Let g\mathfrak g be a .

A derivation of g\mathfrak g is a linear map D:ggD:\mathfrak g\to\mathfrak g satisfying the Leibniz rule

D([X,Y])=[D(X),Y]+[X,D(Y)]for all X,Yg,D([X,Y])=[D(X),Y]+[X,D(Y)] \quad \text{for all }X,Y\in\mathfrak g,

as in . The space of all derivations is a Lie algebra Der(g)\mathrm{Der}(\mathfrak g) under the commutator bracket.

An inner derivation is one of the form adX\mathrm{ad}_X for some XgX\in\mathfrak g, where adX(Y)=[X,Y]\mathrm{ad}_X(Y)=[X,Y] (see and the ). The set of inner derivations is an ideal Inn(g)=ad(g)Der(g)\mathrm{Inn}(\mathfrak g)=\mathrm{ad}(\mathfrak g)\subseteq \mathrm{Der}(\mathfrak g).

A derivation is called an outer derivation if it is not inner. The quotient Lie algebra

Der(g)/Inn(g)\mathrm{Der}(\mathfrak g)/\mathrm{Inn}(\mathfrak g)

measures outer derivations “modulo inner ones.”

Properties
  • If XX lies in the of g\mathfrak g, then adX=0\mathrm{ad}_X=0, so the map ad:gDer(g)\mathrm{ad}:\mathfrak g\to \mathrm{Der}(\mathfrak g) factors through g/Z(g)\mathfrak g/Z(\mathfrak g).
  • For many rigid Lie algebras (notably ones), every derivation is inner, so the outer derivation quotient vanishes. This is one conceptual reason semisimple Lie algebras have very small deformation theory.