How many times do I have to criticize attacks on scientism, all of which use various permutations of the same three claims? Here they are:
1.) There are “other ways of knowing” that don’t involve science. These often involve “why” questions, like “Why am I here? (i.e. what is my purpose?)” or “Why is the universe here”?
2.) The scientific method (or rather, the use of empirical analysis and observation, confirmation, testing, making predictions, and so on) cannot be justified a priori by philosophy, and involves untestable or fallacious assumptions.
3.) Science is trying to take over the humanities, and this unwarranted extension of science to places where it doesn’t belong is true scientism.
The article below that just appeared in Quillette rehashes the same tired old arguments, and I’m tired of refuting them. But I’ll take up the cudgels once again. To see my numerous and previous criticisms of scientism, go here.
Read the article by clicking on the screenshot below. It’s a criticism of an earlier piece by Bo and Ben Winegard (also in Quillette) called “In defense of scientism.”

The nice article by the Winegards uses a narrow definition of scientism: “science based social policy” (SBSP) which they say is “the view that social policy should be based on the best available theory and data; in other words, that social policy should be decided using the weight of the evidence. And that is all scientism is—the view that scientific attitudes and methods can enhance all modes of empirical inquiry and should, therefore, be promoted.” Well, some would disagree with that. Others define scientism as “the extension of science beyond its proper bailiwick.” Two such construals of that are 1.) The claim that science devalues nonscientific realms like art and literature, and 2.) Science tries to construct an objective morality, saying that empirical investigation alone can tell us what and what is not desirable and good.
The Winegards address both of these construals, admitting that science can’t replace art and literature, which convey emotion and experience rather than empirical truth. But they also argue—and I agree—that science can helpfully infuse areas like sociology, literary criticism, and other areas that try to make claims about reality. And they agree with me that no, science can’t tell us objectively what is good and moral and desirable, for those are subjective preferences. But once you agree on those preferences—and in morality many of us do adopt similar consequentialist views—then science can tell us how to best achieve them; for how to achieve a desired goal is an empirical matter.
But Aaron Neil, a researcher at the Canadian think tank Cardus, wants to go beyond what the Winegards say, in particular conveying the first two tropes given at the top. I’ve refuted both of these claims before, most extensively in my book Faith Versus Fact, but I’ll try again, and will also try—and probably fail—to be brief. My refutations of claims 1 and 2 as emitted by Neil are these:
1.) Neil fails to tell us a single bit of knowledge that wasn’t derived by science, though he bloviates at length about how this is possible. If there are “other ways of knowing”, what is the knowledge produced by those ways? If it’s so pervasive, Neil should be able to give us many examples. But he fails miserably, coming up dry.
2.) Indeed, you can’t justify philosophically the use of the empirical method to produce truths about the universe. But the justification is not by philosophy, but by usefulness. In other words, the scientific method works to tell us truths about the universe, and some version of it (the varieties of empirical methods that I call “science construed broadly”) are the only way to find out facts. To paraphrase theology, we justify science by works rather than faith.
Let’s take Neil’s two claims in reverse order (his quotations are indented):
1.) The scientific method can’t be justified by philosophy, and involves untestable (and sometimes failed) assumptions. Here are some of the philosophical attacks on science leveled by Neil:
a. Science is self-refuting in saying that “scientific truth is always provisional.” To wit:
Although the Winegards present an innocuous definition in their essay, they commonly drift into the less benign form of scientism identified by Hayek. The Winegards’ Hayekian scientism manifests itself early in their piece with the claim that “Truth is always provisional.” As they correctly note, scientific “truths” appear to be true so long as they provide “the best available theory” based on the evidence at hand. However, not all truths bear this hypothetical quality. Ironically, the very statement, “Truth is always provisional” is not itself a provisional truth claim. If it is always true that truth is always provisional, this statement is self-refuting. Not all truth claims are theoretical statements that are vulnerable to empirical falsification. Take the proposition, “there are no square circles.” This is not a hypothesis that is true so long as scientists do not discover a square circle. Logically, a circle can never be a square.
Do I need to waste time on this? It’s philosophical pilpul. Instead of saying “truth is always provisional”, let’s restate it as “science doesn’t tell us anything with absolute certainty, but we have degrees of certainty about various things, and are more confident about some scientific truths than others.” That revision is sufficient to refute Neil’s philosophical twiddling.
b. Science is based on untestable metaphysical assumptions. To wit:
. . . as the greatest critics and advocates of modern science have argued, science is full of extra-scientific assumptions.
Friedrich Nietzsche, for example, saw that far from doing away with faith and metaphysics, the scientific enterprise of the “godless anti-metaphysicians” rested upon its own “metaphysical faith.” In The Gay Science, Nietzsche explains that science depends on dispelling personal convictions and replacing them with provisional hypotheses. However, Nietzsche argues, the scientific attempt to disallow a priori convictions is itself based on “some prior conviction…one that is so commanding and unconditional that it sacrifices all other convictions to itself.” For scientific inquiry to occur, the conviction must “be affirmed in advance” that “‘Nothing is needed more than truth.’” Implicit in the modern “scientific spirit” is the metaphysical belief that “truth is divine.” Therefore, he argues, “there is simply no science ‘without presuppositions.’”
If Nietzsche provides an example of a moral assumption implicit in the scientific method, David Hume, the great skeptic and pioneer of the modern empirical project, provides a philosophical one. For Hume, “all inferences from experience suppose that the future will resemble the past.” To observe that a cause follows from an effect, and to conclude that the same effect will always follow from the same cause, assumes that nature remains the same. This assumption is impossible to prove. “It is impossible,” writes Hume, “that any arguments from experience can prove this resemblance of the past to the future; since all these arguments are founded on the supposition of that resemblance.” In other words, arguing for uniformity in nature based on experiences assumes that uniformity already exists. To prove the consistency of the causal relationship would require stepping outside of empirical experience.
Again, this is easily refuted. First, not all scientists think that the pursuit of truth is the most important thing they do. But as scientists it is our job to pursue truth and that’s all, though many of us do value truth above nearly everything. (For example, many of us would prefer to be told we have a fatal disease than be lied to.) Further, if you are interested in solving problems like “How can we stem global warming?” or “How can we quash this outbreak of measles?”, then science is the only way to go. The so-called a priori assumption that the empirical method is the best way to find truth is not a prior conviction, but the result of centuries of experience of what works and what doesn’t.
As far as Hume is concerned, science does not assume that the future will resemble the past. Indeed, evolutionary biologists assume that the future will NOT resemble the past. The resemblance we do see, and this is not an a priori conviction but also the result of experience, is that the laws of physics appear to remain unalterable in our universe, so in that sense future laws and events (i.e., as instantiated in the evolution of stars) can be assumed to resemble the past laws and events. We use uniformitarianism insofar as our experience tells us this applies. We do not assume it a priori.
It’s a common mistake of people like Neil to think that scientists once sat down and constructed a scientific method, complete with dictums like “value truth above all else”, “assume the future will be like the past”, and “empirical investigation, replication, and so on are the best ways to find empirical truth”. No, those procedures developed from experience when people learned about the best ways to find truth.
2.) There are other ways of knowing. I discuss this at length in Faith Versus Fact, concluding that if you want to know facts about our universe, the scientific procedure (“science construed broadly”) is the only way to proceed. This does not denigrate philosophy or mathematics, both of which are logical systems that are very important in doing science and in thinking hard about what you’re doing. Mathematics does not tell us truths about the universe, but truths about the logical system it comprises. That’s why we can sensibly speak of “proof” in mathematics but not in science. Philosophy, by teaching us how to think clearly and logically, can point out errors in our thinking and lead us to conclusions that aren’t obvious. One of them is the Euthyphro Issue, which teaches us that most religious people get their morality not from religion itself, but from secular and extra-scriptural sources. That is not a truth about the universe, but a logical (and valuable!) truth that comes from reflection and perhaps some observation of how people construe morality (the border between this kind of philosophy and science is very tenuous).
What are the other ways of knowing? Neil discusses two areas.
a. Ways to answer “why” questions. To wit:
A notable example of this scientistic shift from method into metaphysics comes from Richard Dawkins and Peter Atkins who, like Dawkins, is a prolific author as well as a fellow at Lincoln College, Oxford University. During the question and answer period following a discussion of The God Delusion, Dawkins was asked whether science provides the answers to the great existential ‘why’ questions. In his reply, Dawkins declared that questions like “why does the universe exist” are “silly” questions that do not deserve answers. Peter Atkins makes a similar point in a recent article. He argues that questions like “Why are we here?” are “not real questions because they are not based on evidence.” Real questions, according to Atkins, are questions “open to scientific elucidation.”
Unfortunately, for Dawkins and Atkins, the belief that all questions must be open to scientific explanation is a metaphysical commitment, not a scientific one. Science does not say that only scientific questions are worth pursuing. Nor does science say that every aspect of reality can be explained by science. Lurking beneath their rejection of the non-scientific lies a fundamentally extra-scientific worldview. In their dismissal of the deepest questions concerning human existence, Dawkins and Atkins speak not as dispassionate scientists, but as partisans to their own philosophical picture of reality.
This can be dispelled easily when we realize that what Dawkins and Atkins (both scientists) are talking about as “fake questions” are “questions that cannot be answered with any certainty.” They are construing “real questions” as questions that have answers that we can all agree on, and can have some certainty about the answers. And for those kinds of questions, Dawkins and Atkins are correct, for only science can answer questions that have answers like that. Sure, you can say, “I had a vision of Jesus,” but there is no way to verify it. Beyond this kind of subjective “truth”, we must turn to science. If questions like “Why are we here?” do have “correct” answers, or answers that most of us can agree on, then pray tell us what the answers are, Dr. Neil! For religions, which occupy themselves with such questions, cannot agree of any answers.
b. Areas that constitute “other ways of knowing.” Disturbingly, besides philosophy, which I’ve already discussed, the only field mentioned by Neil is theology. (I’m not sure whether he thinks theology can really tell us any truths.):
Science is not the only form of knowledge. There are valid non-scientific ways of approaching reality. In fact, before the empirical science of Isaac Newton and Francis Bacon, science (from the Latin scientia) simply meant “knowledge.” For the ancients, natural philosophy (the rough pre-modern equivalent to modern science) and philosophy were ‘sciences’ because each intellectual discipline contributed towards knowledge of reality. Not only were philosophy and theology considered legitimate ways of knowing, the medievals placed natural philosophy below philosophy and theology. It may be tempting to dismiss the medieval hierarchy as an example of pre-modern ignorance. Before too quickly discounting it, consider first the following explanation behind the ordering provided by Thomas Aquinas, one of the greatest theologian-philosophers of the Middle Ages: “Lower sciences,” Aquinas writes, “presuppose conclusions proved in the higher sciences.”
And that’s about it: philosophy and theology (“natural philosophy” is just another word for “science”). Neil doesn’t mention literature or art or music or any of the other classic but bogus “ways of knowing.” (I’m not denigrating these areas, but claiming, as I did in my book, that they are ways of feeling rather than ways of knowing.)
Again, it’s extremely telling that despite Neil’s repeated claim that “science is not the only form of knowledge,” he cannot give us a single example of “knowledge” that comes from outside science. That alone invalidates this part of his argument.
For further discussion of the fallacious “other ways of knowing” claim, read pp. 185-196 of Faith Versus Fact, and for a longer discussion of the scientism canard, read pages 196-224—a section that goes over many of the issues discussed by Neil as well as the Winegards.