The Substitution That Collapses the Algebra

Solve $x^4 - 5x^2 + 4 = 0$. Written out, it looks like a quartic — normally a serious undertaking with no clean formula on the whiteboard. But look closer: the only powers of $x$ that appear are $x^4$ and $x^2$, and $x^4 = (x^2)^2$. The equation is secretly a quadratic wearing a disguise. The whole difficulty was cosmetic.

The idea: name the repeated chunk

Whenever the same expression — or the same expression squared — shows up more than once inside a problem, give it a new name and rewrite everything in terms of that name. Suddenly a quartic becomes a quadratic, a nasty radical expression becomes a linear equation, or a sprawling system becomes two lines you can solve by substitution. The algebra doesn't get easier because you did new mathematics — it gets easier because you stopped re-deriving the same sub-expression's behavior every time it appeared and instead treated it as a single unit.

This is the same instinct as caching a repeated subcomputation: once you notice $x^2$ playing the same role everywhere, solving for it once and translating back at the end is strictly less work than tracking $x$ through every step.

Worked example 1: the disguised quadratic

For $x^4 - 5x^2 + 4 = 0$, let $y = x^2$. The equation becomes $y^2 - 5y + 4 = 0$, which factors as $(y-1)(y-4) = 0$, giving $y = 1$ or $y = 4$. Translating back: $x^2 = 1 \Rightarrow x = \pm 1$, and $x^2 = 4 \Rightarrow x = \pm 2$. Four roots — $\pm 1, \pm 2$ — recovered from a quadratic you could factor by inspection, because the substitution stripped away the part of the problem that only looked hard.

Worked example 2: a radical equation

Solve $\sqrt{x} + \dfrac{6}{\sqrt{x}} = 5$ for $x > 0$. Let $u = \sqrt{x}$, so $u > 0$ and the equation becomes $u + \dfrac{6}{u} = 5$. Multiply through by $u$: $u^2 - 5u + 6 = 0$, which factors as $(u-2)(u-3) = 0$, so $u = 2$ or $u = 3$. Translating back, $\sqrt{x} = 2 \Rightarrow x = 4$, and $\sqrt{x} = 3 \Rightarrow x = 9$. Both check out in the original equation. Without the substitution, this would have meant clearing a radical from a rational equation directly — doable, but far more error-prone than solving a quadratic in $u$.