In logic, a predicate is a symbol which represents a property or a relation. For instance, in the first order formula ${\displaystyle P(a)}$, the symbol ${\displaystyle P}$ is a predicate which applies to the individual constant ${\displaystyle a}$. Similarly, in the formula ${\displaystyle R(a,b)}$, ${\displaystyle R}$ is a predicate which applies to the individual constants ${\displaystyle a}$ and ${\displaystyle b}$.

In the semantics of logic, predicates are interpreted as relations. For instance, in a standard semantics for first-order logic, the formula ${\displaystyle R(a,b)}$ would be true on an interpretation if the entities denoted by ${\displaystyle a}$ and ${\displaystyle b}$ stand in the relation denoted by ${\displaystyle R}$. Since predicates are non-logical symbols, they can denote different relations depending on the interpretation used to interpret them. While first-order logic only includes predicates which apply to individual constants, other logics may allow predicates which apply to other predicates.

Predicates in different systems

• In propositional logic, atomic formulas are sometimes regarded as zero-place predicates^[1] In a sense, these are nullary (i.e. 0-arity) predicates.
• In first-order logic, a predicate forms an atomic formula when applied to an appropriate number of terms.
• In set theory with excluded middle, predicates are understood to be characteristic functions or set indicator functions (i.e., functions from a set element to a truth value). Set-builder notation makes use of predicates to define sets.
• In autoepistemic logic, which rejects the law of excluded middle, predicates may be true, false, or simply unknown. In particular, a given collection of facts may be insufficient to determine the truth or falsehood of a predicate.
• In fuzzy logic, predicates are the characteristic functions of a probability distribution. That is, the strict true/false valuation of the predicate is replaced by a quantity interpreted as the degree of truth.