The 12 Rules That Still Run Animation

The rulebook that never expired

Every few years, a new frontier of animation arrives and someone declares the old rules obsolete. Hand-drawn cels gave way to computer graphics. Computer graphics gave way to real-time engines and motion capture. Anime built an entire visual grammar on the other side of the world. And yet, watch a squirrel launch itself across a branch in a modern CGI film, or a character wind up a punch in a television anime, and you are watching the same handful of ideas that a group of Disney artists formalized decades ago.

That is the argument of this piece, and it is worth stating plainly at the top because everything else follows from it: the 12 classic principles of animation still govern modern CGI and anime because they describe how the human eye reads movement, not a fashion of any particular era. They are not a house style. They are closer to physics — or, more precisely, to the perceptual rules our brains use to decide whether motion looks alive or dead. You cannot outrun them with better hardware, because the bottleneck was never the tool. It was always the eye.

Where the twelve came from

The principles were codified in The Illusion of Life: Disney Animation, a 1981 book by Frank Thomas and Ollie Johnston — two of the animators later known as Disney's "Nine Old Men." But the book was a summary, not an invention. The ideas had been discovered the hard way across the 1930s, as animators pushed past the stiff, rubber-hose cartoons of the silent era and tried to make drawings that felt like they had weight, intention, and life.

The twelve are usually listed like this:

1. Squash and stretch — objects deform to convey weight and flexibility.
2. Anticipation — a small opposite motion prepares the audience for the main action.
3. Staging — the shot directs the eye to the one thing that matters.
4. Straight ahead and pose to pose — two ways to build a sequence of drawings.
5. Follow through and overlapping action — parts of a body keep moving after it stops, and not all at once.
6. Slow in and slow out — motion accelerates and decelerates; it rarely starts at full speed.
7. Arcs — living things move along curved paths, not straight lines.
8. Secondary action — supporting motion that enriches the main action without stealing it.
9. Timing — the number of frames a movement takes, which defines weight and mood.
10. Exaggeration — pushing a pose or action past the literal to read more clearly.
11. Solid drawing — forms that hold up as three-dimensional volumes in space.
12. Appeal — the charisma that makes a character worth watching.

Read that list again and notice something. Almost none of these are about drawing, and none are about a specific technology. They are about perception: how weight registers, how attention is directed, how the eye predicts and then confirms a trajectory. That is precisely why they survived the jump from paper to pixels.

Why the eye, not the tool, is the constraint

Here is the useful way to think about it. When you watch anything move, your visual system is running a constant, unconscious prediction: where is that mass going, and does its behavior match what I know about real objects? A ball that stops dead with no deceleration reads as fake. A face that snaps between expressions with no anticipation reads as robotic. A jump where the character never squashes on landing reads as weightless, floating, wrong.

Squash and stretch works because real flesh and real rubber deform under force, and your brain expects to see that deformation. Slow in and slow out works because inertia is real, and a motion that ignores it looks like it was dragged by a machine. Arcs work because muscles and joints pivot; nothing alive travels in a perfectly straight line. Anticipation works because in the real world, force builds before it releases — you crouch before you jump, you pull back before you throw.

None of that has anything to do with whether the image was drawn by hand, rendered by a computer, or captured off a live actor. The principles are a description of the target — "alive" — and the target does not move when the tools change. This is the whole reason a studio can hand an animator a state-of-the-art rig and still get a lifeless result: the software renders the drawing, but it does not know the rules. The rules live in the choices.

The CGI test: proof by the medium that "should" have killed them

When computer animation went mainstream, a comfortable prediction circulated: the computer would handle the physics, so animators could stop worrying about squash and stretch and arcs and all the rest. The opposite happened. Studios like Pixar became famous precisely for enforcing the old principles inside the new medium.

Consider Toy Story (1995), the first feature made entirely with computer graphics. Its animators fought the software's natural tendency toward stiff, mechanically perfect motion by deliberately building in anticipation, follow-through, and exaggerated timing — hand-authored, pose by pose. The computer could render a bouncing ball with real physics, but a physically "correct" bounce looks boring; the appealing bounce is the exaggerated one. Two decades later, Spider-Man: Into the Spider-Verse (2018) went further and made the principles visible as a style: it animated on "twos" and even threes to evoke the staccato snap of comic panels, and it treated exaggeration and staging as the whole point rather than an invisible craft layer. That film looked radically new. It was radically old underneath.

The lesson is consistent across the CGI era. The realism of the renderer went up; the reliance on the twelve principles did not go down. If anything it went up too, because a more photoreal image punishes any motion that violates the eye's expectations more harshly — this is the uncanny-valley trap, and it is the subject of a companion piece worth reading on why cartoonish beats realistic. The more real the surface, the more the movement has to obey.

The anime test: the same rules, a different accent

Now the harder case, the one that supposedly proves the principles are just a Western Disney convention: Japanese animation. Anime is famous for looking nothing like Disney — flatter, more graphic, willing to hold on a still frame for seconds, willing to blow a whole episode's budget on a few seconds of spectacular motion.

But look at what anime actually does with movement and the twelve are all there, just weighted differently. The celebrated "sakuga" sequences — the bursts of fluid, high-frame-count animation that fans clip and share — are exercises in exaggeration, follow-through, timing, and arcs pushed to their limit. A sword swing in a good action cut has enormous anticipation (the wind-up), extreme exaggeration at the apex, and overlapping action in the hair and cloth that keep moving after the body stops. That is not a Disney import. It is the same perceptual physics, discovered and pushed in a different direction.

Where anime diverges is in economy, and even that is a principle in disguise. Limited animation — holding poses, moving only the mouth, using dramatic still frames — is a masterclass in staging and timing. When a series holds on a single charged pose while the soundtrack swells, it is spending its frames where they buy the most impact, exactly as principle number three (staging) and principle number nine (timing) instruct. The budget forced the discipline, and the discipline turned out to be the craft.

For a generation of Brazilian and international viewers who grew up on both Disney features and afternoon anime blocks, this is intuitive even without the vocabulary: a Ghibli film and a Pixar film feel worlds apart, yet a jump lands with weight in both, a punch winds up in both, a face anticipates a laugh in both. Different accents. Same underlying language.

The two principles that are really about acting

Two of the twelve are worth pausing on, because they expose the deepest layer of why the list endures: appeal and timing. These are the ones that sound the most subjective — charisma and rhythm — and yet they are the most ruthless in practice.

Appeal is often misread as "make it cute." It is not. Appeal is legibility of character: a design and a performance clear enough that the eye instantly grasps who this is and what they want. A villain can have appeal without being pretty; a hero can lose all appeal by becoming muddy or generic. The reason a memorable animated character stays memorable across dubs, decades, and merchandise is that the appeal was built into the movement and the silhouette, not bolted on. It reads before a single line of dialogue lands.

Timing is even more decisive, and it is pure perception. The exact number of frames a movement occupies decides its weight, its comedy, its menace. Hold a pose two frames too long and a joke dies; cut it two frames short and a threat feels weightless. No renderer, no rig, no capture system can make this call, because timing is a reading of the audience's attention, not a property of the object. This is the clearest proof of the whole thesis in miniature: the tool can place the frames, but only a human who understands how the eye experiences duration can decide how many. The craft lives in that judgment, and the judgment is about the viewer.

The principles you notice only when they're missing

A quick diagnostic proves the point better than any praise. The clearest evidence that these rules are perceptual, not stylistic, is what happens when someone breaks them.

• A character that moves without arcs — traveling in straight robotic lines — reads instantly as a machine or a puppet, even in a beautifully rendered scene.
• Facial animation with no anticipation produces the dead-eyed snap we associate with cheap CGI and the uncanny valley.
• A heavy object that lands with no squash, slow-out, or follow-through looks like it weighs nothing, and the illusion of a physical world collapses.
• A busy shot with no staging exhausts the viewer because the eye has nowhere to land.

You do not need to know the names to feel the failure. That is the tell. When a rule can be violated and everyone in the audience — trained or not — senses that something is wrong, the rule is describing the audience, not the artist.

Why "old" is the wrong word

It is tempting to call the twelve principles old, and technically they are: the book that named them is from 1981, and the craft it described was worked out in the 1930s. But "old" implies they might one day be replaced, the way silent film gave way to sound. That analogy is wrong. Sound was a technology. The principles are not a technology; they are a description of a constraint that does not update — the human visual system.

New tools keep arriving. Real-time rendering, physics simulation, machine-generated in-betweens, motion capture that can drive a photoreal face. Every one of them changes how the frames get made. None of them changes what the frames have to accomplish to read as alive, because that job is defined on the viewer's side of the screen, and the viewer's eye has not been upgraded. A simulation still has to be art-directed toward exaggeration and appeal or it looks lifeless. Motion capture still gets cleaned up and pushed by an animator, because raw human motion, oddly, often reads as less convincing than a well-exaggerated performance.

The conclusion

So we end where we started, and the two versions of this argument — for the global reader and for the Brazilian reader who met animation through both Disney and anime — land in exactly the same place. The 12 classic principles of animation are not a period style, a Disney signature, or a stage the medium has outgrown. They are a working description of how the human eye decides that movement is alive. That is why they crossed from cels to CGI without losing force, and why they are as present in a sakuga action cut as in a Pixar chase. The tools will keep changing, faster than anyone can track. The eye will not. And as long as the eye is the audience, the twelve rules will keep running animation — because they were never really rules about drawing. They were rules about seeing.