Mathematics
How Linear Transformations Rotate and Stretch Space
Quick fact
A single 2×2 matrix can represent a rotation, a reflection, or a stretch — and its determinant tells you how much the area changes, with a negative sign meaning a flip!
Why this is interesting
You've probably seen matrices as grids of numbers, but what do they actually do? The secret is that each matrix is a machine that moves the entire space around you in a precise, grid-friendly way.
Read the full explanation
Understanding How Linear Transformations Rotate and Stretch Space
Think of linear transformations as ways to move an entire sheet of grid paper. The origin (0,0) stays put, grid lines stay straight and parallel, and distances are stretched or compressed uniformly along lines. This may sound abstract, but these transformations are all around us: rotating an image on your phone, zooming into a map, or even modeling how a stretchy material deforms. The key idea is that a linear transformation is completely determined by what it does to the basis vectors i (1,0) and j (0,1). Once you know where those two land, you know where every point goes. For example, a 90° counterclockwise rotation sends i to (0,1) and j to (-1,0). The numbers that track these new positions form a matrix: [[0, -1], [1, 0]]. Every other point (x, y) simply becomes x times the new i plus y times the new j. This simple rule is what makes matrices so powerful.
A deeper explanation
The underlying principle is linearity: a transformation T is linear if T(a + b) = T(a) + T(b) and T(c·a) = c·T(a). This means that the image of the entire space is just the span of the transformed basis vectors. When you apply a matrix to a vector, you're effectively adding together scaled copies of the columns. The geometric effect can be visualised as a combination of rotations, stretches (scaling), shears (slanting), and reflections. The determinant of the matrix tells you how areas (or volumes in 3D) change: if it's 1, area is preserved; if it's 2, areas double; if it's 0, the space collapses into a line or point; if it's negative, a reflection occurs. Eigenvectors are those special vectors that don't change direction under the transformation — they only get stretched or compressed (eigenvalues). This geometric view connects matrix algebra to real-world phenomena like stress in materials, rotations in computer graphics, and principal component analysis in data science. Understanding this helps you see matrices not as tables of numbers but as geometric actions that shape spaces.