Linear Algebra: The Language of Geometry

Linear algebra is the language used to describe and manipulate the objects of analytical geometry. Every core concept in linear algebra, from vectors to eigenvectors, has a direct geometric interpretation.

Vectors as Points and Directions

The most fundamental connection is the vector. In analytical geometry, we describe a point with coordinates like $(x, y)$. In linear algebra, this is represented as a column vector. This vector can be seen as either the position of the point or as an arrow from the origin to that point, representing a direction and magnitude.

Matrices as Geometric Transformations

A matrix is a function that transforms geometric space. When you multiply a vector by a matrix, you are applying a transformation to that point.

• Scaling: The matrix $\begin{bmatrix} 2 & 0 \\ 0 & 2 \end{bmatrix}$ doubles the size of any shape.
• Rotation: The matrix $\begin{bmatrix} \cos\theta & -\sin\theta \\ \sin\theta & \cos\theta \end{bmatrix}$ rotates every point in the plane by an angle $\theta$.
• Projection: A singular matrix “projects” or squishes a higher-dimensional space onto a lower one, like casting a 3D shadow onto a 2D wall.

Dynamic visualization of a linear transformation:

Lines and Planes as Subspaces

In analytical geometry, a line or a plane is a set of points satisfying an equation. In linear algebra, these are described as subspaces or the span of vectors.

Eigenvectors as Invariant Directions

An eigenvector of a matrix represents a direction that remains unchanged (except for scaling) during the transformation. In analytical geometry, this defines the principal axes of the transformation.

Imagine a stretching transformation. The eigenvectors are the directions along which the stretching occurs. A point on an eigenvector’s path will stay on that path; it will only move further from or closer to the origin.

The first visualization below shows a full-rank transformation — one that stretches and shears space without collapsing it. Notice how the grid, originally square, becomes a parallelogram, yet the entire plane remains intact. The eigenvectors mark the invariant directions of this transformation, each scaled by its corresponding eigenvalue.

Determinant as Area/Volume Change

The determinant of a matrix has a direct geometric meaning: it’s the scaling factor of area (in 2D) or volume (in 3D) after the transformation.

• If $\det(A) = 3$, the transformation makes any shape three times larger in area.
• If $\det(A) = 0$, the transformation collapses the space into a lower dimension (e.g., a plane onto a line). The resulting “area” is zero, which is why the matrix is called singular.

In contrast, the second visualization shows a singular transformation. Here, the determinant equals zero, meaning all points are projected onto a single line. The grid collapses, and the transformed “area” vanishes. This demonstrates the geometric meaning of a singular matrix — a transformation that destroys one dimension of space.

Systems of Equations as Intersections

Solving a system of linear equations is equivalent to finding the intersection of geometric objects.