the realms of our senses:
In other words, each of the above aspects of our physical world can only be studied via empirical evidence that is certainly very far from anything that we can directly sense. Yet few, if any, persons nowadays seriously doubt that the sun and the planets of our solar system really do exist at roughly the distances scientists claim them to be, or that stars and galaxies populate the universe thousands or millions of light-years away, or that there really are structures known as molecules, atoms and nuclear particles, with roughly the sizes and possessing the properties that scientists assert for them. So why is the study of the distant past in any way fundamentally different?
- The planets and moons of our solar system are much too far away for humans to examine first-hand, although humans may travel to Mars in 20 years or so. And even with the space trips that have been made (e.g., the Apollo missions to the moon), the public must rely on scientific reports, since only a tiny number of humans have made these trips first-hand. Along this line, fully 6% of the U.S. population believes NASA faked the Apollo moon landings [Griggs2009].
- For distant stars and their planets, we rely completely on powerful telescopes and exotic techniques such as measuring subtle changes in the light from a distant star (indicating that a planet has passed between us and the star). Exploration by humans or even robot spacecraft to star systems outside our own solar system is still decades away. And galaxies are so much further away that there is, at present, no foreseeable technology to permit humans or our spacecraft to study these objects close-up and first-hand.
- Atoms and molecules are much, much too small for any human to examine first-hand. Thus we rely on exotic equipment such as electron microscopes and atomic-force microscopes to "see" them. And even these images are merely synthetic aids to visualization, generated say by exquisitely small variations in electric current as a probe passes over the surface. In other words, we "see" these structures only in a very indirect way.
- Chemical reactions cannot yet be observed, at least not in any way that truly displays what is actually happening. Thus the entire field of chemistry relies on indirect laboratory experiments, from which we infer what is happening at the molecular and atomic level.
- The nucleus of an atom is far, far smaller than anything that can be seen, even with exotic equipment such as atomic-force microscopes. Nuclear reactions can only be studied effectively using very powerful, large-scale equipment, whose operation is based on highly mathematical physical theories.
Is there a time machine to study the distant past?
In fact, there is a solid connection between the distances measured by astronomers to stars and galaxies, and the multi-million-year ages asserted by geologists for the fossil layers. This is because viewing distant stars and galaxies is, in effect, a "time machine." For example, when we view the Pinwheel Galaxy, which is 21 million light-years away, the image we see is a record of events that occurred 21 million years ago.
In August 2011, a Type 1A supernova was observed in the Pinwheel Galaxy. It was first discovered by Lawrence Berkeley Laboratory scientist Peter Nugent, using the Palomar Transient Factory (PTF), a remotely-controlled telescope facility near San Diego, California, just 11 hours after it exploded, so scientists were able to study its behavior in unprecedented detail. For example, the data was sufficiently strong to rule out several possibilities, leaving only the explosion of a carbon-oxygen white dwarf as the cause. What's more, careful measurements as this supernova unfolded, such as the intensity and light spectra of elements such as radioactive nickel decaying to cobalt, revealed that the fundamental physical laws in play when this supernova exploded 21 million years ago are indistinguishable from those laws we measure in Earth-bound laboratories today. Among these laws are the laws of radioactive decay, as well as fundamental quantum mechanics and general relativity [Preuss2011]. For these and other reasons, scientists have very good reasons to be entirely confident in the established old-earth picture of geology and evolution. For additional discussion, see Reliability and Time machine.
The reliability of human "eyewitness" evidence
It should be added that human eyewitness accounts are often inaccurate and misleading, as any courtroom lawyer or judge will attest. According to a November 2011 New York Times article, hundreds of studies have documented how human memories can be distorted. For example, in one study, participants viewed films of fender-benders in which no windows or headlights were broken. Later, if they were asked how fast the cars were going when they "smashed" into each other, they were more likely to describe shattered glass (that they never saw!) than when the word "hit" was used instead. In another study, participants were four times more likely to report a memory of a nonexistent event, if they had been asked to imagine such an event one week before [Beil2011].
In a remarkable study at the University of Illinois, subjects were instructed to carefully watch a 25-second video of students with basketballs and count the number of times balls are passed from one person to another. In this video, there is at least one "surprise" -- for several seconds a person in a gorilla suit strolls through the group of students, turns to the camera, thumps his chest, and then exits. Incredibly, the majority of subjects watching the video were so focused on the counting task that they failed to notice the gorilla! There are other more subtle surprises that almost everyone misses [Else2010; Simons1999].
In short, it is not at all clear that not being able to witness some event (such as the splitting of a particular species in the distant past) with our eyes is a significant disadvantage -- scientific instruments and computers are more reliable.
In summary, evolution is indeed a "theory" in the scientific sense (a scientific notion that has been studied in great detail and has survived many rigorous tests), but certainly not in the colloquial sense (a vague hypothetical guess, not yet really tested). Thus creationists and religious fundamentalists who promote the line that evolution is "only a theory" are either completely uniformed on the true nature of science, or else are being deliberately disingenuous to their audience.
For additional discussion on the nature of science, see
What is science?. For detailed discussion of evolution and the evidence that supports it, see