Logarithms — The Invention That Saved Science

Source: "The Invention That Saved Science" (on logarithms)

1. The Naive Question

Imagine you're Kepler in 1600. You have 20 years of Tycho Brahe's astronomical observations — the most precise data ever collected on planetary motion. The data is in there somewhere. A pattern exists. But to find it, you need to multiply and divide 7-digit numbers — thousands of times. By hand. With quill and ink.

At that pace, a single calculation takes minutes. Thousands of calculations take months. Years, even. The discovery of the laws governing the solar system is arithmetically blocked.

The question is: can we make arithmetic fast enough for science to actually work?

2. Strip the Assumptions

Assumption 1: "Logarithms are a mathematical abstraction."
They were invented as a computational tool — a slide rule in book form. The abstract mathematics came later. Napier's motive was purely practical: save scientists from drowning in arithmetic.

Assumption 2: "Multiplication is fast."
For small numbers, yes. For 7-digit astronomical figures, a single multiplication takes several minutes of careful pen-and-paper work with many intermediate steps and opportunities for error. At scale, it is the bottleneck of scientific progress.

Assumption 3: "Logarithms are about exponents."
Today we teach logs through exponents: log₁₀(100) = 2 because 10² = 100. But Napier didn't think of it this way at all. He thought of it as a mapping between two moving points — one constant-speed, one slowing down. The exponent interpretation came later, from Briggs and Euler.

3. Build Back Up

The Core Mechanism: Turn Multiplication into Addition

The key observation: if you have two geometric sequences and their corresponding arithmetic (index) sequences, then:

Multiplying two numbers in the geometric sequence = Adding their indices
Dividing two numbers = Subtracting their indices

Example with powers of 2:

Geometric:   1,  2,  4,  8, 16, 32, 64 ...
Index:        0,  1,  2,  3,  4,  5,  6 ...

To multiply 8 × 16: look up their indices (3 + 4 = 7), 2^7 = 128. Done.

Lookup → Add → Lookup back. Three trivial steps instead of long multiplication.

The Spacing Problem and Bürgi's Fix

Early geometric tables were coarse — numbers spaced too far apart for precision. Clockmaker Joost Bürgi solved this using a ratio of 1.0001 instead of 2, shrinking the steps until entries were dense enough for 9-digit accuracy.

Napier's Continuous Leap

John Napier (who coined "logarithm" — logos + arithmos, "ratio number") took it further by imagining continuous motion: two points moving along two lines simultaneously. One at constant speed (the log), one slowing proportionally (the number). This modelled a smooth function rather than a discrete table — the first hint of what would later become calculus.

Henry Briggs then simplified Napier's system to Base 10, making it universally practical: every integer step in the log corresponds to shifting a decimal point.

Kepler's Discovery

With Briggs' tables, Kepler took the logarithms of the planetary orbital periods and their distances from the sun. The "signal" emerged from the noise visually:

log(period) vs log(distance) → a straight line

A straight line in log-log space means a power law. The slope was exactly 3/2. Therefore:

T² ∝ a³

Kepler's Third Law of Planetary Motion — one of the most important equations in the history of science — was found by reading a graph that logarithms made possible. Without log tables, Kepler might never have finished.

4. The Insight

Logarithms were to the 17th century what computers were to the 20th: they didn't change what was mathematically possible, they changed what was computationally feasible — and in doing so, they unlocked four centuries of scientific discovery.

Every major scientist from Newton to Einstein used log tables and slide rules. Astronauts went to the moon using algorithms descended from Briggs' 1617 tables.

The deeper insight: the bottleneck of scientific progress is often not conceptual — it's computational. Remove the arithmetic friction and discoveries cascade. This is why every generation of computing tools (logarithms → mechanical calculators → digital computers → neural computation) produces an explosion of new science.