Bourbaki

Remark 4

Tag 02OY, permanent. Cite it as [Bourbaki, Tag 02OY].

Théories spectrales, chapter III, § 1. Applications linéaires compactes

Let E be a locally convex space over C\mathbf{C}. Let us denote by E0E_0 the locally convex space over R\mathbf{R} underlying E. In order that E be an approximation space, it is necessary and sufficient that E0E_0 be one. In fact, the condition is necessary; let us prove that it is sufficient. Suppose therefore that E0E_0 is an approximation space. Let C be a precompact subset of E and U a balanced convex neighbourhood of 0 in E. Put C′=C∪iCC'= C\cup iC. There exists a continuous R\mathbf{R}-linear mapping uu of finite rank from E0E_0 into E0E_0 such that x−u(x)x-u(x) belongs to U for every x∈C′x\in C'. Put v(x)=12(u(x)−iu(ix))v(x) =\frac{1}{2}(u(x)-iu(ix)) for every xx in E. Thus one defines a continuous C\mathbf{C}-linear mapping of finite rank from E into E. For every x∈Cx\in C, we have x∈C′,ix∈C′x\in C',ix\in C' and x−v(x)=12(x−u(x))−i2(ix−u(ix))x-v(x) =\frac{1}{2}(x-u(x))-\frac{i}{2}(ix-u(ix)), so that x−v(x)x-v(x) belongs to U. This proves that E is an approximation space.

Bourbaki sets the proof directly after the statement, and the corpus does not separate them, so what is above is the statement and its proof together.

Cited by

Nothing.

Cites

Nothing.

Machine translation of the English, by gpt-5-6-mini, gpt-5-mini, gpt-5.4, against glossary version 34. Not checked by a person. That model is a cut down version of the one the rest of this language was written by.