Why No Quintic Formula Exists: A Theorem, Not a Failure — Epoche B2
The common belief Many students assume that, just as the quadratic equation $ax^2+bx+c=0$ has the familiar formula $x=\frac{-b\pm\sqrt{b^2-4ac}}{2a}$, there must be a similar formula for every degree — and that for degree five, "nobody has managed to find it yet". Formulas do exist for cubics (Cardano, 1545) and quartics (Ferrari), so the pattern seems to invite one more step. This belief is false, and the reason is not a lack of ingenuity. It is a theorem, and the theorem also explains why the three formulas we do have exist. The claim There is no general formula in radicals — that is, using only the coefficients, the four arithmetic operations, and $n$-th roots — for the roots of the general quintic $$x^5 + a_4x^4 + a_3x^3 + a_2x^2 + a_1x + a_0 = 0.$$ "General" is essential here and means that the coefficients are treated as independent symbols, with no numerical relations assumed among them. The vocabulary, built from the quadratic formula A field is a set in which one can add, subtract, multiply and divide by anything except zero, with the usual rules — the rational numbers $\mathbb{Q}$, the reals and the complex numbers are the standard examples. A field extension is a larger field containing a smaller one. The extension that matters here is what happens when a root is added. Take $\mathbb{Q}(\sqrt{2})$, the set of all numbers $a + b\sqrt{2}$ with $a, b$ rational. It is closed under multiplication, since $(a+b\sqrt2)(c+d\sqrt2) = (ac+2bd) + (ad+bc)\sqrt2$, and under division, since $\frac{1}{a+b\sqrt2} = \frac{a-b\sqrt2}{a^2-2b^2}$. So adjoining one square root produces a genuine new field, and only a modest one: everything in it is described by two rational numbers. The quadratic formula, read structurally, says exactly this: the roots of $ax^2+bx+c$ lie in the field obtained from $\mathbb{Q}(a,b,c)$ by adjoining the single square root $\sqrt{b^2-4ac}$. Cardano's cubic formula is a two-storey version — one adjoins a square root, then a cube root of an expression involving it. A formula in radicals just is a tower of such steps, which is why the first premise below is a description rather than an assumption. A group is a set of reversible operations that can be composed, containing a do-nothing operation and an inverse for each element; here the operations are rearrangements of a polynomial's roots. Two elements commute if $gh = hg$, and a group in which every pair commutes is called abelian . A subgroup $N$ of $G$ is normal if $gNg^{-1} = N$ for every $g$ in $G$ — that is, if $N$ looks the same from every element's point of view — and this is precisely the condition under which the cosets of $N$ can themselves be multiplied, forming the quotient group $G/N$. The argument Premise 1. Any expression built from radicals corresponds to a tower of field extensions $K_0 \subset K_1 \subset \dots \subset K_m$, starting from a field containing the coefficients — for the general quintic, $K_0=\mathbb{Q}(a_0,\dots,a_4)$ — where each step adjoins one $n$-th root: $K_{i+1}=K_i(\sqrt[n_i]{\alpha_i})$. Premise 2. Galois theory attaches to each polynomial a group of symmetries of its roots: the permutations of the roots that preserve every equation with coefficients in the base field that the roots satisfy. Over a field of characteristic zero — as here, since $K_0$ contains $\mathbb{Q}$ — a polynomial is solvable by radicals if and only if this Galois group $G$ is solvable , meaning there is a chain of subgroups $G = G_0 \supseteq G_1 \supseteq \dots \supseteq G_k = \{e\}$ running from the whole group down to the trivial subgroup, each normal in the previous one, whose successive quotients are abelian. Premise 3. The Galois group of the general quintic is $S_5$, the group of all $120$ permutations of five roots. Premise 4. $S_5$ is not solvable. Conclusion. Therefore no radical tower reaches the roots of the general quintic. The formula does not exist. Premises 2, 3 and 4 each need their grounds, and they are given in turn below. Why radicals correspond to abelian layers The Galois group measures how far the roots are from being individually pinned down by the base field. For $x^2-2$ over $\mathbb{Q}$ the roots are $\sqrt2$ and $-\sqrt2$; every rational relation they satisfy — their sum is $0$, their product is $-2$ — survives interchanging them, so the group has two elements. For $x^2-4$, by contrast, the relation "the first root equals $2$" is itself rational and is destroyed by any swap, so the group is trivial. The group is large exactly when the roots are interchangeable as far as the base field can tell. Now take one step of a radical tower: adjoin $\beta = \sqrt[n]{\alpha}$, assuming the $n$-th roots of unity are already present. The other $n$-th roots of $\alpha$ are $\zeta\beta, \zeta^2\beta, \dots$, where $\zeta$ is a primitive $n$-th root of unity, so any symmetry of this step multiplies $\beta$ by some power of $\zeta$. Composing two such symmetries multiplies the exponents' sum modulo $n$ — and addition modulo $n$ is commutative. The group of each single radical step is therefore abelian. A tower of radicals is a stack of such steps, and the corresponding chain of groups has abelian quotients at every layer. That is the content of Premise 2 in one direction, and it shows why "solvable" is the right word: a solvable group is precisely one that can be dismantled into commutative layers, which is what a formula built from successive roots is able to build. Why the general quintic has the full symmetric group Expanding $(x-r_1)\cdots(x-r_5)$ shows that each coefficient is, up to sign, an elementary symmetric function of the roots: $a_4 = -(r_1+\dots+r_5)$, $a_3 = \sum_{i<j} r_ir_j$, and so on down to $a_0 = -r_1r_2r_3r_4r_5$. For the general quintic the base field consists precisely of the rational functions of these quantities. By the fundamental theorem of symmetric polynomials, every symmetric rational function of the roots can be written as a rational function of the coefficients, and