Bourbaki

Definition 6

Tag 0584, permanent. Cite it as [Bourbaki, Tag 0584].

Algebra, chapter I, § 3. Actions

An action α↦fα\alpha \mapsto f_\alpha of a set Ω\Omega on a magma EE is said to be distributive if, for all α∈Ω\alpha \in \Omega, the mapping fαf_\alpha is an endomorphism of the magma EE.

If ⊤\top denotes the law of the magma EE and ⊥\perp the law of action associated with the action of Ω\Omega on EE, the distributivity of the latter is then expressed by the formula (6)

α⊥(x⊤y)=(α⊥x)⊤(α⊥y)(for α∈Ω and x,y∈E). \alpha \perp (x \top y) = (\alpha \perp x) \top (\alpha \perp y) \quad \text{(for } \alpha \in \Omega \text{ and } x, y \in E\text{)}.

By an abuse of language, it is also said that the law ⊥\perp is distributive (or right distributive) with respect to the law ⊤\top.

Formula (2) of § 1, no. 2 shows that then, for every ordered sequence (xλ)λ∈L(x_\lambda)_{\lambda \in L} of elements of EE and every α∈Ω\alpha \in \Omega, (7)

α⊥(⋀λ∈Lxλ)=⋀λ∈L(α⊥xλ). \alpha \perp \left( \bigwedge_{\lambda \in L} x_\lambda \right) = \bigwedge_{\lambda \in L} (\alpha \perp x_\lambda).

If an action α↦fα\alpha \mapsto f_\alpha is distributive and an equivalence relation RR on EE is compatible with the law of composition of EE and the action α↦fα\alpha \mapsto f_\alpha, the quotient action on E/RE/R is distributive.

When the law on EE is written multiplicatively, we often use the exponential notation xαx^\alpha for a law of action which is distributive with respect to this multiplication, so that distributivity is expressed by the identity (xy)α=xαyα(xy)^\alpha = x^\alpha y^\alpha. If the law on E is written additively, we often use left (resp. right) multiplicative notation α.x\alpha . x (resp. x.αx . \alpha) for a law of action which is distributive with respect to this addition, the distributivity being expressed by the identity

α(x+y)=αx+αy(resp. (x+y)α=xα+yα). \alpha(x + y) = \alpha x + \alpha y \quad (\text{resp. } (x + y)\alpha = x\alpha + y\alpha).

We may also consider the case where Ω\Omega has an internal law, denoted by T‾\overline{T}, and the law of action is distributive with respect to the first variable, which means that

(8)

(αT‾β)⊥x=(α⊥x)T(β⊥x) (\alpha \overline{T} \beta) \perp x = (\alpha \perp x) T (\beta \perp x)
for all α,β∈Ω\alpha, \beta \in \Omega and x∈Ex \in E. Then, by formula (2) of § 1, no. 2

(9)

(T‾λ∈Lαλ)⊥x=T‾λ∈L(αλ⊥x) \left( \overline{T}_{\lambda \in L} \alpha_{\lambda} \right) \perp x = \overline{T}_{\lambda \in L} (\alpha_{\lambda} \perp x)
for every ordered sequence (αλ)λ∈L(\alpha_{\lambda})_{\lambda \in L} of elements of Ω\Omega and all x∈Ex \in E.

Bourbaki puts the proof directly after the statement, so what is above is the statement and its proof together.

Cited by

Nothing.

Cites

Transcribed from the printed edition, 1998, Springer. Not translated.