Omar Khayyam's geometric solution to the cubic x³ + bx = c
Six centuries before Cardano, the 11th-century Persian poet-mathematician solved every positive-coefficient cubic by intersecting a parabola and a circle.
What this shows
Six hundred years before Cardano and Ferrari brought symbolic cubic-root formulas to Europe, Omar Khayyam (Nishapur, 1048–1131) gave a complete geometric construction for every cubic equation with positive coefficients. His Treatise on Demonstration of Problems of Algebra classifies cubics into fourteen cases and shows how to solve each by intersecting two conic sections.
For the case x³ + b·x = c (with b, c > 0), Khayyam's construction is:
Parabola: y = x² / √b
Circle: x² + y² = (c / b) · x
— passes through the origin,
centre at (c/(2b), 0), radius c/(2b)Substituting the parabola's y² = x⁴/b into the circle's equation and simplifying gives x⁴ + bx² = cx, which factors as x·(x³ + bx − c) = 0. Discarding the trivial root x = 0 leaves x³ + bx = c — so the x-coordinate of the non-trivial intersection of the two conics IS the real root of the cubic.
The figure shows this for x³ + 3x = 14. Khayyam's parabola is y = x² / √3; the circle has centre (14/6, 0) ≈ (2.33, 0) and radius 7/3. They meet (other than at the origin) at x = 2, the exact real root (check: 8 + 6 = 14).
Where it shows up
Khayyam's construction is the high-water mark of pre-symbolic algebra — a complete classification of cubics that wouldn't be superseded until the Italian renaissance algebraists worked out radical formulas in the 16th century. It is also the first systematic use of conic sections as algorithmic tools rather than purely geometric objects.
The technique generalises beautifully: every algebraic equation whose solution can be expressed using square and cube roots admits a ruler-compass-and-conic-section construction, which is exactly the class of constructions Khayyam's method exhibits. This idea — the relationship between algebraic extension degree and geometric constructibility — anticipates Galois theory by 800 years.
Frequently asked questions
Why did Khayyam not just use the algebraic formula?
There wasn't one. Symbolic algebra as we know it didn't exist in the 11th century — equations were stated in words, and Khayyam himself remarked that no Islamic mathematician had found a general algebraic solution for cubics. His geometric construction was the closest thing to a general method until Cardano's formula appeared in 1545.
Does the construction always give a real root?
For x³ + bx = c with b, c positive, yes — the cubic always has exactly one positive real root, and the parabola-and-circle intersection produces it. Khayyam treats the case of negative coefficients (and equations with three real roots) in separate sub-cases, each with its own pair of conics.
Did Khayyam know how many roots a cubic could have?
He understood that a cubic could have one or three real roots depending on the case, and his fourteen-case classification reflects that. He did not, however, accept negative numbers or zero as solutions — the geometric construction inherently produces only positive lengths.
Why is Khayyam famous as a poet and mathematician?
He's the same person in both contexts. Khayyam wrote the Rubaiyat (quatrains) that Edward FitzGerald translated into English in 1859, and he simultaneously did first-rate mathematics — this cubic-equation work, calendar reform achieving accuracy comparable to the modern Gregorian, and the first systematic exposition of parallel-postulate alternatives to Euclid.