What a magic square is
A magic square of order n arranges the whole numbers 1 to n-squared in a grid so that every row, every column and both main diagonals add to the same total. That total is fixed by the order alone: the magic constant is n(n-squared + 1) / 2 — 15 for a 3×3, 34 for a 4×4, 65 for a 5×5.
verified
The constant is not a convention but a consequence: the numbers 1 to n-squared sum to a fixed amount, that sum is shared equally among n rows, and the arithmetic is forced. This is why a claim that a given grid is “magic” is fully checkable — you add it up — and why the construction methods below are reproducible rather than secret.
You can build and verify squares in every order discussed here with this page’s companion magic square generator, which uses the historical construction methods and checks the constant for you.
The Lo Shu
verified
The 3×3 square — 4 9 2 / 3 5 7 / 8 1 6, every line summing to 15 — is the oldest magic square known, and in China it is the Lo Shu. It is embedded in correlative cosmology: the odd (yang) numbers at the cardinal points, the even (yin) numbers at the corners, five at the centre, mapped onto the nine palaces and the five phases.
disputed
Its traditional origin — revealed on the shell of a turtle from the River Luo in the age of the sage-king Yu — is legend, not history. Schuyler Cammann (1961) argued that although the 3×3 arrangement is old, its elaborate identification as the “Lo Shu”, paired with the He Tu diagram, is a later systematisation by Song-dynasty Neo-Confucian scholars. The square is ancient; the full cosmological scheme attached to it is medieval.
The grading splits cleanly: that the square exists and sums to 15 is verified arithmetic; the date and meaning of its cosmological reading are where the scholarship is careful. The same material is treated from the Chinese side under He Tu and Luo Shu.
Wafq and the budūḥ
verified
Mathematicians writing in Arabic made magic squares (wafq al-aʿdād, “harmonious disposition of numbers”) a developed branch of combinatorics from around the tenth century, with general methods for constructing squares of any order. This is genuine mathematics with a documented literature.
verified
The 3×3 square also travelled into talismanic use as the budūḥ. The name is itself a number: the four even corner values 2, 4, 6, 8 correspond to the Arabic letters bāʾ, dāl, wāw and ḥāʾ, which spell b-d-ū-ḥ — a point documented by Cammann (1969). It was inscribed on amulets, bowls and documents, notably to ease childbirth.
exploratory
That the budūḥ or any wafq exerts a talismanic effect is a claim of a different kind, with no evidential support; it is reported here as a documented historical belief, not endorsed. Al-Būnī’s thirteenth-century Shams al-maʿārif is the best-known compendium of such squares. The theoretical side is treated under the science of letters.
Varāhamihira to Khajuraho
verified
The astronomer Varāhamihira (sixth century) gives a 4×4 square in the Bṛhatsaṃhitā, used to enumerate combinations of ingredients for perfume — a magic square put to a concrete combinatorial purpose rather than a mystical one.
verified
The 4×4 square inscribed at the Parśvanātha temple in Khajuraho, dated to about the tenth century, is a pandiagonal square — the broken diagonals sum to the constant 34 as well, along with every 2×2 block, making it one of the most tightly structured 4×4 squares known. It is sometimes called the Chautisa (“thirty-four”) yantra.
The Indian cases are a good antidote to the idea that magic squares were always occult. Here the earliest clear use is a recipe table, and the finest example is architectural ornament that happens to be mathematically exceptional.
Dürer, Agrippa and the planets
verified
Albrecht Dürer’s engraving Melencolia I (1514) contains a 4×4 magic square with constant 34. Its two central bottom cells read 15 and 14 — the date of the work — and the square carries several properties beyond the defining ones. That it is in the engraving, and what it sums to, is plain fact.
verified
Heinrich Cornelius Agrippa’s De occulta philosophia (1533) sets out seven magic squares of orders 3 to 9 and assigns them to the seven classical planets — Saturn to the 3×3, Jupiter to the 4×4, and so on up to the Moon at 9×9. As a primary source it is reliable evidence for what Renaissance occult philosophy believed.
exploratory
The belief itself — that each square channels its planet’s influence in a talisman — has no evidential basis and is reported, not endorsed. Agrippa is cited here as a historical document of the planetary-square tradition, the same way the Western esoteric material is handled throughout the site.
Franklin’s squares
verified
Benjamin Franklin constructed large squares (8×8 and 16×16) in the mid-eighteenth century and described them in correspondence. They are a telling coda: Franklin’s 8×8 is not, strictly, fully magic — its main diagonals do not sum to the constant — yet its rows and columns do, and it has striking “bent diagonal” and half-row symmetries that he engineered deliberately as a recreation.
By Franklin’s hands the magic square has completed its arc: from a cosmogram and an amulet to a piece of pure mathematical play, prized for its structure and credited with no powers at all. That is the whole history in miniature.
What the squares do and do not show
Across China, the Islamic world, India and Europe the same verdict holds. The arithmetic is verified and often independently reinvented, because a magic square is a solvable combinatorial puzzle that any sufficiently numerate culture can find. The meanings — cosmological, talismanic, planetary — are local, later, and unsupported as claims about the world, however well attested as historical beliefs.
verified
This is the house pattern once more: cross-cultural recurrence that needs no diffusion and no mysticism to explain it. Many cultures found the magic square for the same reason many found base-60 — the mathematics was there to be found. Build one yourself in the generator and the inevitability is the point.
Sources
The works below are where each claim on this page can be checked. Primary texts are listed first, then the secondary scholarship.
- Agrippa von Nettesheim, De occulta philosophia libri tres (1533), Book II — the seven planetary squares (primary historical source).
- Al-Būnī (attrib.), Shams al-maʿārif (thirteenth century) — talismanic squares (primary historical source).
- Varāhamihira, Bṛhatsaṃhitā (sixth century) — the 4×4 perfume square.
- Schuyler Cammann, “The Magic Square of Three in Old Chinese Philosophy and Religion”, History of Religions 1.1 (1961), 37–80.
- Schuyler Cammann, “Islamic and Indian Magic Squares”, Parts I and II, History of Religions 8.3 and 8.4 (1968–69).
Magic Square Generator
Build squares of any order by the historical methods and verify the constant.
He Tu & Luo Shu
The Lo Shu in its Chinese cosmological setting.
Sacred Geometry
The companion test of ratio claims in stone.
The Science of Letters
The Islamic letter-number theory behind the budūḥ.
Cite this page
Formatted references for this page. The evidence grade is part of the claim — when quoting a graded statement, carry the grade with it.