Executive summary
The source proves the existence of complex roots through a geometric argument. For a non-constant polynomial with non-zero constant term, consider circles centred at the origin and map each circle through the polynomial. For a sufficiently small radius the constant term dominates, so the image winds zero times around the origin. For a sufficiently large radius the leading term dominates, so the image winds as many times as the polynomial degree. If the polynomial never vanished while the radius changed continuously between these two values, the image curves would avoid the origin and their integer winding number could not change. Because the endpoint windings differ, some intermediate circle must map through zero, producing a root.
What this handbook page teaches
- Form the image of a circle under a complex polynomial.
- Use dominance estimates for small and large radii.
- Calculate image winding in the two limiting regimes.
- Apply the stability of integer winding under root-free deformation.
- Derive factorisation consequences from the existence of one root.
Core concepts
Δarg = 2πk
The integer k records the net number of turns about the reference point, including orientation.
Circle images under a polynomial
Let P(z)=a0 z^n+a1 z^(n-1)+…+an with a0≠0. For each radius R, parameterise the circle by z=R(cos 2πt+i sin 2πt) and examine the closed image curve P(z). If the image passes through zero, the corresponding input point is already a root.
Assume temporarily that no image circle in a chosen radius interval passes through zero. Then each image has a defined winding number about zero, and continuous variation of R deforms one image into another without crossing zero.
Small and large radius regimes
For sufficiently small R, the non-constant terms can be made much smaller in modulus than the non-zero constant term an. The image curve stays in a small neighbourhood of an that does not contain zero. It therefore has winding number zero.
For sufficiently large R, factor the leading behaviour as a0 z^n[1+(a1/a0)z^-1+…+(an/a0)z^-n]. The bracket stays close to one, while a0 z^n winds n times. The image has winding number n.
The contradiction and factorisation
If the polynomial had no root between the selected radii, the winding number would be an integer-valued continuous function of the radius and hence constant. But it is zero at the small radius and n at the large radius. Therefore the root-free assumption fails and P(z)=0 for some complex z.
Once one root z1 exists, polynomial division gives P(z)=(z-z1)Q(z). Apply the same existence theorem to Q and repeat. A degree-n complex polynomial therefore factors into n linear factors when multiplicities are counted.
Working method
Use the following sequence as a repeatable analysis workflow. The steps are ordered to separate definitions and admissibility checks from structural conclusions, so a result can be audited without relying on intuition or an unverified diagram.
- If the constant term is zero, record the immediate root
z=0and reduce the problem. Otherwise continue. - Choose a small radius using a triangle-inequality estimate that makes the sum of non-constant terms strictly smaller than the constant term.
- Choose a large radius so the lower-degree terms, after division by the leading term, have combined modulus strictly smaller than one.
- Compute the image winding in each regime by comparing with the dominant term.
- Assume no root lies on any intermediate circle and infer that winding cannot change during the deformation.
- Use the contradiction between windings zero and
n, then factor out roots recursively.
Dominance estimates without exact optimisation
The source does not need the best possible small or large radius. It chooses radii that make simple inequalities decisive. For small R1, require the total modulus of all terms involving positive powers of R1 to be less than a fixed fraction of |an|. Then P(z) cannot reach zero on that circle because the perturbing terms cannot cancel the constant term.
For large R2, divide by a0 z^n and require the modulus of the sum of all reciprocal-power corrections to be less than a fixed fraction of one. The bracket cannot reach zero and contributes no net winding, so the leading term controls the winding.
The proof is robust precisely because it uses inequalities rather than exact root locations. It establishes existence for every non-constant complex polynomial without solving the polynomial.
Technical reasoning and deeper connections
This proof demonstrates a recurring strategy: compare a difficult map with a dominant simple map on a boundary, then use an integer invariant that cannot change unless a forbidden event occurs. The forbidden event here is passage of the image through zero, which is exactly a polynomial root.
For polynomials with real coefficients, non-real roots occur in conjugate pairs. Combining this with complete complex factorisation shows that every real polynomial factors into real linear factors and irreducible real quadratic factors. This result is used earlier in the source's discussion of real field extensions.
The derivative and multiplicity concepts introduced immediately after factorisation are later used to find where a parameterised polynomial can have multiple roots. A root of multiplicity greater than one is also a root of the derivative, so simultaneous equations locate possible branch values.
The source explicitly notes that the winding proof contains passages treated as proof ideas rather than fully rigorous foundational topology. The handbook therefore presents the logical mechanism and estimates faithfully while marking that limitation.
Quick-reference matrix
| Radius regime | Dominant part | Image winding |
|---|---|---|
Very small R | Constant term an | 0, provided an≠0. |
Intermediate R | No fixed dominant term | Constant if no image crosses zero. |
Very large R | Leading term a0 z^n | n. |
| Crossing event | P(z)=0 | A root exists on that circle. |
Common mistakes
- Trying to solve for the roots rather than proving existence.
- Choosing small or large radii without a strict inequality preventing cancellation.
- Assuming winding changes continuously as a real number; it remains integer-valued while defined.
- Allowing an image curve to cross zero while still claiming the winding is defined.
- Forgetting the special case where the constant term is already zero.
- Claiming complete formal topological rigour beyond the proof level supplied by the source.
Verification checklist
Before accepting a solution or proof, confirm each item below. These checks are intentionally practical: they catch the most common category errors, missing hypotheses and structure mismatches.
- Small-radius dominance is justified quantitatively.
- Large-radius dominance is justified after factoring the leading term.
- All image curves considered for winding avoid zero.
- Winding at each endpoint radius is computed from a simple comparison curve.
- The contradiction identifies an actual zero of the polynomial.
- Recursive factorisation counts multiplicity correctly.
Frequently asked questions
Does this theorem give a formula for the roots?
No. It proves existence and factorisation, not a finite radical expression.
Why use circles centred at zero?
Their simple parameterisation makes the leading monomial wind exactly n times and allows radius to be varied continuously.
What if the polynomial has a root at zero?
Then the theorem is immediately satisfied; factor out the corresponding power of z if further roots are being studied.
How does this connect to the quintic impossibility result?
Existence of five complex roots is compatible with the absence of a universal radical formula. Existence and representability are different questions.
Source scope
The source labels some steps in this topological existence argument as non-rigorous passages that can be made exact with additional topology. This page preserves that qualification.
Source basis: this page is a handbook-style synthesis of the uploaded mathematics source, principally sections 2.8. It paraphrases the supplied material and does not reproduce source artwork or long source passages. Historical names, publisher details and personal details have been intentionally omitted.