Science can answer any question about the natural world that can be tested with observable, measurable evidence and potentially proven wrong. “Does warmer water dissolve sugar faster than cold water?” is a scientific question. “Is it wrong to lie?” is not. The dividing line comes down to whether you could design an experiment or collect data that would settle the matter.
The Two Requirements for a Scientific Question
A question falls within the reach of science when it meets two conditions. First, it must involve something observable or measurable in the natural world. You need to be able to collect data, whether that means reading a thermometer, counting cells under a microscope, or tracking the orbit of a planet. If there’s no way to gather evidence, science has no way to weigh in.
Second, any proposed answer to the question must be falsifiable. That means it has to be possible, at least in principle, for an experiment or observation to prove the answer wrong. The philosopher Karl Popper used Einstein’s theory of general relativity as a classic example: it made specific predictions about how gravity bends light, and astronomers could test those predictions during a solar eclipse. The theory could have been disproven. By contrast, Popper argued that Freud’s theory of psychoanalysis made no specific predictions for any given patient, so no experiment could contradict it. As the physicist Wolfgang Pauli once said about an untestable paper: “This isn’t right. It’s not even wrong.”
If a question passes both tests, science can go to work on it. If it fails either one, the question belongs to philosophy, ethics, personal belief, or some other domain of human thought.
What Scientific Questions Look Like
Scientific questions are descriptive. They ask what is, not what ought to be. They focus on how the natural world works, what causes a particular phenomenon, or what happens under specific conditions. Here are some examples across different fields:
- Biology: Does regular exercise reduce resting heart rate over a 12-week period?
- Chemistry: At what temperature does a specific compound change from solid to liquid?
- Ecology: How does removing a predator species affect the population of its prey?
- Medicine: Does a new vaccine produce antibodies in a higher percentage of patients than the current one?
- Physics: Does the speed of sound change at different altitudes?
Each of these questions points toward something you can measure, and each proposed answer could turn out to be wrong when tested. That’s what makes them scientific. The process typically follows a recognizable path: observe something, form a testable hypothesis, define what you’ll measure and what you’ll hold constant, run an experiment, analyze the data, and draw a conclusion. Other researchers can then repeat your experiment to verify or challenge your results, which is how scientific knowledge builds over time.
Questions Science Cannot Answer
Some of the biggest questions humans ask sit outside the boundaries of science entirely. These tend to fall into a few categories.
Moral and ethical questions ask what people should or shouldn’t do. “Is it wrong to steal?” or “Should wealthy nations share resources with poorer ones?” are normative questions. They deal with values, fairness, and human judgment. Science can tell you the measurable effects of wealth inequality on health outcomes, but it cannot tell you whether inequality is morally acceptable. As several prominent moral philosophers have argued, ethical values are personal interpretations and deliberations, not general principles that can be proven true or false through experimentation.
This is sometimes called the “is/ought” distinction. Science handles “is” statements: the car is red, the river is flowing quickly, this chemical causes cancer at a certain dose. Normative claims handle “ought” statements: jazz is better than pop music, killing an innocent person is wrong, no one should fear death. A descriptive claim like “no one knows what happens after death” is very different from the normative claim “no one should fear death.” Science can investigate the first. The second is a matter of philosophy.
Questions about subjective experience and meaning also resist scientific testing. “What is the purpose of life?” or “What does it feel like to be another person?” involve inner experience that can’t be directly measured from the outside. Consciousness research illustrates this boundary well. Scientists can identify which brain regions are active during conscious experience, but identifying where something happens in the brain doesn’t explain why or how subjective awareness arises from biological processes. Researchers in this field acknowledge that many of their core theoretical claims are framed in terms that don’t translate straightforwardly into testable hypotheses, and decades of work have not yet produced a consensus theory.
Questions about the supernatural fall outside science by definition. Whether God exists, whether there is an afterlife, or whether fate controls events are not questions science can address, because they involve claims that can’t be observed, measured, or falsified. Science can describe the biological stages of dying, but it cannot tell you whether consciousness persists afterward.
Aesthetic judgments are similarly beyond science’s reach. “Is this painting beautiful?” or “Which sunset is more stunning?” depend on personal taste and cultural context, not measurable properties of the world.
The Gray Area: Questions Science Can Partially Answer
Many real-world questions blend scientific and non-scientific elements. Take “Should we allow genetic editing of human embryos?” Science can answer the factual parts: what genetic editing does to DNA, what risks it carries, how often errors occur in current techniques. But the deeper question of whether humanity should use that power involves ethics, cultural values, and political priorities that no experiment can resolve.
Climate policy works the same way. Science can measure how much the planet has warmed, model what will happen at different emission levels, and quantify the effects on ecosystems. But deciding how much economic cost is acceptable to reduce emissions is a value judgment. The scientific evidence informs the decision without making it for you.
Even within purely scientific territory, some questions remain unanswered simply because we don’t yet have the tools or knowledge to test them. How life originated on Earth, whether intelligent life exists elsewhere, and how consciousness works are all scientifically legitimate questions. They involve natural phenomena that are, in principle, observable and measurable. The challenge is practical, not philosophical. These questions sit within the domain of science even though definitive answers haven’t arrived yet.
How to Tell the Difference Quickly
When you’re trying to figure out whether a question can be answered by science, run it through three quick filters. Can you imagine collecting data that would help answer it? Could a proposed answer be shown to be wrong? And does the question ask about how the world is, rather than how it should be? If the answer to all three is yes, you’re looking at a scientific question. If any of those filters catches it, at least part of the question lives outside what science can do.
This doesn’t make non-scientific questions less important. Some of the most meaningful questions humans wrestle with, about purpose, morality, beauty, and meaning, are ones that science respectfully steps aside from. Knowing which tool fits which question is itself a valuable kind of understanding.