Finite-dimensional objects in distinguished triangles (Q852548)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Finite-dimensional objects in distinguished triangles
scientific article

    Statements

    Finite-dimensional objects in distinguished triangles (English)
    0 references
    15 November 2006
    0 references
    Let \({\mathbf D}{\mathbf M}= \text{DM}(k)\) be the triangulated category of motives over a field \(k\), with \(\mathbb{Q}\) coefficients, as constructed by \textit{V. Voevodsky} [in: Transfers and motivic cohomology theories. Ann. Math. Stud. 143, 188--238 (2000; Zbl 1019.14009)]. \({\mathbf D}{\mathbf M}\) is an additive, \(\mathbb{Q}\)-linear, pseudo-abelian, symmetric and monoidal category. A notion of finite-dimensionality for objects in such a category has been considered by \textit{S. I. Kimura} [Math. Ann. 331, No. 1, 173--201 (2005; Zbl 1067.14006)]: a Kimura-finite motive \(M\) splits into an even part \(M^+\) and an odd part \(M^-\) such that \(\bigwedge^m(M^+)= 0\) and \(S^n(M^-)= 0\) for some \(n,m> 0\). In particular, the motive of a smooth projective curve is Kimura-finite. \textit{C. Mazza} [K-Theory 33, No. 2, 89--106 (2004; Zbl 1071.14026)] introduced a notion of Schur finiteness and proves that this property defines a subtriangulated category of \({\mathbf D}{\mathbf M}\) i.e if 2 objects in a distinguished triangle are Schur-finite so is the third. Schur-finite objects maybe not Kimura-finite and this last property is in general not preserved in distinguished triangles. In this paper the author proves the following result: Theorem. Let \({\mathbf T}\) be a triangulated category, which is the homotopy category of a pointed simplicial model monoidal category \({\mathcal C}\). Assume furthermore that \({\mathbf T}\) is \(\mathbb{Q}\)-linear and pseudo-abelian: then for any distinguished triangle \(X\to Y\to Z\to\Sigma X\) in \({\mathbf T}\), if \(X\) is even (respectively odd) and \(Y\) is odd (respectively even) then \(Z\) is odd (respectively even). Note that one cannot make a similar statement if \(X\) and \(Y\) are both odd or both even, as the following example (suggested by O'Sullivan) shows. Let \(S\) be a smooth projective Kummer surface (over \(\mathbb{C}\)). Then the motive \(X= M(S)\) is evenly, finite-dimensional. Let \(Y= M(H)(2)[4]\), where \(H= \{P_1,\dots, P_n\}\) are \(n\) distinct closed points of \(S\) with \(n> 2\). Then in the distinguished triangle: \[ X\to Y\to Z, \] where \(Z= M(U)[1]\), \(U= S- H\), the motives \(X\) and \(Y\) are both even but \(Z\) is neither even nor odd, because it is Schur-finite but not Kimura-finite (see \textit{C. Mazza} in [loc. cit]). In particular the theorem above holds in the motivic stable homotopy category \({\mathbf M}{\mathbf S}{\mathbf H}\) as defined by Morel and Voevodsky. The paper also contains the following corollary, which has independently been proved by \textit{C. Mazza} [loc. cit]. Corollary. Let \(k\) be afield of characteristic \(0\) and let \(X\) be a quasi-projective curve over \(k\). Then its motive \(M(X)\) in \({\mathbf D}{\mathbf M}\) is Kimura-finite.
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references