The Tesseract: Why a 4D Cube Has Six Ways to Rotate

A cube's cube: building the tesseract by analogy

A tesseract (also called an 8-cell or hypercube) is the four-dimensional analogue of a cube, built the same way a cube is built from squares. A point extruded along a new axis sweeps out a line segment; a line segment extruded along a perpendicular axis sweeps out a square; a square extruded perpendicular to itself sweeps out a cube; and a cube extruded along a fourth, mutually perpendicular axis sweeps out a tesseract. Each step adds one dimension and doubles the number of vertices, so the pattern of vertex counts — 2, 4, 8, 16 — is just 2 raised to the dimension.

dimension   vertices   edges   faces   cells
0 (point)   1          -       -       -
1 (line)    2          1       -       -
2 (square)  4          4       1       -
3 (cube)    8          12      6       1
4 (tesseract) 16       32      24      8

A tesseract has 16 vertices (every combination of ±1 in four coordinates), 32 edges, 24 square faces and 8 cubic cells — the 3D cube is literally one of its eight bounding cells, with an identical cube directly "across" the fourth dimension, and six more cubes stitching the two together, the four-dimensional equivalent of the six square faces that stitch a cube's two opposite faces together.

Six planes of rotation, not three axes

In 3D, rotation happens around an axis — the axis itself stays fixed while everything else turns around it. That description stops working in 4D, because a single 1D axis leaves a 3D subspace un-fixed rather than a 2D plane, which does not correctly generalise rotation. The right generalisation is that rotation happens within a plane — the two coordinates spanning that plane rotate into each other while every other coordinate stays fixed. A cube can only rotate in the three coordinate planes XY, XZ and YZ. A tesseract, with four coordinates (x, y, z, w), has six independent coordinate planes to rotate within: XY, XZ, XW, YZ, YW and ZW — the count matches C(4,2) = 6, the number of ways to choose 2 axes out of 4.

Rotations in the three planes that don't involve w — XY, XZ, YZ — look exactly like ordinary 3D rotations and leave the w-coordinate untouched. The three planes that do involve w — XW, YW, ZW — are genuinely new: they rotate part of the shape into and out of the fourth dimension, which is precisely the motion that makes projections of a rotating tesseract look so unfamiliar, with cubes appearing to grow, shrink, and pass through each other.

Why the projection looks like it's turning inside out

A tesseract's rotation can only be shown on a 2D screen by projecting it down twice: first from 4D to 3D, then from 3D to 2D, exactly the way a wireframe cube's shadow is a 2D projection of a 3D object. During a w-involving rotation, vertices that are far from the viewer in the fourth dimension get projected smaller (in a perspective projection) and vertices that approach the viewer in w get projected larger, so the inner and outer cubes of the classic tesseract wireframe drawing appear to swap places, grow through each other, and turn inside out — an optical consequence of projection, not evidence that the shape is literally deforming.

Cross-sections: slicing instead of projecting

An alternative way to visualise 4D structure is to take a 3D cross-section — a slice — instead of a projection, the same relationship a 2D cross-section of a 3D object (like a CT scan slice) has to the full object. Slicing a rotating tesseract with a fixed 3D hyperplane produces a sequence of ordinary 3D solids that grow, shrink and morph as the slicing plane's position within the fourth dimension effectively changes with the rotation — a completely different, and often more intuitive, way to build 4D geometric intuition than watching a shadow.

Frequently asked questions

Try it live

Everything above runs in your browser — open Tesseract and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.