Rigid transformations
Which transformations preserve the notion of shape?
How does restricting morphisms change equivalence?
We added a spatial meaning to linear structure by making distance and angle part of what the space carries. Once that extra layer is present, the next task is to identify the transformations that truly respect it.
That is where rigidity enters.
In ordinary Euclidean space, some changes relocate a figure while leaving its shape intact. A triangle may be translated across the plane, rotated around a point, or reflected across a line. In each case the edge lengths remain the same, the angles remain the same, and the figure is still the same geometric object in a new position.
Those are rigid transformations.
A rigid transformation is one that preserves the full spatial data that geometry has decided to matter. The cleanest formulation is distance preservation:
d(f(p), f(q)) = d(p, q) for every pair of points p, q
Once all distances are preserved, shape is preserved with them. In Euclidean settings, angles and orthogonality come along as well. Geometry often calls such a map an isometry.
This is the exact place where the standard of admissible change becomes more refined than it was in linear algebra. An invertible linear map may send lines to lines and preserve dimension while also reshaping lengths and angles. Geometry gives the name rigid only to the transformations that carry the full metric structure intact.
Geometry also broadens the picture of motion. Space itself comes to us as a field of points, so rigid motions include translations alongside rotations and reflections. Among origin-preserving linear maps, the distinguished rigid family is the one that preserves the metric structure exactly. When the origin is fixed, these are the familiar orthogonal transformations. When the origin may move, translations join them.
V ──iso──▶ V
│ │
│distance │distance
│angle │angle
▼ ▼
V ──iso──▶ V
(commutation = rigidity)
Measure first or move first: the same geometric relations remain in view.
Once the correct morphisms have been identified, the same organizational pattern returns. Rigid transformations compose. If one motion preserves all distances and a second motion does the same, then performing one after the other still preserves all distances. The identity transformation is rigid. Reversing a rigid motion is rigid as well.
So geometry becomes a structured world whose allowed changes close under composition.
This restriction changes the meaning of sameness.
In linear algebra, two spaces can count as equivalent when an invertible linear map connects them. In geometry, sameness is organized by rigid motion. Two figures count as the same geometric figure when a rigid transformation carries one to the other.
That stricter relation is geometric equivalence, often expressed concretely as congruence.
A square and a rotated square are geometrically equivalent. A square and a translated square are geometrically equivalent. A square and a reflected square are geometrically equivalent as well if reflections are allowed among the rigid motions.
A square and a sheared parallelogram illustrate the finer grain of geometric comparison. They live in the same ambient linear setting, while geometry places them in different equivalence classes because rigid motion preserves shape exactly.
rigid: non-rigid:
+----+ +----+
| | rotate | | shear /----/
| | ----------> | | ----------> / /
+----+ +----+ /----/
same shape shape changed
This is why the choice of morphisms matters so much. The invariants are exactly the features preserved by every allowed transformation. Under rigid change, distance, angle, orthogonality, and shape remain invariant. Dimension remains invariant too, and geometry now carries a richer family of invariants alongside it.
Some features have a more delicate status. Orientation, for example, is preserved by rotations and translations and reversed by reflections. Its role therefore depends on which rigid motions the setting includes. Geometry becomes precise by tying invariance to the chosen transformations rather than to intuition alone.
The gain is conceptual clarity. Geometry is the study of what remains the same under rigid change. Once the admissible motions are fixed, the notion of geometric sameness becomes exact.
The structural upshot is this. Geometry has now identified its correct morphisms: rigid transformations, or isometries, acting on spaces and figures within them. These motions compose coherently because distance preservation survives composition, and the resulting invariants are distance, angle, orthogonality, and shape. The defining condition is d(f(p), f(q)) = d(p, q) for all points p, q. With that restriction in place, sameness becomes geometric equivalence rather than the broader linear notion. The organization of all such motions into symmetry structures is the next step.
How are rigid transformations organized into a coherent structure?
And what does it mean for a space to have symmetry?