A misleading name
The digits used in most of the world today are often called Arabic numerals, but their origins lie in India. Indian mathematicians developed a place-value system with ten symbols, including a symbol for zero, in the first millennium. Scholars writing in Arabic learned the system, adapted it, wrote about it and carried it west. So the usual name credits the messenger, not the inventor. A more accurate label is Indian-Arabic or Hindu-Arabic numerals.
The forms of the symbols also changed along the way. The shapes used in Spain and North Africa, often called the western Arabic forms, are the ancestors of the digits used in Europe. The shapes used in the eastern Islamic world, which are still used in many Arabic-speaking regions today, are different.
Why place value mattered
Roman numerals are awkward for calculation. Writing a number such as 1,349 as MCCCXLIX is fine for recording a date or a total, but doing long multiplication or division with such symbols is hard. Medieval Europeans commonly calculated using a counting board or abacus with counters and wrote down only the result.
A place-value system with zero lets a person do arithmetic on paper by following a set of steps. Zero is the key idea: it holds an empty place so that, for instance, 205 differs from 25. The Arabic word for it, sifr, is the origin of both the English words cipher and zero, through Latin and Italian forms.
Al-Khwarizmi and the two books
The Persian-born scholar Muhammad ibn Musa al-Khwarizmi worked in Baghdad in the early ninth century, during the period associated with the House of Wisdom. Two works matter here. One was a text on arithmetic using Indian numerals. The original Arabic is not known to survive in its early form, but a Latin version made in Europe, probably in the twelfth century, begins with the words “Dixit Algorizmi,” meaning “Algorizmi said.” His name gave us the word algorithm.
The other was a book on solving equations, whose Arabic title includes the word al-jabr, referring to the operation of restoring or completing a quantity by moving a term to the other side of an equation. That word became algebra. The book is not written in modern symbols. Problems and solutions are set out in words, with practical cases such as inheritance, land measurement and trade, and with geometric explanations of why the methods work.
It is important not to overstate the credit. Greek, Indian and Babylonian traditions all lay behind this work, and later Arabic-language mathematicians developed algebra much further, including work on polynomials and the solving of cubic equations. Al-Khwarizmi is best seen as an early and influential author, not the sole inventor.
How the knowledge crossed over
The movement west took several routes. One was through Spain and Sicily, where Latin readers had contact with Arabic books. In the twelfth century translators, working in cities such as Toledo, rendered mathematical and scientific texts into Latin. Robert of Chester is traditionally credited with a Latin translation of al-Khwarizmi’s algebra in 1145, and Gerard of Cremona translated a version too.
Another route was commerce. Italian merchants trading around the Mediterranean, including those of the Venetian republic and Genoa, met Arab and North African counterparts, and practical arithmetic was a useful skill. Leonardo of Pisa, known as Fibonacci, grew up partly in North Africa where his father worked, and in 1202 he wrote the Liber Abaci, a large book that explained Indian-Arabic numerals and their use for business problems such as currency exchange and profit sharing. It did not use the numbers to create a new mathematics; it taught a practical method to readers who wanted to calculate.
Some stories mention Gerbert of Aurillac, a scholar who became pope in 999 as Sylvester II, and his use of a counting board with numerals. He is often cited as an early figure in the transmission of the numerals, though how much he really knew of the full system is debated.
A slow, uneven adoption
The new numbers did not replace the old ones overnight. For centuries, arguments continued between those who calculated with an abacus and the algorists, who preferred written arithmetic. Roman numerals stayed standard in many records, on clocks, in church books and on monuments, and they are still used for some purposes today.
There were also reasons for caution. A written 0 can be altered into a 6 or a 9 more easily than a Roman numeral can be changed, which made some officials uneasy. Some city authorities are reported to have restricted the numerals in certain official or banking records, but the details are not well documented, and historians treat the stories with care. Education mattered most: as medieval universities and commercial schools taught the techniques, written arithmetic became ordinary. By the later fifteenth century it was widespread, and the printing press helped fix the forms.
What changed for Europe
The long-term effects were large. Easier calculation helped accounting, tax records, navigation tables, the calculations behind medieval astrology and astronomy, and later science. Algebra provided methods of problem solving that European mathematicians of the Renaissance extended, partly with a new symbolic notation developed in the sixteenth and seventeenth centuries. None of this was inevitable, and it required centuries of teaching and use.
The lesson is that mathematical ideas travel through people: Indian authors, Arabic-language scholars, Latin translators, merchants and teachers. Each group adapted what it received. The label on the numerals is a useful reminder of the middle step, but the full story is a chain.