Science is frequently invoked in public life as a source of authoritative knowledge. “The science says…” has become a common rhetorical move in policy debates, media reporting, and everyday argument. The implication is that science produces clear, stable, and actionable truths.

This picture is both powerful and misleading.

The Productive Role of Uncertainty

At the heart of the scientific enterprise is a particular relationship to ignorance. Good scientific work does not simply replace uncertainty with certainty. It transforms vague ignorance into specific, well-characterized uncertainty. It replaces “we don’t know” with “we don’t know this, but we have narrowed the possibilities to these.”

This is not a weakness. It is the mechanism by which science makes progress. A field that cannot clearly state what it does not yet understand is not in a position to design the next experiment or observation that would be most informative.

The physicist Richard Feynman captured something important when he said that science is the belief in the ignorance of experts. The most reliable scientific knowledge is often accompanied by the clearest statement of its own limitations.

When Uncertainty Is Hidden

Problems arise when this relationship to uncertainty is obscured in communication with the public and with decision-makers.

In some cases, the uncertainty is genuinely small, and presenting results with appropriate confidence is justified. In other cases—particularly in emerging areas, complex systems, or value-laden domains—the uncertainty is large and structural. When such findings are presented without adequate qualification, several distortions follow:

  • Overconfidence in policy. Decisions made on the basis of findings whose uncertainty was downplayed can produce backlash when the limitations become apparent.

  • Polarization. When scientific claims are presented as more settled than they are, disagreement is more easily interpreted as irrationality or bad faith rather than as a reasonable response to genuine ambiguity.

  • Erosion of trust. When the public discovers that scientific claims were presented with more certainty than the evidence supported, trust in the institution of science is damaged.

Communicating What We Do Not Know

There is a genuine tension here. Policymakers and the public often want clear guidance. Scientists are trained to be precise about what they do and do not know. These incentives are not perfectly aligned.

Nevertheless, several practices seem worth defending:

Distinguish different kinds of uncertainty. Not all uncertainty is the same. There is statistical uncertainty within a well-specified model. There is model uncertainty (are we even asking the right question?). There is uncertainty arising from incomplete data or from the inherent variability of complex systems. Good communication tries to be clear about which kind of uncertainty is at play.

Present ranges and scenarios rather than single numbers when appropriate. In climate science, for example, the presentation of ranges of possible outcomes under different emissions pathways has been more useful for decision-making than any single “best estimate.”

Acknowledge when uncertainty is unlikely to be reduced quickly. Some questions are difficult not because we lack data but because the systems are genuinely complex or because the relevant evidence is intrinsically limited. Pretending that more research will soon deliver definitive answers can be misleading.

Separate the communication of findings from the communication of implications. Scientists can be rigorous about what their data show while remaining appropriately modest about what those data imply for policy or personal decisions, which necessarily involve values and trade-offs.

Living with Provisional Knowledge

Perhaps the deepest challenge is cultural rather than technical. Modern societies have inherited from the Enlightenment a picture of science as a steadily accumulating body of reliable knowledge that can serve as a foundation for rational action. This picture is not entirely false, but it is incomplete.

A more accurate picture would include the recognition that scientific knowledge is always, to some degree, provisional; that important domains remain resistant to full quantification and prediction; and that the relationship between scientific understanding and wise action is mediated by values, institutions, and judgment.

This does not mean that we should be paralyzed by uncertainty. It means that we should develop better cultural and institutional capacities for acting under conditions of incomplete knowledge. That includes:

  • Decision frameworks that are robust across a range of possible outcomes rather than optimized for a single predicted future.
  • Institutions that can revise their positions as evidence changes without losing legitimacy.
  • A public culture that can tolerate ambiguity without immediately descending into relativism or authoritarian demands for certainty.

The Virtue of Intellectual Humility

In the long run, the credibility of science may depend less on its ability to project certainty than on its willingness to be honest about the boundaries of its knowledge. The disciplines that have maintained the highest public trust over time are often those that have been most careful not to claim more than they can justify.

Uncertainty, properly understood and communicated, is not the opposite of scientific authority. It is one of its essential conditions.