Evolution of number representation

Every numeral system that has ever existed can be described using the same small set of features, which makes wildly different-looking systems easy to compare.

• Correspondence tools. The earliest and most universal stage: matching one physical mark to one counted object — a pebble, a knot, a tally stroke. Even tribal systems that count only in twos (building every number by repeating the words "one" and "two") belong here, showing that this basic idea was reinvented independently, not copied from one source.
• Base. The number of units collected before the system starts a new group or place. Egypt and India eventually settled on base $10$ (ten fingers); the Maya used base $20$ (fingers and toes); Babylon used base $60$, prized because $60$ has many divisors ($2, 3, 4, 5, 6, 10, 12, 15, 20, 30$), which makes splitting quantities evenly far easier than with base $10$.
• Additive versus place-value structure. In an additive system, a symbol contributes the same value no matter where it sits — Roman $X$ is always $10$. In a place-value system, a symbol's value depends on its position — the $3$ in $39$ means $30$ only because it sits in the tens place, while the same $3$ in $312$ means $300$.
• Presence of zero. Whether the system has a symbol to mark an empty place, and further, whether that symbol is treated as a genuine number that can be added to or subtracted from — a idea that arrived comparatively late, developed fully in Indian mathematics before spreading through the Arab world to Europe.

Combining these features gives the overall historical arc:

where: each arrow marks a real limitation being solved — additive numerals let you write large counts without laying out thousands of tally marks; place value lets you reuse the same few symbols for numbers of any size; zero lets place value work without ambiguity and turns "nothing" into a number you can compute with. This is not one ladder every civilisation climbed in lockstep — Roman numerals stayed additive for over a thousand years, and different cultures reached different stages completely independently — but it captures what problem each new idea solved.

Place value beats an additive system for a concrete, checkable reason, not just tradition. An additive system must give every distinct magnitude its own symbol or repeat symbols to build size — Roman numerals need $I, V, X, L, C, D, M$ and a subtraction convention just to reach into the thousands, and writing a number like $3888$ (\text{MMMDCCCLXXXVIII}) takes twelve characters. A place-value system needs only as many distinct symbols as its base (ten digits for base $10$), because the same ten symbols are reused in every place, with position alone telling you whether a digit means units, tens, hundreds, or more. This reuse is exactly what makes standard column arithmetic possible. Adding two additive numerals has no fixed step-by-step procedure — you must combine symbols freely and re-simplify by hand each time, and the method looks different for every pair of numbers. Adding two place-value numerals, by contrast, is a single repeatable recipe: align digits by place, add each column, carry into the next place when a column reaches the base, and the same recipe works whether the numbers have two digits or twenty. Zero is the piece that makes this recipe unambiguous: without a symbol for "this place is empty," $"305"$ and $"35"$ could be confused, since there would be no way to show that the tens place holds nothing. Once zero marks the empty place and is treated as a genuine number, addition, subtraction and even the concept of "nothing" become things you can compute with rather than special cases you must talk around.

Problem. Compare Roman numeral $\text{XXXIX}$ (which equals $39$) with Hindu-Arabic $39$: count the symbols each uses, classify each as additive or place value, and carry out the addition $39 + 24$ in both systems to see which supports a repeatable method.

Step 1. Count symbols. $\text{XXXIX}$ breaks into $XXX$ (three $X$'s, worth $10$ each, totalling $30$) followed by $IX$ (the subtractive pair worth $10 - 1 = 9$) — five characters in total. Hindu-Arabic $39$ uses just two characters, the digits $3$ and $9$.

Step 2. Classify structure. In $\text{XXXIX}$, every $X$ contributes exactly $10$ regardless of where it sits — this is additive, with a subtraction rule bolted on for cases like $IX$. In $39$, the digit $3$ means $30$ because it occupies the tens place; move it and its meaning changes. This is place value.

Step 3. Add the Roman way. $\text{XXXIX} + \text{XXIV}$ (that is, $39 + 24$) has no column procedure: you must combine all the symbols and re-simplify by hand, $30 + 9 + 20 + 4 = 63$, then convert $63$ back into Roman form as $\text{LXIII}$.

Step 4. Add the Hindu-Arabic way. Align by place: units column $9 + 4 = 13$, so write $3$ and carry $1$; tens column $3 + 2 + 1(\text{carry}) = 6$. Reading off the columns gives $63$ directly, using the same fixed recipe that works for any two numbers, however large.

Result. Both numerals name the same quantity, $39$, but Hindu-Arabic $39$ uses fewer symbols (2 versus 5) and, because it is place value, supports the general column-addition recipe that Roman numerals cannot offer — this is the central practical reason place-value systems eventually replaced additive ones almost everywhere arithmetic mattered.

• Assuming history is a single straight line from "primitive" to "advanced" — sophisticated place-value systems (Babylonian, Mayan) coexisted for centuries with purely additive systems (Roman) used by other contemporaneous cultures, not one replacing the other in strict sequence.
• Calling a tribal counting system (such as counting only in twos) "inferior" rather than recognising it as a complete, self-consistent additive system suited to that community's needs.
• Forgetting that place value's win — fewer symbols and a repeatable calculation method — is a genuine trade-off against the extra idea of "nothing has to be marked," not a free upgrade with no cost of understanding.
• Treating Hindu-Arabic as the one "correct" system rather than one efficient combination (base $10$, place value, and zero) that spread worldwide through trade and scholarship, while other complete systems remain historically valid on their own terms.

Ask the tutor: Ask the AI tutor to help you design your own additive counting system (say, base $5$ or base $12$) and test whether it can support a working column-addition recipe.