Engineering / Mathematics — Rings and Polynomial Rings
Formal Derivatives of Polynomials
The formal derivative as an algebraic operation, its rules, and its use in detecting repeated factors.
Executive summary
The formal derivative is defined by the familiar power rule applied to coefficients, with no limits involved. It is a purely algebraic operation valid over any commutative ring.
Its principal use is detecting repeated factors, which is the basis of square-free decomposition.
Learning objectives
- Define the formal derivative and verify its rules.
- Use it to detect repeated factors.
- Recognise the characteristic p complication.
01Definition and rules
Formal derivative
For f = Σ aᵢX^i, define f' = Σ i · aᵢX^{i−1}, where i · aᵢ means the coefficient added to itself i times.
No limiting process is involved. The definition is combinatorial and applies over any commutative ring, including finite fields where limits are meaningless.
| Rule | Statement |
|---|---|
| Linearity | (af + bg)' = af' + bg' |
| Product | (fg)' = f'g + fg' |
| Power | (f^k)' = k f^{k−1} f' |
| Constant | c' = 0 |
Each is verified by direct coefficient computation. The product rule in particular is a finite rearrangement of the convolution defining polynomial multiplication.
02Detecting repeated factors
Repeated factor criterion
An irreducible h divides gcd(f, f') if and only if h² divides f.
Hence f is square-free exactly when gcd(f, f') = 1, in characteristic zero.
The mechanism is the product rule. Writing f = h²g gives f' = 2hh'g + h²g', and h divides both terms, so it divides the derivative as well as f.
This gives a square-free decomposition algorithm requiring only gcd computations and no factorisation, which is why it is the cheap first step of every polynomial factorisation method.
03The characteristic p complication
Zero derivative in characteristic p
Over a field of characteristic p, f' = 0 if and only if f is a polynomial in X^p, that is f(X) = g(X^p) for some g.
Over a finite field this is handled cleanly. Since the Frobenius map is a bijection, every coefficient has a unique p-th root, so g(X^p) = (h(X))^p for a computable h. The repeated part is extracted by taking that p-th root rather than by a gcd.
Compute gcd(f, f')
If it is 1 and f' is non-zero, f is square-free.
If f' = 0
Then f = g(X^p); take the p-th root of each coefficient to obtain h with f = h^p.
Recurse
Apply the procedure to h and to the gcd factor as required.
Assemble
Combine to obtain the full square-free decomposition.
04Frequently asked questions
Is the formal derivative related to the analytic one?
Over the reals they agree, which is why the rules look familiar. The formal version is defined purely algebraically and remains valid where no notion of limit exists.
Why does square-free decomposition come first in factorisation?
Because it is cheap — a few gcds — and it simplifies the remaining work. Distinct degree and equal degree factorisation both assume a square-free input.
Does the criterion work over Z?
Yes, characteristic zero poses no difficulty. The complication arises only in positive characteristic, where multiplying a coefficient by p annihilates it.
Sources and method
Structural reference: Victor Shoup, A Computational Introduction to Number Theory and Algebra, Version 1, Cambridge University Press, 2005 — book pages 228-229.
This page carries the durable method layer only: definitions, constructions, algorithms, complexity results and selection criteria, authored originally for KEVOS. No text is transcribed or paraphrased from the source, and no numeric tables or benchmark data are reproduced — these are routed to live authoritative sources instead.
Author: Kevin Jogin. Last reviewed 2026-08-07.