Manifolds Unfolded

“Eisenstein’s criterion is famous. Perron’s and Cohn’s are quietly powerful.”

Irreducibility of polynomials over $\mathbb Z[X]$ lies at the heart of algebra and number theory. From constructing number fields to understanding Galois groups, from algebraic geometry to arithmetic dynamics, the question:

$\text{Is a given polynomial irreducible?}$

appears relentlessly.

Every student of abstract algebra meets Eisenstein’s Criterion early on—and rightly so. It is elegant, efficient, and devastatingly effective when applicable. But Eisenstein has a limitation that every practitioner knows: it requires a suitable prime, and many natural polynomials simply do not cooperate.

In this article, we explore two less celebrated but remarkably useful irreducibility criteria:

1. Perron’s Irreducibility Criterion — rooted in coefficient dominance and geometric configuration of the zeros.
2. Cohn’s Irreducibility Criterion — arising from digit expansions and primality of values.

These criteria are not replacements for Eisenstein. They are complements—often succeeding precisely when Eisenstein fails. Our goal is to explain them conceptually, rigorously, and practically, with intuition, proofs sketches, and worked examples.

---

1. The landscape: irreducibility beyond Eisenstein

Let us recall the basic setting. A polynomial $f(x)\in \mathbb Z[X]$ is irreducible over $\mathbb Z$ if it cannot be written as a product of two non-constant polynomials with integer coefficients.

Eisenstein’s criterion exploits divisibility by a prime $p$. But what if:

• no such prime exists?
• coefficients are small but arranged asymmetrically?
• the polynomial arises from a combinatorial or digital construction?

This is where Perron and Cohn enter the story—each offering a strikingly different philosophy of irreducibility.

---

2. Perron’s Irreducibility Criterion

We begin with Perron’s criterion, which applies to monic polynomials and uses a dominance condition on coefficients.

Theorem (Perron).

Let$f(X)=X^n+a_{n-1}X^{n-1}+a_{n-2}X^{n-2}+\dots+a_1X+a_0\in \mathbb Z[X]$

with $a_0 \neq 0$

If either$$|a_{n-1}| > 1 + |a_{n-2}| + \cdots + |a_0|,$$or$$|a_{n-1}| = 1 + |a_{n-2}| + \cdots + |a_0| \quad \text{and} \quad f(1) \neq 0,\; f(-1) \neq 0,$$

then $f(X)$is irreducible over $\mathbb Z$.

---

3. Intuition behind Perron’s criterion: geometry of roots

Perron’s criterion is deeply geometric.

The key idea is this:

A strong dominance relation of the coefficients $$|a_{n-1}| > 1 + |a_{n-2}| + \cdots + |a_0|,$$forces almost all roots of the polynomial to lie strictly inside the unit circle, with at most one root outside.

This is not accidental. One proves (using Rouche-type arguments or analytic bounds) that under Perron’s inequality:

• exactly one root satisfies $|\alpha| > 1$,
• all remaining roots satisfy $|\alpha| < 1$.

Now suppose $f(X)=g(X)h(X)$ in $\mathbb Z[X]$.
Then each factor must have an integer constant term. But a polynomial whose all roots lie inside the unit circle must have constant term of absolute value strictly less than 1, contradicting integrality.

Thus, no such factorization can exist.

This irreducibility criterion uses geometric configuration of zeros rather than modular arithmetic.

---

4. Perron in action: worked examples

Example 1.

Consider $f(X)=X^5+7X^4+2X^3-X+1$.
Here,$$|a_4| = 7,\quad |a_3| + |a_2| + |a_1| + |a_0| = 2 + 0 + 1 + 1 = 4.$$

Since $7 > 1 + 4$, Perron’s criterion applies, showing that$\text{ } f(X) \text{ is irreducible in } \mathbb Z[X].$

No primes, no modular arithmetic—just a sharp inequality.

---

Example 2 (boundary case).

Let $f(X)=X^5+5X^4+2X^2+X-1$

Here,$|a_4|=5, \text{ and }1+|a_3|+|a_2|+|a_1|+|a_0|=1+|0|+|2|+|1|+|-1|=1+2+1+1=5=|a_4|$

Check:$f(1)=1+5+2+1-1=8\neq 0\text{ and } f(-1)=-1+5+2-1-1=4\neq 0$

Thus, Perron’s second condition applies, and $f(X)$ is irreducible.

---

5. A Non-Monic Version of Perron’s Theorem

There is a non-monic version of the Perron’s Irreducibility Criterion:

Theorem (Non-Monic Version; Perron).

Let$\text{ }f(X)=a_nX^n+a_{n-1}X^{n-1}+\dots+a_0\in \mathbb Z[X]$

Suppose the coefficients satisfy the condition that $|a_0|>|a_1|+\dots +|a_n| \text{ and } |a_0| \text{ a prime number, }$then $f(X)$is irreducible over $\mathbb Z$.

Example:

Consider $f(X)=2X^3-2X+7$.
Here, $|a_0|=7, \text{ a prime number and, }|a_1|+|a_2|+|a_3|=2+0+2=4, \text{ thus }|a_0|>|a_1|+|a_2|+|a_3|$as $7>4$. Therefore Perron’s criterion applies, showing that$\text{ } f(X) \text{ is irreducible in } \mathbb Z[X].$

No primes, no modular arithmetic—just a sharp inequality.

6. Strengths and limitations of Perron’s criterion

Strengths

• No primes required (for monic version).
• Works well for polynomials with a dominant coefficient.
• Excellent for constructed examples and asymmetrical polynomials.

Limitations

• Fails if coefficients are evenly distributed.
• Is sufficient, not necessary.

Still, Perron is a remarkably efficient test when applicable.

---

7. Cohn’s Irreducibility Criterion

Cohn’s criterion comes from a completely different direction: number representation in a base.

Theorem (Cohn).

Let $b \ge 2$ be an integer, and let $f(X)=a_nX^n+a_{n-1}X^{n-1}+a_{n-2}X^{n-2}+\dots+a_1X+a_0\in \mathbb Z[X]$

with$$0 \le a_i \le b – 1 \quad \text{for all } i.$$

If the integer $f(b)$ is prime, then $f(X)$ is irreducible in $\mathbb Z[X]$.

---

8. Intuition behind Cohn’s criterion: digits and factorization

Think of $f(b)$ as a base-$b$ expansion:$$f(b) = a_n b^n + \cdots + a_0.$$

If $f(X)=g(X)h(X)$ in $\mathbb Z[X]$ then$$f(b) = g(b)h(b).$$
But under the coefficient constraints, both $g(b)$ and $h(b)$ are positive integers greater than 1, forcing $f(b)$ to be composite.

Thus primality of $f(b)$ forbids polynomial factorization.

This is irreducibility certified by arithmetic evaluation.

---

9. Cohn in action: classic examples

Example 3 (base 10).

Let$\text{ }f(X)=X^4+7$

Then$$f(10) = 10007,$$

which is prime.

Since the coefficients $1,0,0,0,7$ are valid base-10 digits, Cohn’s criterion applies:
$f(X)=X^4+7 \text{ is irreducible in }\mathbb Z[X].$

---

10. Why these criteria matter (and why they deserve more attention)

Perron and Cohn remind us that irreducibility is not only a modular phenomenon. It can emerge from:

• geometry of roots,
• asymmetry of coefficients,
• arithmetic evaluation,
• combinatorial structure.

They are especially useful in:

• research-level constructions,
• algorithmic testing,
• pedagogical enrichment beyond Eisenstein.

---

11. Final remarks

If Eisenstein is the hammer, then Perron and Cohn are the precision tools—subtle, elegant, and devastatingly effective when used in the right setting.

Every serious student of algebra should know them.

Irreducibility has many faces. Prime divisibility is only one of them.