One tactic often used by those seeking to discredit evolutionary science is to label it as “historical science.” They claim historical sciences make no testable predictions and can only make educated guesses about the past, since the past is not directly observable and is not repeatable. Other sciences (sometimes called “operational” or “observational” or “experimental”) are purported to be more reliable because they involve observing things in the present and can directly test their hypotheses. This distinction allows people to affirm science insofar as it is responsible for technology and modern medicine, while rejecting the science that has established the ancient age of the earth and the reality of evolution.
Is there any merit to distinguishing historical science from other kinds of science?
First note that the term “historical science” does not refer to an object like the moon or the Eiffel Tower or Charles Darwin. It is a label given to a set of activities that supposedly have some important characteristic in common. There are other ways of grouping those activities that appeal to different characteristics and use different labels. For example, we can talk about physical sciences vs. life sciences, or empirical sciences vs. theoretical sciences, or natural sciences vs. social sciences (to name just a few). These are useful distinctions in their contexts, but the fact that there are so many different ways of categorizing science should caution us from thinking that any of them is the definitive way to do so.
So this question is not about whether we can legitimately call some activities “historical sciences” in certain contexts. It may be useful to do so. But what is at stake in this question is whether science can reliably tell us things about the past.
Reasoning about the past
Let’s start by considering an ordinary, everyday example of reasoning about a past event by someone who did not observe it. Upon returning home from work one day, Sarah observes that her son’s bicycle is missing from the garage. There are two explanations for this that spring to mind immediately: her son took the bicycle and went to his friend’s house, or the bicycle was stolen. These two explanations both fit with the immediate data and cause Sarah to look into things more carefully. She reasons that if he left to go somewhere else, he would have texted to let her know, so that makes her lean toward the theft option. But the doors and windows of the garage were all locked, and there is no sign of forced entry (and nothing else seems to be missing). So now she wonders whether her son just forgot to text. But he’s never done that before, so which explanation is more plausible? She needs more data to decide between these options. So she reasons that if he went somewhere but forgot to text, he won’t be in the house. Here is a prediction that can be tested. Sarah goes into the house and calls her son’s name, but gets no answer. Does that settle things? Maybe not; sometimes when she calls for him he doesn’t respond because he has his headphones on in another room. Still more data is needed. Then, unexpectedly, Sarah finds his cellphone on the counter next to a note that reads, “I went to Michael’s. I’ll be home for supper. Sorry, my phone died.” Now all the observable facts fit into nice theory (which she’s had to alter slightly, since it appears he didn’t forget to call). Does this prove that her son took his bike from the garage to go to his friend’s? No, it’s still consistent with the facts that the bike was stolen and he just walked to his friend’s house. And further observations might possibly alter the theory again (if he doesn’t come home for supper, perhaps she starts to worry that the bike was stolen, and her son was kidnapped and forced to write the note!). But given the current evidence, it seems reasonable to conclude that the best explanation for what was observed is that he took his bike to Michael’s house. And it is not too far-fetched to postulate other evidence in this hypothetical example that could make one of the explanations virtually certain.
This story of reasoning about the past might not qualify as science. But if the observations are more precise and systematic, and if they are about aspects of the natural world, it starts to sound a lot like the process of doing “historical science.”
Forensic science provides clear examples of reasoning to conclusions about the past by generating testable predictions from observations in the present. DNA evidence has become particularly persuasive in reconstructing the past to convict or exonerate suspects of horrific crimes.
Observation and prediction
Like the examples of the missing bike and forensic detectives, the theory of evolution begins with observations. Darwin introduces his work in The Origin of Species as follows:
When on board H.M.S. ‘Beagle,’ as naturalist, I was much struck with certain facts in the distribution of the organic beings inhabiting South America, and in the geological relations of the present to the past inhabitants of that continent. These facts, as will be seen in the latter chapters of this volume, seemed to throw some light on the origin of species—that mystery of mysteries, as it has been called by one of our greatest philosophers. On my return home, it occurred to me, in 1837, that something might perhaps be made out on this question by patiently accumulating and reflecting on all sorts of facts which could possibly have any bearing on it. After five years’ work I allowed myself to speculate on the subject, and drew up some short notes; these I enlarged in 1844 into a sketch of the conclusions, which then seemed to me probable: from that period to the present day I have steadily pursued the same object.
Based on his observations, Darwin put forth a number of hypotheses about past events in various fields of study, and subjected them to empirical tests. For example, prior to the publication of his Origin, he worked for years on barnacles to test the implications of his hypothesis for things like variation within a species and vestigial traits. And one of his most famous predictions, which was not resolved in his lifetime, was that marine mammals like whales and dolphins must have evolved from land-dwelling mammals that gradually adapted to life in the water. From fossil discoveries in the last few decades, there is now impressive evidence that confirms his prediction.
Another stunning example of the predictive power of evolutionary explanations for events in the very distant past is the discovery of the Tiktaalik fossil. Based on known fossils and their dates, according to the theory of evolution, land animals evolved from aquatic animals some 375 million years ago. That generates the prediction that we might be able to find transitional fossils if there are layers of rock from that era that formed under suitable conditions for preserving the life forms then. One of the only places on earth where such rock layers are accessible to human researchers is extreme northern Canada. In the late 1990s an expedition to the area was organized specifically to hunt for transitional fossils, and in 2004 three Tiktaalik fossils were found, which showed clear intermediate features between fish and amphibians. This provided strong evidence that our basic understanding of evolutionary theory and earth’s natural history is correct. It is worth watching this PBS video series that documents the discovery.
Today, DNA provides a powerful way to test and confirm the predictions of evolutionary hypotheses. By comparing the DNA of two currently existing species, scientists can make a prediction of when their most recent common ancestor existed, and researchers can go out into the field to look for fossils from that time period, similar to the Tiktaalik discovery. Or the predictions may come from observations of fossils and comparative morphology of species to make a specific tree of common ancestry for certain species. There had long been discussion about which of the great apes were more closely related to humans based on morphology. Being able to compare specific mutations in DNA of the species resolved this beyond reasonable doubt.
Limitations of Science and Natural History
Just as there are limitations to what science can explain in the present, so too we shouldn’t expect that science can uncover and explain everything that has happened in the past. Science is particularly well suited to uncover general patterns or law-like behavior in the natural world; that is what we call the natural history of the world. But we also believe that God can act miraculously in history and that miracles might leave behind evidence that doesn’t fit with scientific expectation (for example, the resurrection of Christ). The belief that God has sometimes acted miraculously does not prevent us from applying scientific methods to understanding natural history. It just means that natural history does not tell the whole story of what God has done. For much of the history of the universe, though, astronomers, geologists, and biologists have found that God’s creative activity can be discovered and described by science.
Science is a process by which people gain knowledge and can employ a variety of different methods. Read More >
(Video #5 in the BioLogos Basics video series.) A central feature of young earth creationism is the claim that the earth and universe were created no more than 10,000 years ago. ... Read More >
Many different and complementary scientific measurements have established with near certainty that the universe and the Earth are billions of years old. Read More >
Hypotheses about the past can generate predictions that can be tested against observations in the present. And conversely, present-day observations and experiments spur new hypotheses... Read More >
The geological principle of uniformitarianism does not necessarily conflict with a faithful interpretation of scripture. Read More >
One really powerful piece of evidence for the old age of the universe consists of the leftovers of dwarf galaxies that were consumed by our Milky Way a long time ago. Read More >
Science goes where the data leads, flood geology does not. Read More >
First published on December 1, 2017.