What Algebra Really Is (And Why It's Weirder Than You Think)
Imagine you have a rule that says: “If you add two numbers, it doesn’t matter what order you add them in.” So 3 + 5 = 5 + 3, and 100 + 2 = 2 + 100. That seems obvious, right? Now imagine finding a world where that rule doesn’t work — where 3 + 5 gives you something different from 5 + 3. That would be strange. But mathematicians have actually discovered such worlds. And that’s where the real story of algebra begins.
What Most People Think Algebra Is
When you hear “algebra,” you probably think of things like solving for x in 2x + 3 = 7, or maybe graphing lines like y = 3x + 5. That’s what’s called elementary algebra, and it’s what most people learn in school. It works like this: you have numbers (constants like 0, 1, π, and variables like x and y), you combine them with operations like + and ×, and you get terms like x + 3y or √x. Then you put those terms into equations.
Some equations are laws — they’re always true, like x + y = y + x. Other equations are constraints — they’re only true for certain values, like 2x² − x + 3 = 5x + 1. Solve the constraint, and you get the specific x that makes it work. In school, these constraints show up as “word problems.” For example: if Xavier will be three times his present age in four years, how old is he? You write 3x = x + 4, solve it, and find x = 2. Easy enough.
There’s also Cartesian geometry, named after René Descartes, who figured out that you could draw equations. The equation y = 3x + 5 gives you a straight line. The equation x² + y² = 1 gives you a circle. The set of all points that satisfy an equation is called a variety — a shape made by the equation. Lines are degree 1, circles are degree 2. If you add a third variable z, you get planes and spheres. And if you keep adding variables, you get things like four-dimensional hyperplanes, which nobody can picture but mathematicians can describe perfectly well with equations.
That’s the part everyone knows. But that’s just the beginning.
When Algebra Gets Weird
Here’s where things get interesting. Elementary algebra assumes you’re working in some fixed system — usually the real numbers or the complex numbers. But what if you flipped that around? What if instead of starting with a domain (like the reals) and seeing what equations hold there, you started with a set of equations and asked: “What kinds of systems could satisfy these equations?”
That’s called abstract algebra, and it changes everything.
The key idea is simple but strange. In elementary algebra, you take it for granted that multiplication is commutative — that 3 × 5 = 5 × 3. But in abstract algebra, you study systems where that might not be true. For example, consider the six ways to shuffle three objects. If you do one shuffle followed by another, the order matters. Doing shuffle A then shuffle B might give you a different result than shuffle B then shuffle A. So the “multiplication” of shuffles is not commutative. And yet groups of shuffles form a perfectly legitimate algebraic system — one that mathematicians study seriously.
This is what abstract algebra does: it finds the hidden structure beneath the arithmetic you already know.
The Zoo of Algebraic Structures
Abstract algebra organizes itself around different kinds of structures. Here are the main ones, from simplest to richest.
Semigroups. A semigroup is just a set with an operation that’s associative — meaning (a × b) × c = a × (b × c). That’s it. The set of all words you can make from letters, under the operation of sticking them together (concatenation), forms a semigroup. “al” + “gebra” = “algebra.” The order doesn’t matter for grouping: (“al” + “geb”) + “ra” gives the same result as “al” + (“geb” + “ra”). So semigroups capture the idea of “just keep combining things.”
Monoids. A monoid is a semigroup that also has an identity — a special element that does nothing when you combine it with anything else. For words, the identity is the empty word (a word with no letters). For addition, the identity is 0. For multiplication, it’s 1.
Groups. A group is a monoid where every element has an inverse — something that undoes it. For the integers under addition, every number x has the inverse −x. For shuffles, every shuffle has an inverse shuffle that puts things back. Groups are everywhere in mathematics, physics, and chemistry. They describe symmetry. The rotations of a square form a group. The possible moves of a Rubik’s Cube form a group. The symmetries of a crystal form a group. If you want to understand symmetry, you need groups.
Rings. A ring is like having addition and multiplication together, with the rule that multiplication distributes over addition: a × (b + c) = (a × b) + (a × c). The integers form a ring. So do polynomials. So do matrices. But in a ring, multiplication doesn’t have to have inverses. You can multiply two matrices and get zero even if neither matrix is zero — that can’t happen with ordinary numbers.
Fields. A field is a ring where you can also divide (except by zero). The rational numbers form a field. The real numbers form a field. The complex numbers form a field. Fields are the richest structures — they let you do everything you learned in school arithmetic.
Why Bother With All These?
You might wonder: why do mathematicians create all these categories? Why not just study the real numbers and be done with it?
Here’s why, and it’s a genuinely cool idea.
Suppose you prove something about the integers using only equations that involve addition, subtraction, and the constant 0. That proof will also work for any system that satisfies those same equations. So if you prove something about the integers using only addition and subtraction, it automatically holds for the rational numbers too — because the rationals satisfy the same addition equations. The proof doesn’t care what the numbers are. It only cares about how they behave.
Now suppose you want to know what makes the integers special. If you find a system that satisfies all the same equations as the integers but has some different property — like being finite, or having division — then you’ve learned something. You’ve discovered that no set of equations can capture everything about the integers.
This is how mathematicians test the limits of what equations can say.
There’s a practical side too. When you solve a system of linear equations in school, you’re doing the same mathematics whether the numbers are real, rational, or complex — because all that matters is that they form a field. The same algorithm works for all of them. And fields other than the reals turn out to be incredibly useful. For example, finite fields — fields with only finitely many numbers — are the foundation of modern cryptography, which keeps your messages secure on the internet.
Abstracting Even Further
If abstract algebra studies particular classes of structures like groups and fields, universal algebra studies the patterns that all of these classes share. It asks questions like: “What does it mean for a class of algebras to be definable by equations alone?” and “What operations do you need to describe all possible structures?”
One classic result is Birkhoff’s Theorem, which says that a class of algebras is definable by equations if and only if it’s closed under three operations: taking subalgebras, taking homomorphic images (like shrinking or squishing the structure in a way that preserves the operations), and taking direct products (like making a new algebra out of pairs of elements from two old ones). This is a neat way of saying that equational logic — the logic of “always true” statements — has a clean mathematical structure.
Another big idea in universal algebra is the free algebra. A free algebra on some set of generators is the algebra that satisfies only those equations forced by the laws of its class, and no others. For example, the free monoid on two generators is just the set of all finite strings you can make from those two symbols — with no identifications except the ones required by associativity. The free commutative monoid on two generators identifies strings that differ only in order, so “xy” and “yx” become the same. Free algebras are useful because they represent the “purest” version of a structure.
Algebra and Geometry Meet Again
One of the most surprising things about abstract algebra is how it comes back to geometry. Algebraic geometry studies shapes defined by polynomial equations, but it does so over arbitrary fields — not just the reals. A circle over the integers mod 7 has only eight points, not infinitely many. And yet many theorems about real circles also hold for those eight-point circles, because the proofs use only algebraic reasoning that works in any field.
This is how algebra connects to number theory, too. Andrew Wiles’s famous proof of Fermat’s Last Theorem — which says there are no whole number solutions to xⁿ + yⁿ = zⁿ for n > 2 — used deep ideas from algebraic geometry. The proof took over three hundred years to find, and it relied on understanding the shapes defined by certain equations.
Where Things Stand
Algebra is not a finished subject. Philosophers and mathematicians still argue about what it means for something to be “algebraic,” about what counts as a legitimate algebraic structure, and about how algebra relates to the rest of mathematics and the physical world.
One open question is whether every mathematical structure can be captured algebraically. Some structures — like the real numbers themselves — are extremely hard to pin down with equations alone. Another question is about the relationship between algebra and logic: can every logical system be turned into an algebraic one, or are there limits?
And then there’s the mystery of why algebra works at all. Why should manipulating symbols according to rules tell us anything about the real world? That’s a philosophical puzzle that goes back to Plato and is still alive today.
Key Terms
| Term | What it does in this debate |
|---|---|
| Abstract algebra | Studies the general structures (groups, rings, fields) that satisfy certain sets of equations |
| Group | A set with an associative operation, an identity, and inverses for every element |
| Ring | A set with addition and multiplication, where multiplication distributes over addition |
| Field | A ring where you can divide by any nonzero element |
| Free algebra | The purest example of a structure: it satisfies only the laws forced by its class |
| Variety | A class of algebras definable by equations (in the sense of universal algebra) |
Key People
- René Descartes (1596–1650) — French philosopher and mathematician who connected algebra to geometry by inventing coordinates, letting equations describe shapes.
- Garrett Birkhoff (1911–1996) — American mathematician who proved that varieties of algebras are exactly the classes closed under subalgebras, homomorphic images, and direct products.
Things to Think About
- If groups describe symmetry, what kinds of symmetry do non-commutative groups describe? Can you think of a real object whose rotations don’t commute?
- The integers mod 7 form a field. The integers mod 8 do not (because 2 × 4 = 8 ≡ 0, so 2 has no inverse). What goes wrong when you try to divide by 2 in mod 8 arithmetic?
- Free algebras are “pure” because they satisfy only the laws forced by their class. But how would you prove that a particular algebra really is free — that it doesn’t secretly satisfy some extra law?
- Algebraic geometry studies shapes defined by polynomial equations over any field. Does it matter that a circle over a finite field has only a few points? Is it still a “circle” in any meaningful sense?
Where This Shows Up
- Cryptography — The security of your messages online depends on finite fields and the difficulty of solving certain equations in them.
- Rubik’s Cube — The moves of a Rubik’s Cube form a group, and group theory tells you how to solve it systematically.
- Physics — The symmetries of physical laws are described by groups, especially in relativity and quantum mechanics.
- Chemistry — Crystals are classified by their symmetry groups; the shape of a molecule determines how it reacts.