Mathematics

Why You Cannot Divide by Zero

Division asks what times the divisor gives the answer, and nothing times zero gives 12. Why the answer is not infinity, and where the rule bites in exams.

Short answer

Why can you not divide by zero?

Because division asks what number times the divisor gives you back the original, and nothing times zero gives 12. There is no answer to find, so the operation is undefined rather than merely difficult.

The short answer
Because division is multiplication asked backwards. 12 ÷ 3 = 4 because 4 × 3 = 12, so 12 ÷ 0 asks what number times zero gives 12 — and nothing does, since everything times zero is zero. There is no answer to find, which is why a calculator returns an error rather than infinity.

What changes the answer

12 ÷ 0 and 0 ÷ 0 fail for opposite reasons: the first has no answer at all, the second has every answer at once.
In calculus you can describe what happens as a divisor approaches zero, but that is a limit and not a division by zero.
Computers using floating-point arithmetic return Infinity for 1/0 and NaN for 0/0, which is a programming convention rather than a mathematical result.

What division is actually asking

There are two ways to read a division, and it is worth having both, because one of them explains the feeling and the other settles the argument.

The first reading is grouping. 12 ÷ 3 asks how many groups of 3 fit inside 12, and the answer is 4. Now ask how many groups of nothing fit inside 12. You can take nothing away for the rest of your life and there will still be 12 there. The question does not run out; it never finishes.

The second reading is the one to hold on to. Division is multiplication turned round: 12 ÷ 3 = 4 precisely because 4 × 3 = 12. So 12 ÷ 0 is asking for a number that, multiplied by zero, gives 12. Every number multiplied by zero gives zero. There is no such number — a fact you can check.

That is the whole reason. Dividing by zero is not forbidden by a rule someone invented; the question simply has no answer.

Zero divided by zero fails the other way round

Ask the same question of 0 ÷ 0. What number, multiplied by zero, gives zero? Five does. Minus three does. A million does. Every number in existence satisfies it.

So this one fails not for lack of an answer but because there is no way to choose between them. That is why the two cases have different names: 12 ÷ 0 is undefined, 0 ÷ 0 is indeterminate.

Note what is fine: 0 ÷ 5 = 0, because 0 × 5 = 0 and no other number works. Zero on the top is ordinary arithmetic. Zero on the bottom is the problem.

Why the answer is not infinity

Most students arrive having half-heard that the answer is infinity, and from one direction it does look true: shrink the divisor and the quotient grows without limit.

But approach from the other side and the same descent runs the other way: 12 ÷ (−0.000001) is −12,000,000. As the divisor closes on zero from below, the answer plunges rather than climbs. The two directions disagree by as much as it is possible to disagree, so no single value can be assigned that respects both.

There is a second objection, and it is the more serious one. Infinity is not a number you can do arithmetic with. If 12 ÷ 0 were infinity, then infinity × 0 would have to be 12 — and by the same argument it would also have to be 5, and 1, and everything else. Putting infinity in the answer does not fix the problem; it moves it one line down.

12 ÷ 1 = 12
12 ÷ 0.1 = 120
12 ÷ 0.01 = 1,200
12 ÷ 0.000001 = 12,000,000
12 ÷ (−0.000001) = −12,000,000

Where this actually costs marks in an exam

This is not a curiosity. Four standard exam situations are this rule wearing a different hat.

The last one is the most expensive, because it looks like good algebra and quietly destroys a solution.

Excluded values. The expression 1/(x − 3) is undefined at x = 3. Questions asking for the value of x for which an expression is not defined are asking exactly this.
Cancelling in algebraic fractions. Removing a factor of (x − 2) from the top and bottom silently assumes x is not 2, and a careful answer says so.
Gradients. Gradient is rise over run, and a vertical line has a run of zero. That is why a vertical line has no gradient rather than an infinite one, and why its equation is x = 4 rather than something of the form y = mx + c.
Losing solutions. Given x² = 3x, dividing both sides by x gives x = 3 and quietly loses x = 0. Take everything to one side and factorise instead: x(x − 3) = 0 gives both.

The 1 = 2 proof, and where it cheats

Here is an argument that appears to prove 2 = 1. Start with two equal numbers, a and b, so a = b. Multiply both sides by a to get a² = ab. Subtract b² from both sides: a² − b² = ab − b². Factorise each side: (a − b)(a + b) = b(a − b). Divide both sides by (a − b) and you are left with a + b = b. Since a = b, that says 2b = b, so 2 = 1.

Every line is correct except one. The step that divides both sides by (a − b) is dividing by zero, because a and b were equal from the start, so a − b is zero. Everything after that line is meaningless.

Do this with a student rather than describing it. Dividing by zero produces no error message in algebra; it produces a confident, tidy, completely false answer. The error message is a convenience computers give you. On paper nothing warns you at all.

Common questions

What does my calculator mean by Math ERROR?

That there is no value for it to display. The calculator has not failed and the sum is not too hard for it; the operation has no answer, so there is nothing to show. Spreadsheets say the same thing in their own words with #DIV/0!, and a programming language will usually raise an exception.

Is 0 ÷ 5 allowed?

Yes, and the answer is 0. Zero on the top is completely ordinary: it asks what number times 5 gives 0, and the answer is 0, uniquely. Only zero on the bottom causes trouble. Students who have half-remembered this rule often refuse both, and lose marks for it.

Is dividing by zero the same as dividing by a very small number?

No, and the difference is the whole point. Dividing by a very small number gives a very large answer, and you can make it as large as you like. Dividing by zero gives no answer, because the two directions of approach disagree and no single value satisfies both.

Why is 0 ÷ 0 called indeterminate rather than undefined?

Because the two failures are different. 12 ÷ 0 has no answer — nothing works. 0 ÷ 0 has every answer — everything works, and there is no way to choose. Undefined means the question has no solution; indeterminate means it has too many for any one of them to be the answer.

Does this ever change at a higher level?

In advanced mathematics you can build number systems that add a point at infinity and give division by zero a meaning inside them, but those systems give up other properties in exchange, and limits in calculus describe approach rather than arrival. None of it changes the answer expected in a GCSE or IGCSE paper.

Last updated August 21, 2026