2
$\begingroup$

I am reading a book of Arthur's book "Introduction to the trace formula".

In page 24, Arthur defined height function $H_P:G(\mathbb{A}) \to \mathcal{a}_P$ by setting $H_p(nmk)=H_{M_p}(m)$ where $n\in N(\mathbb{A}), m\in M(\mathbb{A}), k\in K$. (here, $P$ is a standard subgroup of $G$ and $NMK$ is the Iwasawa decomposition of $G$, $a_P$ is Lie algebra of $P$.)

If $P_1 \subset P$ is a standard subgroup of $G$, I am wondering whether $p \circ H_{P_1}(x)=H_P(x)$ where $p:a_{P_1} \to a_P$ is the projection map.

It seems true because Arthur used it implicitly in many parts. But I cannot prove it.

Does it really true? If so, how to prove it?

I am sorry for not explaining all the notation on my question. It is too long!

Any comments will be great helpful!

$\endgroup$

1 Answer 1

-1
$\begingroup$

This is natural because $H_{G}$ for reductive $G$ is defined by $\langle H_{G}(x),\chi\rangle=|\log(x^{\chi})|$ for $x\in G(\mathbb{A})$ and $\chi\in X(G)_{\mathbb{Q}}$, and because the projection $p:\mathfrak{a}_{P_{1}}\rightarrow\mathfrak{a}_{P}$ is the dual of the injection $X(M_{P})_{\mathbb{Q}}\rightarrow X(M_{P_{1}})_{\mathbb{Q}}$ which is just the restriction. Namely for $\chi\in X(M_{P})_{\mathbb{Q}}$, $\langle p\circ H_{P_{1}}(x),\chi\rangle=\langle H_{P_{1}}(x),\chi\rangle$ where $\chi$ in the RHS is seen as a character of $M_{P_{1}}$ so the RHS becomes $|\log(x^{\chi})|$.

$\endgroup$
1
  • $\begingroup$ Thank u very much! But since $x$ varies in $G$, I think we should boils down it to $X(A_p)$ instead $X(M_p)$. Right? $\endgroup$
    – Monty
    Aug 20, 2019 at 11:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

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