History of numbers

The Ishango bone, a baboon fibula carved with grouped notches, is among the oldest widely-cited physical evidence of tallying — but what those groupings actually represent is genuinely debated, not settled fact.

Discovered in the 1950s in what is now the Democratic Republic of the Congo, the bone carries three columns of notches arranged in groups. Some researchers have proposed the groupings reflect prime numbers or doubling; others read it as a six-month lunar tally; others caution against over-interpreting a single artifact whose maker left no explanatory record. What's uncontested is that it demonstrates deliberate, structured tally marking tens of thousands of years before writing.

Why it matters: It shows that recording quantity — the most basic building block behind every number system that came later — predates written language, agriculture, and settled civilisation by a wide margin.

Uncertainty: The specific meaning of the notch groupings (calendar, arithmetic exercise, or something else) is disputed among researchers; only the fact of deliberate tallying is well established.

One-to-one tally marking — a notch or mark for each item counted — appears independently across unconnected prehistoric cultures worldwide, because it requires no shared transmission, only a simple correspondence between marks and things.

Alongside the Ishango bone, the older Lebombo bone (found in Eswatini, with a proposed date of roughly 35,000 BCE) shows similarly deliberate notching. Because tallying needs no borrowed idea — just matching one mark to one object — it's one of the clearest cases in the historical record of the same basic technique being reinvented repeatedly rather than spreading from a single origin point.

Why it matters: It's the foundational counter-example to any 'numbers were invented once, by one people' narrative: the simplest form of counting is something independent human groups arrived at on their own, over and over.

Independent parallel development: Tally marking is attested independently across multiple prehistoric cultures with no plausible contact between them, making it a textbook case of convergent invention rather than diffusion from one source.

Ancient Egyptian numerals used a distinct symbol for each power of ten (1, 10, 100, 1,000, and so on) repeated and added together — there was no positional place value, so the order symbols were written in didn't change their meaning.

To write 300, an Egyptian scribe drew three coiled-rope symbols (each meaning 100) side by side; the total was found by simply adding every symbol present, in whatever order they were arranged. This additive approach is simple to read but grows unwieldy for large numbers — writing 9,999 required 36 separate symbols.

Why it matters: It's a clear illustration of what a numeral system looks like before place value exists, making the later invention of positional notation (where a digit's position, not just its shape, carries meaning) easier to appreciate as a genuine leap rather than an obvious default.

Babylonian scribes developed one of the earliest true positional number systems — using base 60 rather than base 10 — where a symbol's position, not just its shape, determined its value.

Written in cuneiform on clay tablets, the Babylonian system combined just two basic symbols (for 1 and 10) into groups that stood for values up to 59, then used position to represent multiples of 60, 3,600, and beyond — the same principle underlying modern decimal place value, just with 60 as the base instead of 10. For centuries the system had no symbol at all for an empty position, which could make numbers genuinely ambiguous; a placeholder symbol was only introduced later, and it still didn't function as a true zero you could compute with.

Why it matters: Its legacy survives directly in how we measure time and angles today — 60 seconds in a minute, 60 minutes in an hour, 360 degrees in a circle — making it one of the few ancient number systems still in everyday use, even though almost nobody using it realises where it came from.

Common misconception: That base-10 is the 'natural' or only sensible base for a positional number system — Babylonian base-60 shows positional place value works with any sufficiently practical base, and predates most base-10 positional systems by well over a thousand years.

The Maya developed a base-20 (vigesimal) positional number system with its own shell-shaped symbol for zero, used as a placeholder in calendar calculations — entirely independently of the zero concepts developing in Asia at a similar and later time.

Maya numerals combined dots (for 1) and bars (for 5) with a distinct shell glyph marking an empty position, letting scribes write large calendar counts positionally in base 20. This development had no contact with, and owes nothing to, the separate emergence of zero in Babylon, India, or the Islamic world — it's one of the clearest known cases of the same core idea being invented twice.

Why it matters: It directly undercuts any claim that zero was 'invented' by one civilisation and spread everywhere else from that single point — the Maya reached a working placeholder zero without any link to the Old World traditions.

Independent parallel development: Developed completely independently of Old World zero concepts (Babylonian, Indian) — no evidence of contact or transmission in either direction.

Common misconception: That zero has one single origin story — the Maya case shows a placeholder zero was reached at least twice, independently, on opposite sides of the world.

The Indian mathematician Brahmagupta was the first known to give zero formal arithmetic rules in his 628 CE text Brahmasphutasiddhanta — treating it as a number you could add, subtract, and multiply with, not merely a placeholder for an empty position.

Earlier systems (Babylonian, Maya) used a symbol to mark an empty place in a positional number — genuinely useful, but not the same as treating 'nothing' as a number with its own defined behaviour. Brahmagupta set out explicit rules: a number plus zero is unchanged, a number minus itself is zero, and so on. Full profile at /number/0 covers zero's broader story, including its later transmission and modern conventions.

Why it matters: This is the conceptual leap from 'a symbol that marks an empty column' to 'a number you can compute with' — the difference between a placeholder and an actual number, which is what let zero eventually become foundational to algebra.

The ten digits used worldwide today (0–9) descend from Brahmi numerals developed in India, refined over centuries into a positional base-10 system before being transmitted westward through the Islamic world.

Early Brahmi numeral forms gradually evolved through several regional Indian scripts into shapes recognisably close to today's digits, combined with true positional place-value notation and a functioning zero. This combination — ten digit shapes, positional value, and zero — is what made the system so much more efficient for calculation than the additive systems (like Egyptian or Roman numerals) that came before it.

Why it matters: It's the direct ancestor of the numeral system almost the entire world now uses, making its Indian origin one of the most consequential and least widely known facts in the history of mathematics.

Common misconception: That the numerals are 'Arabic' in the sense of originating in the Arab world — the Islamic world's essential role was transmitting and refining the system (see the al-Khwārizmī milestone), not originating the digit shapes themselves.

The Persian mathematician al-Khwārizmī wrote an influential treatise on calculating with Indian numerals in Baghdad around 825 CE, a key link that carried the Indian decimal system into the Islamic world and, eventually, into Europe.

Working at the House of Wisdom in Baghdad, al-Khwārizmī's book explained how to perform arithmetic using the Indian digit-and-place-value system. Later Latin translations and adaptations of his work — and of the numerals it taught — carried the system toward Europe. His name, Latinised as 'Algorithmi', is the direct root of the modern word 'algorithm'; his book on equation-solving, al-Jabr, gives us 'algebra'.

Why it matters: Without this transmission link, the Indian system might have remained regional — it's the connective step between zero's conceptual origin in India and its eventual adoption across the Islamic world and Europe.

Leonardo of Pisa (Fibonacci) introduced Hindu-Arabic numerals to European merchants in his 1202 book Liber Abaci, but adoption was far from instant — some cities actively banned the new numerals for official record-keeping for generations afterward.

Liber Abaci demonstrated the practical calculating advantages of positional decimal numerals over Roman numerals for trade and bookkeeping. Despite this, resistance was real: Florence's merchant guild banned the use of Arabic numerals in official ledgers in 1299, requiring numbers to be spelled out in words instead, partly out of concern that digits like 0 and 1 could be altered to commit fraud more easily than Roman numerals could. Widespread European adoption took roughly three more centuries after Fibonacci's book.

Why it matters: It's a useful corrective to the idea that better tools are adopted as soon as they're available — a genuinely superior numeral system still faced generations of institutional resistance before it won out.

Common misconception: That Hindu-Arabic numerals swept across Europe immediately once introduced — in practice, adoption was gradual and, in places, actively resisted for political and practical reasons.

Chinese mathematicians were computing with negative numbers (represented with counting rods of a different colour) by around 200 BCE, and Brahmagupta gave them formal arithmetic rules in India in 628 CE — centuries before European mathematicians widely accepted negative numbers as legitimate.

The Chinese text The Nine Chapters on the Mathematical Art describes using red counting rods for positive quantities and black rods for negative ones, applying them to practical problems involving debts and surpluses. In India, Brahmagupta's same 628 CE text that formalised zero also gave explicit rules for negative numbers (called 'debts', as opposed to positive 'fortunes'). European mathematicians, by contrast, were still routinely describing negative numbers as 'absurd', 'fictitious', or meaningless well into the 17th and even 18th centuries.

Why it matters: It's one of the clearest examples that mathematical progress isn't a single steady line — a concept fully mastered elsewhere can sit unresolved and actively disputed in another tradition for over a thousand years.

Independent parallel development: Chinese and Indian mathematicians developed working rules for negative numbers independently of one another, both well before European mathematics caught up.

Common misconception: That negative numbers are a 'modern' mathematical idea — they were in practical use, with formal rules, over a thousand years before most European mathematicians accepted them.

Fractions have a long, plural history: ancient Egyptians almost always wrote fractions as sums of distinct unit fractions (like 1/2 + 1/3), Chinese mathematicians used decimal fractions in precise calculations centuries before Europe, and it was the Flemish mathematician Simon Stevin whose 1585 book De Thiende popularised decimal fraction notation in the West.

The Rhind Mathematical Papyrus (c. 1650 BCE) shows Egyptian scribes expressing almost every fraction other than 2/3 as a sum of distinct unit fractions — a workable but cumbersome system. Separately, Chinese mathematicians including Liu Hui and Zu Chongzhi used decimal fractions to compute remarkably precise approximations of pi by the 5th century CE. In Europe, Simon Stevin's De Thiende (1585) is widely credited with popularising a systematic decimal fraction notation for general use, building on earlier partial approaches by other mathematicians rather than starting from nothing.

Why it matters: It shows a familiar pattern in this history: a genuinely useful notation is rarely a single 'eureka' invention — it's typically refined and popularised by one identifiable figure after being anticipated, in different forms, elsewhere.

Independent parallel development: Decimal fraction techniques were in documented use in China roughly a millennium before Stevin's work popularised comparable notation in Europe.

Physical calculating devices — abacus-like counting tools — arose independently in multiple ancient civilisations, and the line from those devices through 17th-century mechanical calculators to today's electronic computers is a continuous story of building better tools to compute with numbers, not a single invention.

Counting boards and abacus-like devices are independently attested in Mesopotamia, China, ancient Rome, and among Mesoamerican and Andean cultures, each developed with no contact between the others. Centuries later, Blaise Pascal built one of the first mechanical calculators (1642) capable of automatically carrying digits during addition, and Charles Babbage designed (though never completed in his lifetime) far more ambitious mechanical computing engines in the 19th century. Electronic computers in the 20th century continued the same underlying goal — performing arithmetic on numbers faster and more reliably than a person can by hand.

Why it matters: It ties the abstract history of number concepts to the physical history of the tools used to compute with them, showing that 'inventing better ways to calculate' is a thread that runs continuously from prehistory to modern computing.

Independent parallel development: Abacus-like calculating devices were independently developed in multiple, unconnected ancient civilisations.