How Do Roman Numerals Work?
Seven letters, added together, with a smaller letter before a larger one meaning subtract. The rules, the common mistakes, and why clocks often show IIII instead of.
In short
Roman numerals use seven letters: I is 1, V is 5, X is 10, L is 50, C is 100, D is 500 and M is 1000. Values are added when written largest to smallest, so XVI is 16. When a smaller letter comes before a larger one it is subtracted, so IV is 4 and XC is 90. Only I, X and C are ever subtracted.
Roman numerals survive on clock faces, film credits, monarch names, book chapters and building dedications, long after they stopped being used for arithmetic. The system is simple enough to learn in ten minutes and has two rules that account for nearly every mistake people make with it.
The seven letters
| Letter | Value |
|---|---|
| I | 1 |
| V | 5 |
| X | 10 |
| L | 50 |
| C | 100 |
| D | 500 |
| M | 1000 |
The pattern alternates between 1 and 5 at each power of ten, which is thought to reflect counting on hands: I for a finger, V for the shape of an open hand, X for two hands crossed.
Rule one: add when descending
Write letters from largest to smallest and add them up.
- VI is 5 + 1 = 6
- XVII is 10 + 5 + 1 + 1 = 17
- LXXX is 50 + 10 + 10 + 10 = 80
- MDCLXVI is 1000 + 500 + 100 + 50 + 10 + 5 + 1 = 1666, which happens to use every letter once in descending order
Rule two: subtract when a smaller precedes a larger
When a smaller value sits immediately before a larger one, subtract it.
- IV is 5 − 1 = 4
- IX is 10 − 1 = 9
- XL is 50 − 10 = 40
- XC is 100 − 10 = 90
- CD is 500 − 100 = 400
- CM is 1000 − 100 = 900
This exists to avoid four identical letters in a row. Without it, 4 would be IIII and 9 would be VIIII, which are harder to read at a glance.
The restrictions that trip people up
Only I, X and C are subtracted. V, L and D never are. There is no VL for 45; it is XLV.
Subtract only from the next two values up. I can precede V and X, but not L, C, D or M. So 49 is not IL. It is XLIX: XL for 40, then IX for 9.
No more than three identical letters in a row. III is fine, IIII is not in the standard form. This is the rule that forces subtractive notation to exist at all.
Work in descending groups, left to right. 1994 is MCMXCIV: M for 1000, CM for 900, XC for 90, IV for 4. Each group is complete before the next begins.
Reading a long numeral
Break it into groups by power of ten, then add.
MMXXIV:
- MM = 2000
- XX = 20
- IV = 4
- Total: 2024
MCMLXXXIV:
- M = 1000
- CM = 900
- LXXX = 80
- IV = 4
- Total: 1984
The trick is to spot the subtractive pairs first, since they are the only places where a letter does not simply add.
Writing one from a number
Work down from the largest value, taking out as many of each as fit.
For 1,847:
- 1000 fits once: M, leaving 847
- 500 fits once: D, leaving 347
- 100 fits three times: CCC, leaving 47
- 40 is XL, leaving 7
- 5 is V, leaving 2
- 2 is II
Result: MDCCCXLVII.
The clock face question
Look at almost any traditional clock and four is IIII, not IV.
The usual explanations are practical. IIII balances VIII directly opposite it, so the dial looks symmetrical to the eye. And a clockmaker casting numerals needs only moulds containing I, V and X in particular quantities, which works out neatly when four is IIII.
A third explanation is chronological: clock faces were standardised before subtractive notation became universal in writing, and the convention simply stuck. Whatever the reason, both forms are correct and IIII on a clock is not an error.
Where the system runs out
Standard Roman numerals stop at 3,999, written MMMCMXCIX. There is no symbol larger than M, and no more than three may repeat, so 4,000 cannot be written as MMMM in strict form.
For larger numbers the Romans drew a bar above a numeral to multiply it by a thousand. A barred V meant 5,000; a barred X meant 10,000. That notation is rarely seen now outside inscriptions and academic use.
Why arithmetic was so hard
There is no zero and no place value. The position of a letter carries no information about magnitude beyond the add or subtract rule.
Multiplying XLVII by XIX is genuinely difficult, which is why Roman merchants used a counting board for calculation and wrote only the result in numerals. The arithmetic happened elsewhere, on an abacus, and the numerals recorded the answer.
That limitation is the real reason Hindu-Arabic numerals displaced the system across Europe. Roman numerals are perfectly good for recording a number and nearly useless for working with one.
Where you still meet them
Monarch and pope names, film and television copyright dates, book chapters and prefaces, Super Bowl numbering, building cornerstones, legal document sections, and the dials of clocks and watches.
In every case the job is to record a number with a certain formality or permanence, which is exactly what the system does well. Nobody chooses Roman numerals because they are convenient; they choose them because they look considered.
Common questions
Why do clocks show IIII instead of IV?
Tradition rather than rule. IIII balances the dial visually against VIII opposite it, and it lets a clockmaker cast the numerals using a simple set of moulds. Both forms are correct; the subtractive IV became standard in writing long after clock faces had settled on IIII.
What is the largest number in Roman numerals?
In standard form, 3,999, written MMMCMXCIX, because there is no symbol above M and no more than three may be repeated. Larger numbers used a bar above a numeral to multiply it by a thousand, so a barred V meant five thousand.
Is there a Roman numeral for zero?
No. The system has no zero and no concept of one, which is part of why arithmetic with Roman numerals is so cumbersome. Zero arrived in Europe with Hindu-Arabic numerals centuries later.
Try it yourself
Free, instant, and nothing you enter leaves your browser.