Take the 2-minute tour ×
MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

Let $Q$ be a anisotropic quadric of dimension $d$ over $k$. We work in the category of effective Chow-Motives over $k$. Let $T$ be the Tate-Motive. For a motive $M$ we write $M(l)$ for its $l$-th Tate-Twist.

Assume we have a motivic decomposition $M(Q) = A \oplus B$. Assume further that over the algebraic closure of $k$, the motive $A$ decomposes as $T(a) \oplus T(b)$,while $a\leq b$. The motive $A$ is then called a binary summand of M(Q). The dimension of the motive $A$ is defined as $b-a$.

Question: Is it true that the dimension of every binary summand of a quadric is odd ? $*$

I think i once red it in a paper (of Vishik?), but i need the reference.

$*$ Only exception may be the case if its zero. For example the motive of a zero dimensional quadric splits in two trivial Tate-Motives and thus its motive has dimension zero.

share|improve this question
    
I don't know much about motives, but shouldn't the motive of any smooth quadric surface $S$ be something like $T(0)\oplus \mathrm{Pic}(S)\oplus T(−2)$? Does this not give a counter-example, on taking $A=T(0)\oplus T(−2)$? –  Daniel Loughran Aug 18 at 17:22
    
The motive of a SPLIT quadric is does indeed have the form $T(0) \oplus T(1) \oplus T(1) \oplus T(2)$, while the $T(1)$ basically represent CH$^1(Q) = $Pic$(Q)$. This is a well known fact.But i look at a anisotropic projective quadric.Assume it has a decomposition into $A \oplus B$ over $k$. Then none of these two motives can contain a trivial Tate-Motive because this happens iff it is isotropic over $k$. Take for example a two dimensional Pfister-quadric. It has a decomposition into two Rost-Motives iff its anisotropic. But these split into $T(0) \oplus T(1)$ and $T(1) \oplus T(2)$! –  nxir Aug 18 at 18:23

1 Answer 1

up vote 4 down vote accepted

The binary motive theorem states that a binary summand of the motive of a quadric has dimension $2^r-1$ for some $r$. This should be Theorem 6.1 in:

O. Izhboldin and A. Vishik. Quadratic forms with absolutely maximal splitting. Contemp. Math. 272 (2000), 103-125.

share|improve this answer
    
Yes,thats it. Thanks! –  nxir Aug 18 at 19:33

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.