---
title: "Wall Forms"
canonical: "https://docs.magma-maths.org/v2.30-1/MatricesLinearAlgebra/PolarSpaces/wallform.html"
source: "https://docs.magma-maths.org/v2.30-1/MatricesLinearAlgebra/PolarSpaces/wallform.md"
part: "Matrices and Linear Algebra"
chapter: "Polar Spaces"
breadcrumbs:
  - {"title": "Magma Handbook", "source": "https://docs.magma-maths.org/v2.30-1/index.md"}
  - {"title": "Matrices and Linear Algebra", "source": "https://docs.magma-maths.org/v2.30-1/MatricesLinearAlgebra/index-matrices-linear-algebra.md"}
  - {"title": "Polar Spaces", "source": "https://docs.magma-maths.org/v2.30-1/MatricesLinearAlgebra/PolarSpaces/index-polar-spaces.md"}
prev: {"title": "Lie Algebras and Bilinear Forms", "source": "https://docs.magma-maths.org/v2.30-1/MatricesLinearAlgebra/PolarSpaces/liealg.md"}
next: {"title": "Invariant Forms", "source": "https://docs.magma-maths.org/v2.30-1/MatricesLinearAlgebra/PolarSpaces/invform.md"}
intrinsics:
  - {"name": "WallForm", "signature": "WallForm(V, f): ModTupFld, Mtrx -> ModTupFld, Map"}
  - {"name": "WallIsometry", "signature": "WallIsometry(V, I, mu): ModTupFld, ModTupFld, Map -> Mtrx"}
  - {"name": "WallDecomposition", "signature": "WallDecomposition(V, f): ModTupFld, Mtrx -> Mtrx, Mtrx"}
  - {"name": "SemiOrthogonalBasis", "signature": "SemiOrthogonalBasis(V): ModTupFld -> SeqEnum"}
  - {"name": "GeneralisedWallForm", "signature": "GeneralisedWallForm(V, f): ModTupFld, Mtrx -> ModTupFld, Map"}
---

<a id="wallform"></a>
<a id="sectwall"></a>
# Wall Forms

Given an isometry $f$ of a quadratic, symplectic, or unitary space $V$ with bilinear or sesquilinear form $\beta$, the Wall form of $f$ is the form $\theta$ defined on the image $I$ of $1-f$ by $\theta(u,v) = \beta(w,v)$, where $u = w(1-f)$. In general, the Wall form is not reflexive.

<a id="function-wallform-modtupfld-mtrx"></a>
## `WallForm(V, f): ModTupFld, Mtrx -> ModTupFld, Map`

The space of the Wall form of the isometry $f$ and its embedding in $V$.

<a id="function-wallisometry-modtupfld-modtupfld-map"></a>
## `WallIsometry(V, I, mu): ModTupFld, ModTupFld, Map -> Mtrx`

The inverse of `WallForm`. This is an isometry corresponding to the embedding $\mu : I \to V$, where $V$ is a quadratic, symplectic, or unitary space.

<a id="function-walldecomposition-modtupfld-mtrx"></a>
## `WallDecomposition(V, f): ModTupFld, Mtrx -> Mtrx, Mtrx`

An isometry $f$ of a quadratic or symplectic space $V$ is *Wall-regular* if the restriction of $1-f$ to the image of $1-f$ is invertible. If $f$ is any isometry of $V$ this function returns a Wall-regular element $f_r$ and a unipotent element $f_u$ such that $f = f_rf_u = f_u f_r$.

<a id="function-semiorthogonalbasis-modtupfld"></a>
## `SemiOrthogonalBasis(V): ModTupFld -> SeqEnum`

If $V$ is a vector space with a bilinear form $\beta$, a basis $e_1$, $e_2$, …, $e_n$ for $V$ is semi-orthogonal if $\beta(e_i,e_j) = 0$ for $i < j$. This function returns a semi-orthogonal basis with respect to the non-degenerate, non-alternating form attached to $V$. If the base field is ${\bf F}_{2}$, the form should be symmetric.

<a id="function-generalisedwallform-modtupfld-mtrx"></a>
## `GeneralisedWallForm(V, f): ModTupFld, Mtrx -> ModTupFld, Map`

This function returns the space of the generalised Wall form of the similarity $f$ and its embedding in the quadratic space $V$. Suppose that the quadratic form $Q$ of $V$ is nondegenerate and let $\beta$ be its polar form. Then $Q(vf) = \eta Q(v)$ for some $\eta$. Suppose that $\eta = \zeta^2$ and let $V(f,\zeta)$ denote the $\zeta$-eigenspace of $f$. The generalised Wall form $\theta$ of $f$ is defined on the orthogonal complement of $V(f,\zeta)$ by $\theta(u,v) = \beta(w,v)$, where $u = \zeta w - wf$.
