Steven Pinker has kicked up a cloud of dust with a seemingly mild claim, addressed to people in the humanities: Science Is Not Your Enemy. And he’s right, it’s not! Science is merely an extremely effective method for gaining empirical knowledge of the world, and empirical knowledge of the world should not strike fear into any self-respecting intellectual person. Or if it does, perhaps you should contemplate a different form of employment, like U.S. Senator.
The devil is in the details, of course, and plenty of people have objected to the specific ways in which Pinker has argued that science is your friend, and others have defended him. Here are takes by Jerry Coyne, Eric MacDonald, and Massimo Pigliucci. I don’t mean to add anything deep or comprehensive to the debate, but I do want to make a suggestion that, if adopted, would make the world a better place: the word “scientism” should be dropped from the vocabulary of this discussion.
Now (like Pinker), I am a descriptivist rather than a prescriptivist when it comes to language. Word usage is not “right” or “wrong,” it’s just “useful” or “unhelpful.” So the point here is that use of the word “scientism” is unhelpful, not that people are using the “wrong” definition. It’s unhelpful because it’s ill-defined, and acts as a license for lazy thinking. (It wasn’t too long ago that I acknowledged the potential usefulness of the term, but now I see the error of my ways.)
The working definition of “scientism” is “the belief that science is the right approach to use in situations where science actually isn’t the right approach at all.” Nobody actually quotes this definition, but it accurately matches how the word is used. The problem should be obvious — the areas in which science is the right approach are not universally agreed upon. So instead of having an interesting substantive discussion about a real question (“For what kinds of problems is a scientific approach the best one?”) we instead have a dopey and boring definitional one (“What does the word `scientism’ mean?”).
I don’t know of anyone in the world who thinks that science is the right tool to use for every problem. Pinker joins Alex Rosenberg, who has tried to rehabilitate the word “scientism,” claiming it as a badge of honor, and using it to mean a view that “the methods of science are the only reliable ways to secure knowledge of anything.” But even Alex firmly rejects the idea that science can be used to discover objective moral truths — and others think it can, a view which is sometimes labeled as “scientism.” You can see the confusion.
Someone might respond, “but `scientism’ is a useful shorthand for a set of views that many people seem to hold.” No, it’s not. Here are some possible views that might be described as “scientism”:
- Science is the source of all interesting, reliable facts about the world.
- Philosophy and morality and aesthetics should be subsumed under the rubric of science.
- Science can provide an objective grounding for judgments previously thought to be subjective.
- Humanities and the arts would be improved by taking a more scientific approach.
- The progress of science is an unalloyed good for the world.
- All forms of rational thinking are essentially science.
- Eventually we will understand all the important questions of human life on a scientific basis.
- Reductionism is the best basis for complete understanding of complicated systems.
- There is no supernatural realm, only the natural world that science can investigate.
The problem is that, when you use the word “scientism,” you (presumably) know exactly what you are talking about. You mean to include some of the above supposed sins, but not necessarily all of them. But if you aren’t completely explicit about what you mean every time you use the term, people will misunderstand you.
Indeed, you might even misunderstand yourself. By which I mean, using vague words like this is an invitation to lazy thinking. Rather than arguing against the specific points someone else makes, you wrap them all up in a catch-all term of disapprobation, and then argue against that. Saves time, but makes for less precise and productive discussion.
Given that the only productive way to use a word like “scientism” — something vaguely sinister, ill-defined, used primarily as an accusation against people who would not describe themselves that way — would be to provide an explicit and careful definition every time the word is invoked, why use it at all? I’m not saying you can’t disagree with specific claims made by Pinker or anyone else. If you think people are making some particular mistake, that’s fine — just say what the mistake is.
I take the main point of Pinker’s piece to be the same as Feynman’s discussion of the beauty of a flower, or Dawkins’s Unweaving the Rainbow — science is not opposed to the humanities or the arts, but enhances them by giving us a deeper understanding. With that, I couldn’t agree more. We can disagree with some of the specific contentions in a constructive way, but lumping everything we don’t like into one catch-all word isn’t useful.
TL;DR: The word “scientism” doesn’t helpfully delineate a coherent position, it unhelpfully flattens important distinctions and creates a false target. We can do better.
“Humanities and the arts would be improved by taking a more scientific approach.”
I’m a humanities/social science student and while I don’t disagree with the idea in practice, I would argue that that is what most of us, in fact, do (or at least try to do). However, there is a tendency, of which Pinker is the almost perfect example, of imagining that there are quick fixes to solving the problems of the humanities and social sciences by incorrectly pulling over the ideas from ‘science’ (as generally understood) into the humanities and social sciences and saying they are the solution – without often really ever understanding what the problems are. That is “scientism”.
“The working definition of “scientism” is “the belief that science is the right approach to use in situations where science actually isn’t the right approach at all. … Indeed, you might even misunderstand yourself. By which I mean, using vague words like this is an invitation to lazy thinking. … Given that the only productive way to use a word like “scientism” — something vaguely sinister, ill-defined, used primarily as an accusation against people who would not describe themselves that way — would be to provide an explicit and careful definition every time the word is invoked, why use it at all? … but lumping everything we don’t like into one catch-all word isn’t useful.”
You are exactly right. It is not helpful by using such a catch-all poorly defined word.
Yet, for Pinker to have a talk with the title “Science Is Not Your Enemy”, it shows that there is an issue much deeper than a single word “scientism”. Science instead of being *merely* an extremely effective method for gaining empirical knowledge of the world, it was claimed by very many as the “only” way of gaining empirical knowledge of the world, and this is the problem. There is no crackpot-philosopher or crackpot-musician, only crackpot-physicist. The problem is the definition of “science” by many physicists. If a theory is not a part of the mainstream menu, it is automatically labeled as crackpot. Then, many philosophers and artists are worse than crackpots but are idiots, and this was evident in the “Nothingness debate”. Why should those idiots not fighting back?
I do see a major problem in the definition of “science”, especially on two of its terms, “prediction and theory”.
“Pre- (before)” of prediction was referring to with a “time-frame”. When a “new” data confirms the calculations of an old theory, that old theory made good “prediction”. When the calculation of a new theory fits the old data, it is a “post-“diction, and a postdiction has no scientific value.
Then, there are two types of theories.
Type A — a hodgepodge of test data, with the “best-fit” mathematic equation and with many hand-put-in parameters. This type theory does not truly make “theoretical prediction” but predicts from not “fitting” well, such as there must have a third generation of quark in order for a better fit of the known data.
Type B — an axiomatic system, with a “base” (definitions, axioms, procedures) and consequences (sentence, theorem, etc.).
As the hand-manipulated theory, the type A theory can always be tweaked to make postdiction. On the other hand, the consequences of type B theory can never make any postdiction without changing its base. Thus, the “prediction” of type B theory is not referring to a time-frame but is about its base. If its base “below (or beneath)” the data’s theoretic-reference-frame, its consequences still “predict” that old-data.
The new meaning for “prediction” is very important. That is, the “base” of theory B is verified without itself being subject to “direct-testing”. With this, science “could” well be the right tool to use for every problem, including in the “Nothingness debate”, and even the issues of consciousness and intelligence. A not-testable “base” can be verified via its consequences.
“Scientism” is the use of theory to explain reality. “Science” is the use of reality to explain theory.
Can’t you say the same thing for every -ism ever? What makes “Scientism” so special that it should be immune from criticism?
I think a problem we have in debates between scientists and religious types is that we scientists try to adopt and use their language. “Objective moral values” is the most glaring. This is nothing but a supernatural standard that can never be demonstrated. So when scientists attempt to engage theists in debate the scientists paint themselves into a corner by trying to borrow the theist’s terminology.
This is a partial misreading (or at least not a full reading) of the concept of scientism as typically deployed within the social sciences and philosophy. the notion is that “science” (not as a practice but as a concept) has come to be a signifier rather than a practice- that is, that science has come to be seen (in many quarters) as the only legitimate intermediary between humans and the natural, non-human world, and thus is the ultimate arbiter in social debates. this obviously leads to a great deal of bad scientific claims (by non-scientists mostly) and cherry-picking data in order to prove ideological claims about the world within a pseudo-scientific framework (as in how most climate change deniers rely on arguments couched in scientific language)
“scientism” further, encompasses a type of positivist reductionism whereby (for example) human society is to be understood solely through structures and processes in the brain rather than through interactions with other humans and with the environment. social problems in this paradigm are thus to be solved e.g. through pharmaceuticals rather than through community building and social negotiations. or in the case of Dawkings, assuming that you understand theology and philosophy as well as someone who has spent a lifetime studying it because you are an expert in another discipline operating under an entirely different episteme.
i respect you enormously Sean, and love your work, and the same goes for most of the scientists mentioned in this article, but the scientific community would be far better served to read a corresponding article from a famous social scientist entitled “social sciences (or humanities) are not your enemy”. i have graduate degrees in both Environmental Sciences and Anthropology from Columbia, and i can say that there is at least as much misunderstanding of positions in the social sciences by scientists as vice versa. ultimately we are trying to strengthen scientific practice, not weaken it. you are guilty of exactly the same thing of which you accuse Feynman and Dawkin’s critics: relying on a popularized notion of a concept without engaging with the body of academic literature supporting it.
I agree with Sean that the term should be dropped or at least avoided. Some religious folks like to use it as a straw man to undercut the value of science (particularly in education), as if that makes dogma more valid. I have previously written about that here: Straw Man Scientism.