Definition
A subset of a measure space that has measure zero with respect to the given measure; such sets are negligible for integration and for 'almost everywhere' statements because they do not contribute to integrals and can be omitted in measure-theoretic properties.
Principle
Principle
Properties that hold outside a null set are said to hold almost everywhere; null sets can be covered by arbitrarily small total measure, so altering a function on a null set does not change its integrals or its equivalence class in L^p spaces.
Demonstration
Demonstration
In the real line with Lebesgue measure, the set of rational numbers is a null set (countable), the Cantor set has Lebesgue measure zero though it is uncountable, and any singleton in R^n has measure zero.
Misapplication
Misapplication
Assuming a null set is topologically small (for example, that it must be nowhere dense or empty), or treating non-measurable sets as null without verifying measurability, or assuming pointwise values on a null set are irrelevant in contexts where pointwise definitions matter.
Consequence
Consequence
Null sets underpin the almost-everywhere notion used in integration, differentiation a.e., and convergence theorems: functions that differ only on a null set are identified in L^p, and many theorems permit exceptions on null sets without affecting conclusions.
Reversal
Reversal
The complement of a null set is a full-measure set; statements that fail on a full-measure set indicate genuine failure, opposite to failures confined to null sets which are often ignorable in measure-theoretic contexts.
Boundary
Boundary
Null sets are defined relative to a specific measure; a set may be null for one measure and full-measure for another. Non-measurable sets are not classified as null. The concept applies inside measure spaces, not in purely topological settings unless a measure is specified.
Semantic Tension
Semantic Tension
Tension exists between measure-theoretic 'null' and category-theoretic 'meager' notions: a set can be null but topologically large (dense) or comeager but have measure zero, so category and measure capture different notions of negligibility.
Synthesis
Synthesis
A null set is a measure-dependent negligible subset: if it can be covered by sets of arbitrarily small total measure then exceptions confined to it are ignorable for integration and almost-everywhere properties, but measurability and the ambient measure always determine that classification.