**natural deduction** metalanguage, practical foundations

**type theory** (dependent, intensional, observational type theory, homotopy type theory)

**computational trinitarianism** =

**propositions as types** +**programs as proofs** +**relation type theory/category theory**

An *axiom* is a proposition in logic that a given theory requires to be true: every model of the theory is required to make the axiom hold true. The sense however is that an axiom is a *basic* true proposition, used to prove other true propositions (the theorems) in the theory.

Given a language $L$ (perhaps specified by a signature: a collection of types, function symbols and relation symbols), a theory is the collection of assertions which are derivable (using the rules of deduction of the ambient logic or deductive system) from a given set of assertions, called **axioms** of the theory. In other words, a theory is generated from a set of axioms, by starting with those axioms and applying rules of deduction, much as terms in an algebraic system may be generated from a set of basic terms by applying operations. Axioms should therefore be considered as *presenting* a theory; different axiom sets may well give the same theory.

In terms of a deductive system, axioms can be regarded as “rules with zero hypotheses”. The form of such axioms depends on the details of the deductive system used: it could be natural deduction, sequent calculus, a Hilbert system, etc. If we take sequent calculus, for instance, then any collection of sequents written in the given language $L$

$\vec{\phi} \vdash_{\vec x} \vec{\psi}$

(asserting that “If every proposition $\phi_i$ is true in context $\vec{x}$ then also some $\psi_i$ is/has to be true”) can be taken as a collection of axioms for some theory. Models of the theory will then be those structures of the language in which the axioms are interpreted as true statements. For example, a model of group theory is a structure in the language of groups for which the group theory axioms hold, which is (of course) a group.

Assuming the deductive system is sound?, every sequent which is the conclusion of a valid sequent deduction, starting from the axioms, will also be true in every model. And if the deductive system is also complete, then every sequent of the language which is true in every model will in fact be provable from the axioms.

(…)

(…)

- Hilbert's sixth problem asks for an axiomatization of physics.

For instance def. D1.1.6 in

Last revised on October 30, 2022 at 14:48:21. See the history of this page for a list of all contributions to it.