Feline Union · The Invariant

Identity emerges from invariance across allowable transformations.

Symbolic systems evolve to preserve what matters — and break symmetry only where meaning requires it.

X
Identity holds
b
Boundary crossed

Both glyphs are turning the same 180°. One survives the turn. The other becomes a different letter. That difference is the whole subject.

↓ Scroll to turn the glyphs yourself
01
Perception favors invariance
02
Compression drives reuse
03
Transformations act
04
Some preserve identity
05
Some cross the boundary
06
Meaning emerges
01 · Natural invariance

Rotate a chair → still a chair.

Your visual system treats a rotated object as the same object. It throws away orientation almost for free, because most of the time orientation doesn't change what a thing is. Invariance is the default; the work is deciding where it should stop.

still a chair
Rotation
Angle: — and your brain doesn't blink. The label chair is invariant across the rotation.
Rotate a chair. It is still a chair. This is the default.
02 · The identity boundary

Some transformations preserve identity. Some don't.

Turn both glyphs by the same angle. X has a 180° rotational symmetry, so it lands back on itself — invariant. b does not: at 180° it becomes q, a different letter with a different meaning. The boundary isn't in the glyph — it's learned, and meaning decides where it sits.

XX
reads X · invariant
bb
reads b
Rotate both —
The ghost behind each glyph is its starting position. When the live glyph lands back on its ghost, identity is preserved.
X rotated 180° still looks like X. b rotated 180° becomes q. The boundary is learned. Meaning decides where it is.
03 · Transformations & the orbit

Rotation is multiplication by e.

Groups act on symbols — rotate, reflect, mirror. On the complex plane the rotation states live on the unit circle: z → ez. Drag θ and watch the point sweep its orbit; the four cardinal states are exactly 1, i, −1, −i. Crank n and De Moivre's theorem turns one turn into n.

0° = 1 90° = i 180° = −1 270° = −i
Angle θ —
Power n — 1
z = 1.00 + 0.00 i
(cos θ + i sin θ)n = cos nθ + i sin nθ
The yellow point is e; the violet point is zn, sweeping n× faster. Rotation = phase change. Structure can stay the same.
04 · The dihedral group D₄

Eight ways to move a square and land on a square.

D₄ — four rotations and four reflections — is the symmetry group of the square, and of many glyphs. Apply an element to the symbol x; the set of everything you can reach is its orbit, 𝒪(x) = { g·x : g ∈ G }. A high-symmetry glyph barely moves; a low-symmetry one (try F) shows all eight.

F
e · identity
Glyph:  — orbit reached: 1 / 8 distinct
Rotations + reflections form the foundation of glyph symmetry. X and O sit still under most of D₄ — that is exactly why they're cheap to read.
05 · The geometry hierarchy

Grow the group, and invariants fall away.

D₄ is finite. Loosen what counts as an allowable transformation and the group grows: rigid motions ⊂ similarities ⊂ affine ⊂ projective. Felix Klein's Erlangen program says a geometry simply is the study of what stays invariant under its group. Climb the ladder — each rung adds freedom and sheds an invariant. The survivors get deeper.

grid & parallels inscribed circle diagonals
Rotation θ —
Scale — 1.00×
Shear — 0.00
Perspective — 0.00
Invariants this group preserves
06 · Continuous groups

From four states to a sphere of them: SO(2) and SO(3).

D₄ has eight elements. The orbit on the circle, e, is the continuous group SO(2) — every planar rotation on one smooth dial, and it commutes. Step into 3D and you get SO(3): three degrees of freedom, a whole sphere of orientations — and it does not commute. The order of rotation becomes part of the answer.

Drag the cube — every orientation you reach is an element of SO(3).
Yaw — about Y —
Pitch — about X —
SO(2) is abelian: turn 30° then 50°, or 50° then 30°, you land in the same place. SO(3) is non-abelian — below, the same two 90° turns in swapped order land on different faces.
Order matters · two 90° turns, swapped
X then Y
Y then X
A sphere is invariant under all of SO(3); a labelled cube only under its 24-element rotation subgroup; a generic object, only under the identity. The bigger the symmetry group, the fewer features carry identity — and the more configurations count as the same.
07 · Examples on the boundary

Same strokes, different meaning.

The same shape, moved through the same group, lands on different letters. Literacy is largely the work of un-learning rotational invariance exactly where it would cost you meaning.

b / d / p / q

b · identity

One bowl-and-stem, four placements: b↔d is a mirror, b↔p is a flip, b↔q is a 180° turn. Same object; different orientation; different meaning.

Ambigram

swims

Engineered to sit on the identity boundary: designed so a 180° turn reads the same. Deliberate symmetry where convention usually forbids it.

Phase triangle

0° = 1

Four orientation states — 0°, 90°, 180°, 270° = 1, i, −1, −i. A single rotating mark is enough to see phase, rotation, and state transitions at a glance.

08 · What carries identity

Invariants: the features that survive the orbit.

An invariant is whatever stays fixed across every allowable transformation. These are what a symbol can be recognized by — and what a good representation should store instead of raw pixels.

We store not raw pixels, but transformation-invariant representations. These are what carry identity.
09 · Orbit & stabilizer

The counting law: |G| = |orbit| × |stabilizer|.

Back to D₄ from §04, now with its deepest accounting identity. Act on a glyph with all eight elements: the ones that leave it unchanged form its stabilizer (a subgroup); the distinct results it can reach form its orbit. The orbit–stabilizer theorem says their sizes always multiply to |G| = 8 — the symmetry you keep and the symmetry that moves you are exact complements. The eight tiles below split into orbit colour-groups, each of stabilizer size.

Orbit and stabilizer are perfect complements: the more symmetry a glyph keeps, the fewer distinct images it has. Their product never changes.
10 · Representation theory

Every pattern is a sum of irreducible symmetries.

Put a value on each of the square's four corners — that is a function on the cyclic group C₄, the rotations of §03–§04. Representation theory says it decomposes uniquely into four irreducible pieces, one per character: under a 90° turn r, each piece is simply multiplied by 1, i, −1, or −i — the very numbers this whole site is coloured by. The decomposition is the discrete Fourier transform; the piece-magnitudes are exactly the rotation invariants of the pattern.

Irreducible content — |coefficient| per character (rotation-invariant)
11 · Topological invariance

Euler characteristic: the invariant that survives any deformation.

The §05 ladder — rigid, similarity, affine, projective — is nothing next to the full freedom of topology, where every continuous stretch and bend is allowed. Almost no quantity survives that. But V − E + F does. Refine this mesh however you like — subdivide edges, add chords, drop in interior points — and the Euler characteristic χ = V − E + F = 2 never budges. It is the deepest invariant on the site: it doesn't even need straight lines.

verticesedgesfaces (+ outer)
This mesh triangulates a disc, so V − E + F = 2 = χ(sphere), however finely you cut it. Bend it into a torus and it would fall to 0 (χ = 2 − 2g). For a glyph the matching invariant is its hole-count — O→1, B→2, L→0 — untouched by any stretch (§08).
12 · Symmetry → conservation

Noether's theorem: every continuous symmetry hides a conserved quantity.

The deepest answer to why anything is conserved at all. Emmy Noether proved that each continuous symmetry of a system's dynamics forces something to stay constant: rotation-invariance ⇒ angular momentum, time-invariance ⇒ energy, shift-invariance ⇒ momentum. A planet orbits a star below — pick a symmetry and watch its invariant hold, or break exactly where the symmetry is absent.

star (central force) planet
13 · Invariance in the brain

The visual cortex wires invariance with Gabor filters.

Primary visual cortex (V1) simple cells behave like Gabor filters — a grating under a Gaussian window — sharply tuned to orientation but sensitive to phase (where the light/dark bands fall). Pool two of them in quadrature and take their energy and you get a complex cell: same orientation tuning, but invariant to phase. Drag the phase below — the simple cell flips through zero; the complex cell barely moves. Invariance, built from biology. Modern equivariant CNNs rediscover the very same trick.

Stimulus grating
V1 receptive field
Stimulus orientation —
Stimulus phase —
Simple cell — phase-sensitive
Complex cell — energy of a quadrature pair
Spin the phase: the simple cell swings through zero and flips sign; the complex cell holds steady. Rotate the orientation away and both fall silent — selective to orientation, invariant to phase.
Gabor filter bank → feature vector — magnitude response per orientation (a CNN's first layer, by hand)
14 · Invariance in language

A child and a baritone say the same vowel — the ratios match.

A vowel is fixed by its first two formants F1, F2 (resonances of the vocal tract). A child's tract is short and a grown man's is long, so their formants sit at wildly different absolute frequencies — yet we hear the same vowel. The phoneme is invariant under speaker scaling because identity lives in the pattern of ratios, not the raw hertz. Slide the vocal-tract length: in raw Hz the whole vowel cloud slides and spreads; switch to normalized and every speaker collapses onto one template. (The same scaling move as the similarity group in §05.)

ee / ah highlighted — watch their F2/F1 ratio
Vocal-tract length — 17.5 cm · adult ♂
15 · Why this matters

Everyone who builds with meaning chases the invariant.

Brain

Perception seeks invariance; learning sets the boundaries — where orientation starts to matter.

Typography

Type design exploits symmetry and transformation: few shapes, reused, many meanings.

Mathematics

Groups, representations, and orbits formalize it — invariants are the fixed points.

Physics

Symmetries govern conserved properties. (Noether: every continuous symmetry yields a conservation law.)

Machine learning

Equivariant models preserve structure across transforms; latent spaces encode orbits, not pixels.

16 · Games as groups

In a game, the invariant isn't argued. It's computed.

Every section so far read a glyph. But a game is the purest case of the whole idea: its rules are the allowable transformations, and the game is whatever they leave invariant. In these three, the invariant is not a metaphor — it literally decides what's reachable, what's solvable, and who wins. The full gaming reading →

A · the 15-puzzle — a parity invariant

Sliding never changes parity, so half of all arrangements are forbidden.

Every legal slide swaps the blank with a neighbour — one transposition, and one step of the blank. Each flips a parity; together they leave permutation parity ⊕ blank parity invariant. Scramble with legal moves all day: it stays solvable. Swap two tiles by hand and you flip the invariant — into the unreachable half the puzzle can never enter.

Inversions: 0  ·  blank row from bottom: 1
Same sixteen tiles, two disconnected worlds. The invariant — not the picture — decides which one you're in.
B · Lights Out — a subspace over GF(2)

Pressing toggles five lights at once; the solvable boards are a vector subspace.

Each press flips a cell and its neighbours — addition mod 2. The boards you can clear are exactly the column space of the 25×25 toggle matrix over GF(2), which has rank 23: only one board in four is solvable. Press mode plays the game (always stays solvable); switch to Paint to flip a single light and build a board by hand — three times out of four you'll land outside the subspace, and a hidden invariant (a “quiet pattern”) refuses it.

Lights on: 0 / 25
Solvability isn't about how many lights are on — it's whether the pattern lives in the reachable subspace.
C · Nim — the nim-sum decides everything

One XOR of the heap sizes tells you who wins, before a move is made.

Take any number from one heap; last to take a token wins. The whole game collapses to a single invariant — the nim-sum, the bitwise XOR of the heaps. Nim-sum zero is a lost position for whoever must move; the Sprague–Grundy theorem says every impartial game hides exactly such a value. The computer plays it perfectly: if you ever leave a nonzero nim-sum, it drives it back to zero.

Nim-sum (XOR): 1
Your move — click a token to remove it and everything to its right in its heap.
You move first, and 3·5·7 has nim-sum 1 — not zero — so you can force the win. But only if every move you make returns the nim-sum to zero. Slip once and the machine takes over and never lets go.
§ Share · pass it on

Identity is what survives the transformation.

Pick a hook — each is a different door into the same idea.