Get AI summaries of any video or article — Sign up free
e to the pi i, a nontraditional take (old version) thumbnail

e to the pi i, a nontraditional take (old version)

3Blue1Brown·
5 min read

Based on 3Blue1Brown's video on YouTube. If you like this content, support the original creators by watching, liking and subscribing to their content.

TL;DR

Exponentials are defined by how they convert addition-like “sliding” actions into multiplication-like “stretching/rotating” actions.

Briefing

The equation e^(πi) = −1 stops looking like black magic once exponentials are redefined as a bridge between two kinds of actions on numbers: sliding (addition) and stretching/rotating (multiplication). Instead of treating e^x as “repeated multiplication,” the explanation builds exponentials from a single functional rule: applying exponentials turns adders into multipliers in a way that forces the familiar identity e^(x+y) = e^x · e^y. That rule isn’t a consequence of counting; it’s the defining behavior of the exponential function, with the special role of e emerging from how naturally the function fits this action-based framework.

The core reframing starts by treating any number as three simultaneous roles on a line: a point, an adder (which slides the line), and a multiplier (which stretches the line). Addition is defined purely by how adders compose: sliding by one amount after another is equivalent to sliding by a single combined amount. Multiplication is defined similarly, but with a different constraint—zero stays fixed while the point for one is moved to where the multiplier’s point would land, preserving spacing. With these action rules in place, exponentials become the mechanism that converts “sliding” behavior into “stretching” behavior.

That conversion property—e^(x+y) = e^x · e^y—acts like a blueprint. Multiple functions can satisfy the same rule, but the most natural one is singled out by an explicit infinite-sum construction. The constant e is then described as the value of that function at 1, and the notation e^x is portrayed as a historical leftover from the earlier idea of repeated multiplication.

The leap to e^(πi) = −1 comes from extending the same action logic into the complex plane. Real numbers already correspond to sliding and stretching on a line; complex numbers add new actions because they can also rotate. The imaginary unit i is introduced as the point that, as an adder, slides upward, and as a multiplier, rotates the plane by a quarter-turn. Applying that rotation twice matches multiplication by −1, making i the “square root of −1” in this action language.

From there, exponentials are expected to translate the new “vertical” adder direction (movement involving i) into the new “rotational” multiplier direction. Geometrically, the exponential e^x maps points on a vertical line of adder-amounts into points on the unit circle of multipliers (radius one). The natural wrap-around is crucial: going around a circle once corresponds to an angle of 2π because circumference is 2π times the radius. Therefore, moving by π corresponds to half a revolution—landing at the point opposite 1 on the unit circle, which is −1. The result follows: e^(πi) lands exactly where the multiplier action represents a half-turn, giving −1.

Cornell Notes

The explanation treats numbers as actions, not just quantities. Addition becomes “sliding” a line, multiplication becomes “stretching” while keeping zero fixed, and exponentials are defined as the transformation that turns adders into multipliers. This leads to the key rule e^(x+y) = e^x · e^y, with e identified as the function’s value at 1. Extending the same action idea to the complex plane, i acts like a quarter-turn rotation when used as a multiplier. Since exponentials convert vertical adder motion into rotation on the unit circle, πi corresponds to a half-turn, placing e^(πi) at −1.

Why does the explanation avoid defining e^x as repeated multiplication?

Repeated multiplication only makes straightforward sense when the exponent is a countable integer, and even then it depends on how e itself is defined. Instead, the framework defines exponentials by their action-conversion property: exponentials turn the “sliding” behavior of adders into the “stretching/rotating” behavior of multipliers. That functional behavior is captured by e^(x+y) = e^x · e^y, which is treated as defining the exponential function rather than being derived from counting.

How are addition and multiplication defined using “actions” on a line?

A number is treated as a point plus two associated actions. As an adder, it slides the line so that the point labeled 0 ends up where the adder’s own point started. Successive adders combine into a single adder, which defines addition. As a multiplier, it stretches the line while fixing zero in place and moving the point for 1 to the multiplier’s point, preserving spacing. Successive multipliers combine into a single multiplier, which defines multiplication.

What role does the identity e^(x+y) = e^x · e^y play?

It expresses the central conversion idea: applying two adders (x and y) and then converting via the exponential is equivalent to converting each adder separately and then applying the resulting multipliers in sequence. In this account, that identity is not a side-effect of repeated multiplication; it’s the defining behavior that exponentials must satisfy. The constant e is then the value of this uniquely natural exponential function at x = 1.

How does i connect to rotations and why does i^2 = −1 matter here?

In the complex-plane extension, i acts as a multiplier that rotates the plane by a quarter-turn. Multiplying by i twice applies two quarter-turn rotations, which matches the multiplier action for −1. That’s why i is treated as the square root of −1: applying the “i-multiplier” action twice lands on the “−1-multiplier” action.

Why does πi correspond to −1 on the unit circle?

Exponentials are expected to convert the new adder direction (movement involving i, i.e., sliding up and down) into the new multiplier direction (rotation). Points on the vertical line of adder amounts map to points on the unit circle of multipliers (radius one). Wrapping the line around the circle without distortion means a full trip corresponds to 2π radians because circumference is 2π times the radius. Therefore π radians is half a revolution, landing at the opposite point from 1 on the unit circle—namely −1.

Review Questions

  1. In the action-based framework, what specific rule defines how adders combine, and how does that differ from how multipliers combine?
  2. How does the identity e^(x+y) = e^x · e^y function as a definition of exponentials in this explanation?
  3. What geometric reasoning links πi to a half-turn on the unit circle, and how does that determine the value of e^(πi)?

Key Points

  1. 1

    Exponentials are defined by how they convert addition-like “sliding” actions into multiplication-like “stretching/rotating” actions.

  2. 2

    The functional rule e^(x+y) = e^x · e^y is treated as the defining behavior of e^x, not as a consequence of repeated multiplication.

  3. 3

    The constant e is identified as the value of the natural exponential function at x = 1.

  4. 4

    Complex numbers extend the action picture by adding rotational behavior to multiplication.

  5. 5

    Multiplying by i corresponds to a quarter-turn rotation, so i^2 matches the multiplier action for −1.

  6. 6

    Exponentials map vertical adder motion (multiples of i) onto the unit circle, making e^(πi) correspond to a half-turn.

  7. 7

    Because a full rotation is 2π radians, π radians lands at −1 on the unit circle, yielding e^(πi) = −1.

Highlights

Reframing e^x as an action-converter makes e^(x+y) = e^x · e^y feel inevitable rather than mysterious.
In the complex-plane action view, i is a quarter-turn rotation, so applying it twice produces −1.
The unit-circle geometry supplies the punchline: πi corresponds to half a revolution, placing e^(πi) at −1.

Topics