incertitudes

Incertitudes, plural of incertitude, describe the lack of complete certainty about a value, proposition, or outcome. In science and engineering, the term is most often attached to measurements: it denotes the estimated interval within which the true value lies with a stated level of confidence.

Measurement uncertainty arises from multiple sources: instrument precision, environmental conditions, observer effects, and methodological choices. The Guide to the Expression of Uncertainty in Measurement (GUM) provides a framework to model, combine, and report these components as standard uncertainty and expanded uncertainty (a confidence interval obtained by multiplying the standard uncertainty by a coverage factor, k).

Two broad kinds of incertitude are commonly distinguished: aleatoric (random) uncertainties, due to natural variability and generally irreducible by more information, and epistemic (systematic) uncertainties, arising from imperfect knowledge or bias, which can be reduced with improved methods or data.

Uncertainty propagation describes how individual sources affect the overall uncertainty of a result. Methods include analytical error propagation using partial derivatives and Monte Carlo simulations, which sample from the distributions of inputs to produce a distribution for the output.

In statistics and decision making, incertitudes appear as confidence intervals, credible intervals, or probability distributions that summarize the uncertainty surrounding a parameter or hypothesis. Transparent reporting of uncertainties is central to scientific credibility and risk assessment, as it communicates the reliability of conclusions.

Beyond measurement, incertitudes reflect a broader epistemic position: knowledge is provisional and contingent on evidence, methods, and assumptions. In everyday language, the term captures the recognition that outcomes and beliefs may be uncertain or contested.